Fullscreen HTML5Med Res (??MB)HTML4

APRIL 2025 ISSN 1030-2662 04 9 771030 266001 $ 00* NZ $1390 The VERY BEST DIY Projects! 13 INC GST INC GST HDMI video up to 1280 x 720 Four USB Type-A connectors 3.5-inch audio socket Programmable using MMBasic DS3231 real-time clock - PICO/2/COMPUTER 433MHz Transmitter Module » drop-in replacement for commercial equivalents » easy-to-build, with just a few parts » good for short range communications adjustable rotation speed, direction and LED brightness www.jaycar.com.au Contents Vol.38, No.04 April 2025 11 3D-MID and IMSE 3D-MID (mechatronic integrated devices) and IMSE (in-mould structural electronics) are two similar processes used to create three-dimensional parts with integrated mechanical and electronic functions. By Dr David Maddison, VK3DSM Manufacturing technology 34 The Power Grid’s Future, Part 2 Continuing from last month, we look at how solar and wind generators are connected to the grid to match demand and improve stability. This is done via various types of inverters, with voltage & inertia control for grid stability. By Brandon Speedie Electricity generation 48 Antenna Analysis, Part 3 In the final part of this series, we explain how to calculate the bandwidth of an antenna matching network. Bandwidth is important for antennas, as it determines the range of frequencies at which they are effective. By Roderick Wall, VK3YC Radio antennas 64 Precision Electronics, Part 6 So far we have covered analog circuitry, but nowadays a lot of circuits are designed with digital components too, such as a microcontroller. And these digital/analog signal conversions introduce their own errors. By Andrew Levido Electronic design 24 The Pico /2/ Computer Not to be confused with our previously published Pico Computer from December 2024, the Pico/2/Computer has an HDMI-compatible video output, four USB host ports and is optimised for use with MMBasic. By Peter Mather & Geoff Graham Computer project 58 Rotating Light for Models This simple circuit acts as a rotating LED light display, with adjustable brightness, speed and direction. It’s a flexible project that uses SMD or through-hole LEDs of any colour, and works great with model kits and more. By Nicholas Vinen Model/toy project 72 433MHz Transmitter Module The 433MHz band is good for short-distance communications and does not require a licence. You can build this 433MHz transmitter module that is a drop-in replacement for commercial versions. By Tim Blythman Radio control project 82 Power LCR Meter, Part 2 This isn’t just any old LC Meter; it can deliver currents up to 30A for measuring core saturation of inductors rated up to 1H. And it measures capacitors to 1F and resistors down to small fractions of an ohm. By Phil Prosser Test equipment project Page 11 3D-MID & IMSE Manufacturing Antenna Analysis and Optimisation Part 3: Page 48 Page 58 Rotating Light for Models 2 Editorial Viewpoint 5 Mailbag 20 45 57 Mini Projects 63 Silicon Chip Kits 71 Online Shop 80 Circuit Notebook 90 Serviceman’s Log 96 Vintage Radio 101 Ask Silicon Chip 103 Market Centre 104 Advertising Index 104 Notes & Errata 1. Discrete 555 timer 2. Weather monitor Subscriptions 1. 8x8 RGB LED matrix display with WiFi 2. Interchangeable dual triode valves Astor APK superhet by Jim Greig SILICON SILIC CHIP www.siliconchip.com.au Publisher/Editor Nicholas Vinen Technical Editor John Clarke – B.E.(Elec.) Technical Staff Bao Smith – B.Sc. Tim Blythman – B.E., B.Sc. Advertising Enquiries (02) 9939 3295 adverts<at>siliconchip.com.au Regular Contributors Allan Linton-Smith Dave Thompson David Maddison – B.App.Sc. (Hons 1), PhD, Grad.Dip.Entr.Innov. Geoff Graham Associate Professor Graham Parslow Dr Hugo Holden – B.H.B, MB.ChB., FRANZCO Ian Batty – M.Ed. Phil Prosser – B.Sc., B.E.(Elec.) Cartoonist Louis Decrevel loueee.com Founding Editor (retired) Leo Simpson – B.Bus., FAICD Silicon Chip is published 12 times a year by Silicon Chip Publications Pty Ltd. ACN 626 922 870. ABN 20 880 526 923. All material is copyright ©. No part of this publication may be reproduced without the written consent of the publisher. Subscription rates (Australia only) 6 issues (6 months): $70 12 issues (1 year): $130 24 issues (2 years): $245 Online subscription (Worldwide) 6 issues (6 months): $52.50 12 issues (1 year): $100 24 issues (2 years): $190 For overseas rates, see our website or email silicon<at>siliconchip.com.au * recommended & maximum price only Editorial office: Unit 1 (up ramp), 234 Harbord Rd, Brookvale, NSW 2100. Postal address: PO Box 194, Matraville, NSW 2036. Phone: (02) 9939 3295. ISSN: 1030-2662 Printing and Distribution: Editorial Viewpoint Ferrite beads are not inductors I often see ferrite beads drawn in circuit diagrams as if they are inductors, with “Lx” designators. While many circuit designers likely realise that they are not true inductors, treating them as such could cause confusion, especially for those reading the diagrams. This might lead them to assume that a ferrite bead is just another type of inductor, when in reality, it serves a very different purpose. Ferrite beads clearly exhibit some inductance – as do most components, including wires and PCB tracks – but their operation does not rely on it. At their simplest, ferrite beads are just a piece of wire passing close to (or through a hole in) a piece of ferrite. Some of the confusion may stem from the fact that ferrite is used as a core material in high-frequency inductors and transformers. However, in those applications, the ferrite core is surrounded by multiple turns of wire to create significant inductance. In contrast, a ferrite bead typically has just one or a few turns and thus a relatively low inductance. Ferrite is a ceramic material that contains iron oxide. Like other magnetic core materials, it provides a path for magnetic flux, but only up to a certain frequency. Beyond that, ferrite becomes highly ‘lossy’, converting much of the magnetic energy to heat, due to hysteresis and eddy current losses within the ferrite material. Ferrite beads take advantage of this property to suppress unwanted highfrequency signals by dissipating their energy, effectively acting as a frequencydependent resistor rather than an inductor. Unlike an inductor, a ferrite bead does not store energy or resonate. It simply increases its effective resistance in a targeted frequency range to block unwanted signals. Ferrite beads are available with all sorts of resistances and curves, with the resistance peaking at different frequencies depending on the exact construction of the bead. At very high frequencies, the impedance of the ferrite bead drops as the parasitic capacitance across it starts to cause the signal to bypass it. While you can make a ferrite bead yourself, by passing a wire through a ferrite core, they are also available as pre-built SMD ‘chip’ devices that you can simply solder across pads on a board. Pre-formed through-hole beads are also available but are less common these days. A bead’s peak resistance can range from a few ohms up to a few kilohms, although most fall between 100W and 1kW. Many have a DC resistance well under 1W and can handle from a few hundred milliamps to several amps. However, those with a higher peak resistance usually also have a higher base resistance at DC. The impedance peak is usually between 100MHz and 1GHz and can be fairly broad, allowing the bead to block RF signals over a wide range of frequencies (to some extent, at least). Much of the impedance is real resistance, but not all. The accompanying figure from the TDK MPZ1608 data sheet should give you some idea of the behaviour of a range of different ferrite beads. So, rather than thinking of ferrite beads as inductors, it’s more accurate to consider them as a lossy impedance element that selectively dampens high-frequency signals. That distinction matters. The 2020-2024 block of Silicon Chip PDFs on USB is now available (see p95). Order the set at siliconchip.au/ Shop/digital_pdfs by Nicholas Vinen 9 Kendall Street, Granville NSW 2142 2 Silicon Chip Australia's electronics magazine siliconchip.com.au MAILBAG your feedback Letters and emails should contain complete name, address and daytime phone number. Letters to the Editor are submitted on the condition that Silicon Chip Publications Pty Ltd has the right to edit, reproduce in electronic form, and communicate these letters. This also applies to submissions to “Ask Silicon Chip”, “Circuit Notebook” and “Serviceman’s Log”. How to stop CDI failures leaving you stranded I refer to Vincent Stok’s letter concerning his Programmable Ignition tribulations published on page 8 of the January 2025 issue. My first forays into alternative ignition system building began with a CDI while I was an electrical engineering student back in 1970. It struck me at the time that it would be unwise to install it permanently wired. Instead, I installed it via a Mate-and-Lock connector pair. I kept and continue to keep an extra “dummy” male connector in my glove box, with the terminals bridged in such a way that whenever it is substituted for the CDI connector, the vehicle is instantly returned to the Kettering ignition system with which the car was built. This strategy has served me well because, some years later, the CDI suffered a series of inverter oscillator transistor lockups on very hot days. The fast reset strategy was to switch the ignition off momentarily and then carry on until the car could be parked and the quick change act performed. When the weather cooled, the CDI could be just as quickly returned to service. Of course, this state of affairs couldn’t be tolerated for long, so the transistors were soon swapped for other types with far better temperature tolerance. As you have probably guessed, thanks to the quick change system, I could still drive the car while the CDI was on the workshop bench having its transistors swapped. That CDI has now run faultlessly since 1980, although in the meantime, I tested and proved several other prototype CDIs, dwell extenders and programmable ignitions on my old faithful, and on other vehicles, but always with the quick change wiring arrangement. All this has led me to discover another benefit of the quick change. It can be used to quickly determine whether an ignition failure has taken place within the alternative ignition system or in the original vehicle wiring. Therein        www.icom.net.au siliconchip.com.au Australia's electronics magazine April 2025  5 lies a couple of amusing original vehicle ignition system wiring failure stories for another day. In conclusion, I strongly recommend that a quick change arrangement be installed whenever anyone fits any sort of after market ignition system. Make sure those who regularly drive your car are aware that the system exists and show them how to perform the quick change. George Clauscen, East Oakleigh, Vic. Making a variable AC power supply The ‘One Identity’ Factor A time machine would work on individual atoms, not on larger items like living bodies. When those atoms travel back in time each of them can only exist once; each atom with take up the identity it had before and proceed to exist in its previous role. This could have quite unexpected results for living bodies. The group of atoms that you called your body five years ago is significantly different from the group that you now own. The Loop You make a time machine and to test it out you set it to take you back one second. You press ‘start’ and it takes you back to where you were a second ago; you relive that second then the machine triggers again and you spend the rest of your life living that one second loop over and over again. Galactic Movement Astronomers tell us that the Milky Way galaxy is moving in relation to other galaxies at 580 kilometres per second. No-one knows where stationary actually is. If you set your time machine for one second forward it will take you to where your starting point will be in one second’s time. That could be a few hundred kilometres into outer space or underground. It is recommended that you time your first trial for when the constellation Virgo is just above the horizon, take a parachute and if there is water anywhere near have an EPIRB in your pocket. Get your copy for just $5.50: https://moonglowpublishing. com.au/store/p48/bewarethe-loop-jim-sinclair Beware! The Loop is available as an EPUB, MOBI & PDF RRP $5.50 | available as an EPUB, MOBI and PDF 6 Silicon Chip E-ISBN 9780645945669 I read with interest your reply to Andrew Hannan regarding the use of an audio amplifier as the basis for a variable 50Hz AC power supply. I agree that, for most applications, using a variac is a simple solution. Many years ago, I tried using an audio amplifier as an inverter and promptly destroyed several 2N3055 transistors. Recently, I required a reliable electronically controlled variable 0-500V AC 50Hz power supply for a test jig. I used two Silicon Chip CLASSiC-D amplifiers as an inverter driving two 60-to-240V 160 VA transformers in bridge mode. I used an XR2206 function IC to generate the 50Hz sinewave. An Arduino Mega 2560 module provides DC modulation to the XR2206, and the two transformer 240V output windings are in parallel for 0-250V <at> 1A or in series for 0-500V <at> 500mA. The inverter drives both inductive and capacitive loads with no ill effects and has proved to be indestructible. One change I made to the Classic-D amplifier modules was to remove the 5.6kW dead-time resistor to increase the Mosfet dead time. The AC current waveform when driving a capacitive load is a sinewave. With the Mega2560, the test program uses pin 44 (PWM) running at 31kHz integrated into a variable DC modulation voltage. By using analogWrite(PWM, n); the output voltage from the inverter secondary windings when in parallel is n = the required voltage, ie, analogWrite(PWM, 120); gives an output voltage of 120V AC. Andrew Fraser, via email. Educational articles appreciated The more theoretical articles, such as “Precision Electronics” and “Antenna Analysis” are a good addition to Silicon Chip. I am enjoying them. Such material does not always have to live inside textbooks. Paul Howson, Warwick, Qld. Comment: we try to keep this type of article as accessible as possible. For example, we avoid heavy mathematics, although an understanding of basic algebra will help the reader. On the COMPAC and SECOM undersea cables I was very interested in David Maddison’s article on undersea cables in the December 2024 issue (siliconchip. au/Article/17304). In 1963, the Commonwealth Overseas Pacific Cable (COMPAC) was opened by the Queen and I recorded this on my old Pyrox wire recorder. In 1967, the S-E Asian Commonwealth Cable System (SECOM) cable was also opened by her, and I was also able to record this on the same machine. These historic recordings contain quite an amount of technical information in the introduction. There are links to each of them (underneath the photo) at the website: siliconchip.au/link/ac4c On the COMPAC recording, after the introduction by Australia's electronics magazine siliconchip.com.au Discover New Technologies in Electronics and Hi-Tech Manufacturing See, test and compare the latest technology, products and explore turnkey solutions SMCBA CONFERENCE Hear from international experts presenting the latest design and assembly innovations at the Electronics Design and Manufacturing Conference Details at www.smcba.asn.au In Association with siliconchip.com.au Supporting Publication Organised by Australia's electronics magazine Co-located with April 2025  7 the Queen (which, by the way starts after a long pause – apparently they couldn’t switch her in quick enough), the Prime Ministers of the United Kingdom (Douglas-Home), Canada (Pearson), Australia (Menzies) and New Zealand (Holyoake) each have their say. This may be of interest to anyone wishing to learn more about these two projects. Christopher Ross, Tübingen, Germany. Another method for extracting ROM data Reading Dr Hugo Holden’s article on extracting ROM data from old microcontrollers in the January 2025 issue (siliconchip.au/Article/17609) brought back memories of a similar experience I had back in 2010, when I needed to extract the EPROM data from a pre-programmed Motorola MC68705R3 microcontroller. What helped was that in the 1980s and 1990s, I was using the Motorola MC68705P3, MC68705U3 and MC68705R3 microcontrollers and had built the recommended programming board as described in the data sheets. These microcontrollers are programmed in a similar way to the MC1468705G2 as described by Dr Holden – they program themselves by running a built-in Bootstrap Program, which copies the contents of an external EPROM holding your program into its own internal EPROM. The programming board uses a 4040 CMOS counter connected to the external EPROM’s address lines so the Bootstrap Program can sequentially access each byte of the external EPROM by pulsing the 4040’s clock input. So, to extract the data from an already-programmed microcontroller, I looked for any weakness in the program/ verify procedure. I discovered that as long as I didn’t apply the 21V programming voltage to the microcontroller, the program/verify cycle would still run, but no bytes in the internal EPROM were changed. Next, I needed a copy of the Bootstrap Program to see what it did. On a breadboard, I connected nine LEDs to the output lines of a spare microcontroller, having programmed it to read each byte of the Bootstrap Program and display it slowly on eight LEDs. The ninth LED was used as a ‘data ready’ indicator, which flashed to indicate a new byte was being displayed. Then I simply wrote the bytes down in hexadecimal (115 bytes for the P3, 120 bytes for the U3 and R3). The circuit and code I used can be found here: http://matthieu.benoit. free.fr/6805.htm 8 Silicon Chip Disassembling the Bootstrap Program revealed that when it runs, it first copies itself into RAM and continues running from there. After it finishes programming the internal EPROM, it lights the ‘programmed’ LED. It modifies a few lines of its own code so that when it runs again, it compares the contents of the external EPROM with its internal EPROM. If all bytes are correct, it lights the ‘verified’ LED, then stops. My disassembled listing of the Bootstrap Program for the MC68705U3 and R3 is here: siliconchip.au/link/ac49 Looking closely at the verify procedure, I noticed two things. Firstly, if all bytes verified correctly, the Bootstrap Program executes a bclr (bit clear) command at the end to output a 0, which lights the verified LED. However, if a byte did not verify, the Bootstrap Program would change the bclr command to a bset (bit set) and the LED would not light. Secondly, unlike Dr Holden’s MC1468705G2, which stops when a byte does not verify, with the MC68705P3, U3 and R3, the verify procedure continues running until the whole internal EPROM has been tested. That’s where I found the weakness: each time a byte does not verify, it changes the bclr command to a bset, whether or not it has already been changed. So I figured that if I measured the period between clock pulses used to increment the 4040 counter, I’d get a certain time for bytes which verify but a slightly longer time if a verify failed, because the Bootstrap Program performed the extra step of changing the bclr into a bset (it takes 28us longer). I unplugged the 4040 counter and external EPROM from my programming board and connected the required lines to a National Instruments USB6009 interface, and the clock pulse from the 4040 to a PCI6023E interface. Using LabView, I could fake the external EPROM and run the program/verify procedure for each byte of data from 00 hex to FF hex in turn. This produced 256 files containing the time intervals between clock pulses, one file for each value tested. The whole procedure took about 15 minutes. Another LabView program found the times in each of the 256 files where verification was correct, and before long, I had the contents of the EPROM of the pre-programmed MC68705R3 (except for the last byte). More detail on this procedure can be found here: siliconchip.au/link/ac4a Since I was only interested in getting the EPROM contents from one microcontroller, I didn’t continue to develop a stand-alone reader but others have gone further, as can be seen here: siliconchip.au/link/ac4b Peter Ihnat, Wollongong, NSW. Windows 11 and planned obsolescence I agree with your comments about Windows 11 requiring a Microsoft account to log on and not being able to use it offline (Editorial viewpoint, February 2025). That would make the PC unusable if the internet went down and you wanted to do something on it you didn’t need internet for. However, there are ways around this. Check out Tom’s Hardware for solutions to this problem. I do not have and never will have Windows 11, unless I buy a new PC, which I am not likely to do anytime soon or most likely ever. I have several older computers and laptops, and I have Windows 10 on some of them. Windows 10 also requires a Microsoft account to log on if you just go through the normal Australia's electronics magazine siliconchip.com.au installation steps. But if you tell it that you don’t have internet, then it will let you log on with a local account instead. Apparently, this does not work for Windows 11, so it needs a work-around to enable logging on with a local account. The hardware requirements for Windows 11 ‘in the name of security’ are overboard and are the reason older hardware will not support Windows 11. Then you have the hardware/software scenario where new peripherals such as printers do not have drivers for older operating systems. My mate bought a new printer when his old printer carked it and he could not get drivers for Windows 7, so he had to upgrade to Windows 10, which he found to be sluggish compared to Windows 7 and new updates every few minutes (almost). One way out of this is to use Linux, but as you pointed out, if you need a particular software package that is not available on Linux, you have to stick with Windows. I have various versions of Linux that I installed on several older laptops to have a look at them, but I mostly only use one with Lubuntu that we use for catch-up TV when our recorded program is corrupted and unwatchable. On top of all this, we now have subscriptions for new versions of software, whereas previously it was a one-off license fee. So if you need new features, you have to switch to a yearly subscription with ongoing cost. Personally, I have no need for any software that requires a license and I use freeware programs such as OpenOffice and many others. There’s a lot of good free software out there for the home user. I can see no end to all this, and even mobile phones become redundant when existing standards are discontinued. Even 4G phones won’t work on 4G if they don’t have a specific feature. We found this with an iPhone 5S, which won’t work on 4G, even though it is a 4G phone. Bruce Pierson, Dundathu, Qld. Can Windows and Linux coexist? I agree with your editorial in the February issue on staying on Windows 10 rather than ‘upgrading’ to Windows 11. I too have my PC set and customised with many programs that would take countless hours to install on Windows 11 (if they would work at all). Then I would have to change all the settings again. We use Windows 11 at work, and I hate it with a passion. You mentioned Linux. I was wondering if it is feasible to have Linux and Windows 10 installed on the same PC with a dual boot system, and only connect to the internet when using Linux for downloads, emails etc. Would it be that easy, or would your PC still be at risk? Thanks for your time and a great magazine. Geoff Coppa, Toormina, NSW. Comment: You can definitely dual-boot Linux and Windows 10/11, and share files between them, but it’s inconvenient to reboot whenever you need to run a Windows-only program. A better solution may be to run Windows in a Linux VM. It works, but the performance is not great for highly interactive programs like Altium Designer, CorelDraw, Adobe Photoshop/Illustrator/InDesign etc unless you can get GPU pass-through working (typically requiring a second GPU in the computer). GPU pass-through on virtual machines works in theory, but we found that there are lots of obstacles in practice. SC 10 Silicon Chip Australia's electronics magazine siliconchip.com.au M echatronic I ntegrated D evices I n M ould S tructural E lectronics This article discusses the new and related multidisciplinary technologies of Mechatronic Integrated Devices (MIDs) and In-Mould Structural Electronics (IMSE). Both techniques involve depositing metal tracks on three-dimensional plastic surfaces, with components soldered on top, but they are made differently. By Dr David Maddison, VK3DSM Photos above from top-to-bottom: an injection-moulded part prior to structuring laser activation and structuring of the part; visible conductive tracks added; the conductive tracks have been metallised; an SM4007 diode has been soldered to the conductive tracks. Source: LPKF – siliconchip.au/link/ac3x M echatronic Integrated Devices were known as Moulded Interconnect Devices until they were renamed in 2010. Mechatronic is a portmanteau of mechanical and electronics. They are devices with integrated mechanical and electronic functions, comprising an injection-moulded or 3D-printed plastic body (the “circuit carrier”) onto which are printed electrically conducting tracks, similar to those on a circuit board. Electronic components can then be soldered to those tracks. Also known as 3D-MIDs, these devices have been likened to three-­ siliconchip.com.au dimensional printed circuit boards (3D PCBs). By integrating both electrical and mechanical functions, space and volume can be saved. That’s especially useful in miniaturised devices like smartphones and tablets, mechatronic modules in motor vehicles (such as accelerometers) and medical devices (such as implantable prosthetics or hearing aids). The mechanical aspects of a 3D-MID relate to the moulded carrier substrate, which may form part of a connector, support structure for a high-powered LED, interconnector to another Australia's electronics magazine component, carrier for a specialised sensor or printed antenna etc. Examples of these will be given later. Apart from miniaturisation, additional advantages of MIDs include the integration of mechanical and electrical or electronic components into the one assembly and the possibility of new functionality not achievable in other ways. Almost any shape can be made. MIDs can provide a reduction in the number of parts required, a reduction in manufacturing cost, fewer materials required overall, reduced assembly cost and time, optimal placement of components and reductions in development time and ultimate weight. Related to 3D-MIDs are In-Mould Structural Electronics (IMSEs) or In-Mould Electronics (IMEs). With IMSEs, conductive tracks are incorporated into the item being fabricated at the time of moulding, typically using electrically conductive inks, rather than conductive metallic tracks being added after moulding, as with 3D-MIDs. Discrete electronic components such as LEDs, switches and capacitive controls can also be incorporated at the time of moulding. A typical application for this technology might be a control panel or a lighting panel such as in a car or aircraft (we’ll give examples of these later in the article). The history of MIDs & IMSEs 3D-MIDs were first developed in the 1980s, but they were not initially a success, perhaps because there was not a sufficient demand for the advantages they offered at the time. Also, the technology was not sufficiently well developed. Challenges included: • It was expensive. • It took a long time for a product to get to market. • Design changes were difficult due to tooling being hard to change. • There was a lack of production infrastructure. • There was a separation of specialists who did not work together, such as those working on electronics and those working on the moulded components and the metallisation aspects. • Engineers were not very familiar with the technology. Today, there is an increasing demand for 3D-MIDs due to their advantages. Those include more electronic packaging options and a greater April 2025  11 Fig.1: an example of IMD. Unlike IMSE, no electronic tracks are printed here (this was the predecessor of IMSE). The untrimmed film is on the left (note the print registration markings). On the right is the film after it has been moulded into a plastic body and trimmed. Source: www.dekmake. com/in-mold-decoration Fig.2: a 3D-MID antenna module made using laser direct structuring. The components are on the inside surface. Source: www.lpkf.com/en/ industries-technologies/electronicsmanufacturing/3d-mids-with-laserdirect-structuring-lds Fig.3: a 3D-MID from CIS containing a cell, ICs, capacitor, LED, resistors and switch. Source: https://cis.de/ en/products/electromechanicalcomponents-3/mid Fig.4: a 3D-MID component for a CCD sensor (left) and integrated into a system with standard PCBs (right) by Distant Focus Corporation. Source: https://3d-circuits.com/wp-content/uploads/2022/01/ Sensor-platform-for-a-large-format.pdf Fig.5: a 3D-MID sun sensor for an automotive climate control system. Source: HARTING; siliconchip.au/link/ ac3y Fig.6: a 3D-MID position sensor component for Adaptive Cruise Control (ACC) system from Continental AG. This version is smaller and cheaper than a PCBbased version and allows for the optimal location of components. Source: https://3d-circuits.com/wpcontent/uploads/2022/01/Positionsensor-for-adaptive-speed-control.pdf 12 Silicon Chip flexibility in design and miniaturisation of assemblies. The technology has improved with better materials, better processes, rapid prototyping and reduced development time due to CAD/CAM (computer aided design and manufacturing). It also helps that 3D printing has become commonplace and engineers are now more familiar with it. A major development in the field of 3D-MIDs was laser direct structuring (LDS), which enabled an electrically conducting track in any desired pattern to be created without contact using a laser beam. This process was developed by LPKF and will be discussed later. IMSEs are a newer but related technology to 3D-MIDs. They have their origins with in-mould decoration (IMD), a process introduced in the 1970s where a printed decorative pattern is incorporated into a moulded plastic part. With IMD, a carrier film with the desired pattern is put into a mould designed for plastic injection moulding, then plastic is injected into the mould. The pattern is incorporated into the moulded part (see Fig.1). The area of the mould where the pattern film is placed should be as flat as possible to avoid excessive distortion (we’ll have more details on this later). This process was rapidly adopted after its introduction. Then, with the mobile phone boom of the 1990s and the requirement to print keypads and buttons with labels moulded into them, its adoption was further expanded. An enhancement of in-mould decorating was to print the pattern with an electrically conducting ink, which enabled the direct moulded-in integration of electrically conducting tracks. Electronic components such as LEDs for lighting or backlighting could then be attached to these tracks. Thus, IMSE was born. Digital printing techniques further enhanced design possibilities. IMSE has become very popular since the early 2000s. Examples of 3D-MID and IMSE devices Fig.7: a light assembly with LED made using 3D-MIDs. Laser direct structuring was used to print the tracks. Source: LPKF; siliconchip.au/link/ac3x Australia's electronics magazine As few people are familiar with 3D-MID or IMSE technology, we will start by presenting a few examples. 3D-MIDs can have complex 3D shapes with extensive conductive siliconchip.com.au tracks on both sides of the device; for example, the antenna element shown in Fig.2. With regard to antennas, 3D design allows them to be optimised for beam pattern, gain, efficiency and for millimetre-wave frequencies, due to the small size possible and high-­ precision of the printed tracks and device shape. Another example is shown in Fig.3. Fig.4 shows a 3D-MID to mount a CCD (charge-coupled device) image sensor, while Fig.6 shows a vehicular cruise control component, Fig.5 shows a sensor for a vehicular climate control system and Fig.7 shows a light assembly. These examples demonstrate the versatility of this technique, and its ability to make components that would be difficult, expensive or impossible to create with other processes. It also allows miniaturisation compared to conventional methods. In Fig.9, the foreground shows the front and the background shows the rear of the panel with printed conductive tracks. The settings can be changed via touch and movement of the rubberised switch membrane. Touches are detected by capacitance changes in the printed tracks. Fig.8 shows a “smart surface” in the form of a panel for an aircraft cabin, while Fig.11 is an example of a control panel for an electric car and Fig.10 is a circuit board with 3D structure. Fig.8: an aircraft interior lighting and indicator panel ‘smart surface’ made with IMSE technology by Tactotek. Source: www.tactotek.com/ industry-aviation Fig.9: an example of a control panel made using IMSE. Source: www. eastprint.com/wp-content/uploads/InMold-Electronics.pdf Fig.10: an IMSE circuit board with 3D structure by DuPont, made using their thermoformable electronic inks and pastes to produce a 1.5mm-thick 3D plastic surface. Some small electronic components have been fitted. Source: https://semiengineering.com/getready-for-in-mold-electronics Making a 3D-MID The basic steps for making a 3D-MID are: 1. A computer-aided design (CAD) drawing is created of the plastic body and the conductive track layout. The section on Altium Designer below has more details on this. Also see Fig.12. 2. Injection moulding or, in the case of low-volume or prototype devices, 3D printing is used to make the plastic body – see the lead images. 3. Structuring is performed, more specifically known as laser direct structuring or activation. This is the first part of the process by which conductive circuit traces are created. An infrared laser is used to write the desired pattern on the injection moulded part. Chemical additives which had previously been mixed with the plastic are activated by the localised heat of the laser, converting a non-conductive metal compound into isolated ‘islands’ siliconchip.com.au Fig.11: an example of a control panel for an electric car with touch-sensitive backlit switches and backlit indicators made using IMSE by SunChemical. Source: www.sunchemical.com/el/download-suntronic-for-in-mold-electronicsmaterials-brochure Australia's electronics magazine April 2025  13 of conductive metal, which become nuclei for the plating process in step 4. The traces can be quite fine. Harting (https://3d-circuits.com/en) states they can produce conductive traces down to 75µm (0.075mm) width and spacing. 4. Metallisation – the conductive metal track from step 3 has additional metal such as copper, nickel or gold (or a combination) added by an electroless method (no electrode), joining together the metal islands described in step 3. Additional metal can then be plated on using electroplating. This is a similar process to that of creating vias on a PCB. 5. Assembly – surface mount devices (SMDs) are attached to the conductive tracks by fully or semiautomatic processes. We will now discuss other process steps for making 3D-MIDs and IMSE devices in further detail. Design and prototyping Silicon Chip readers will be familiar with Altium Designer, both because we use it for our PCB designs and because we regularly review it as it is updated. Our last review, in the August 2024 issue (siliconchip.au/Article/16425) mentioned its new 3D-MID capability (on page 66). Altium Designer can now be used for the design of electronic aspects of 3D-MIDs and IMSE devices (Figs.12 & Fig.13). The mechanical components themselves are designed in a CAD tool for 3D mechanical design like SolidWorks. The Altium product is designed to integrate with such software. There is a video on this at https://youtu.be/c8Ld82LEHi8 LPKF ProtoLaser 3D The LPKF ProtoLaser 3D is an example of a machine for creating PCBs and 3D-MID prototype components using laser direct structuring to write conductive tracks onto plastic – see Figs.14 & 15. The ProtoLaser 3D can import designs from conventional layout software. The part might first be 3D printed. In 3D printing, a three-dimensional structure is built up one layer at a time. For prototyping and low-volume production, components can be produced by 3D printing and then processed to incorporate conductive tracks with laser direct structuring or by chemical means. LPKF is one company that offers technological solutions for this process. Once printed, the part is sprayed with LPKF ProtoPaint LDS. This paint contains additives to enable the LDS process. The paint is cured for three hours at 70°C, then the part is ready for LDS. Once the conductive tracks are written by the laser, the part is removed and the tracks thickened by an electroless plating process using LPKF’s ProtoPlate LDS solution. This machine and process can also be used for low-volume manufacturing of custom parts. For example, Boris Yasinov from Elcom Technologies said he could produce 500 custom filters in one week using this machine. Also see the video on the process at https:// youtu.be/THushdmY5Tc Note that normally, for mass production, the chemical components Figs.12 & 13: a 3D-MID being designed in Altium Designer and a rendering of an assembled 3D-MID. Source: www.altium.com/ altium-designer/features/ true-3d-circuit-design Fig.14: the LPKF ProtoLaser 3D for laser direct structuring. It can write conductive tracks onto prototype components. Source: www.lpkfusa. com/pls 14 Silicon Chip siliconchip.com.au ◀ Fig.15: the LPKF prototyping process for 3D-MID components. Source: www.lpkf.com/fileadmin/mediafiles/ user_upload/products/pdf/EQ/3DMID-LDS/brochure_lpkf_laser_ direct_structuring_en.pdf for LDS on 3D-MIDs are incorporated into the plastic feedstock for injection moulding and don’t have to be sprayed on. See below for further information on LDS. 1 2 3 4 5 6 5 Injection moulding Injection moulding is the process most used to fabricate 3D-MIDs and IMSE devices, except for prototyping or low-volume production runs. In fact, this is the most common method of mass production of solid plastic components of any kind. The process of injection moulding involves feeding plastic pellets from a hopper into a heated screw feed mechanism, which melts the plastic and injects the required amount into a mould (Fig.16). The mould is custom made for the required part (see Fig.17). A typical small injection moulding machine is shown in figure Fig.18. Two-shot injection moulding is a variation of injection moulding. A moulded part is first made as per the conventional injection moulding process. Then, the part is put into another siliconchip.com.au Fig.16: a simplified diagram of an injection moulding machine. The parts are: 1) screw feed with heated barrel to melt & inject plastic into mould, 2) hopper for plastic granules, 3) nozzle, 4) & 6) mould, 5) moulded part. Source: https://w.wiki/Cg6g section of the mould, which is a different shape to the first, into which additional material is injected to form the final shape of the part. The additional material may be the same type of plastic in a different colour, or a different type of plastic. For example, a rubbery compound can be added to the first part, as is commonly done with power tool housings. Those principles apply for all types of Australia's electronics magazine two-shot injection moulding, regardless of whether it is used for 3D-MIDs or not. One of the biggest costs for injection moulding is the cost of moulds, which are finely machined to high levels of accuracy and can come in complex shapes. Significant cost savings can be made by machining moulds from aluminium rather than stainless or hardened steel, but they have lower April 2025  15 Fig.17: the basic scheme of injection moulding. In this case, the charge of molten plastic is injected at the top into the mould P and two parts are produced simultaneously. Source: https://w.wiki/Cg6e charge nozzle durability, less longevity and worse dimensional accuracy. Nevertheless, aluminium moulds might be perfectly acceptable for many or most applications. Plastic choices sprue runner gates parts ejector pins A wide range of injection mouldable plastics are possible for 3D-MID, including: • acrylonitrile butadiene styrene (ABS) • polycarbonate • polyphenylene ether • polyetherimide • polybutylene terepthalate • polyethylene terepthalate (PET) • polyamide 66 (Nylon 66) • polyamide 6 (Nylon 6) • polyphenylene sulfide (PPS) • liquid crystal polymer • polyether ether ketone The specific choice of plastic depends on factors such as cost, thermal stability, mechanical properties, UV stability, chemical stability and compatibility with metallisation methods and additives. A variety of plastics are suitable for the fabrication of IMSEs, including polycarbonate, polyester, acrylic (Perspex), acrylonitrile butadiene styrene (ABS) and polyurethane. Metallisation methods for 3D-MID Fig.18: a typical small injection moulding machine. The cone-shaped hopper contains plastic granules. Beneath that is a horizontal screw feed. The mould goes inside the yellow cage and the product exits via the chute to the left of the yellow control panel. Source: https://w.wiki/Cg6h Fig.19: the process of laser direct structuring in which a laser creates metal particles by chemically transforming an additive precursor while also roughening the surface. Source: www.kyoceraavx.com/docs/techinfo/ Application-Based/LDSWorking-Principles-Benefitsfor-RF-Apps.pdf List of Important Acronyms (3D-)MID | (three-dimensional) Mechatronic Integrated Devices IM(S)E | In-Mould (Structural) Electronics IMD | In-Mould Decoration 16 Silicon Chip Australia's electronics magazine Metallisation of 3D-MIDs is generally done using laser direct structuring, but for two-shot injection moulding, it is done via chemical means. In LDS, the plastic compounds used in injection moulding have special additives of chemical compounds such that when a laser is directed at them, they undergo a chemical change to reduce them to pure metallic atoms, which are electrically conductive. These form a nucleation centre for additional subsequent metal deposition. In the case of 3D-printed prototypes or low-volume production, these can be instead sprayed onto normal plastic as a paint. This paint undergoes a similar chemical reaction when exposed to a laser beam, creating conducting pure metal atoms. The laser used in LDS is infrared and has a spot diameter typically of 50–100µm (0.05–0.10mm). Chemical additives typically used in LDS include cuprous oxide (Cu2O), cupric oxide (CuO) and copper chloride (CuCl2). These are reduced to siliconchip.com.au In-mould structural electronics (IMSE) Fig.20: two-step or two-shot injection moulding. The plastic to be metallised contains a special catalyst. Source: www.contag.eu pure copper nuclei by action of the laser, typically in the form of copper nanoparticles – see Fig.19. Other metal complexes can also be used. The surface of the plastic is also roughened by the laser, enhancing adhesion of the subsequent metallisation. The conducting pathways created by laser direct structuring are not thick enough to be used as-is; additional metallisation is required to thicken them and join the islands. Therefore, after the laser process, the components are dipped in a special chemical bath containing catalysts and a copper or other metallic compound. More copper (or another metal such as nickel, silver or gold) is deposited on the pathways modified by the laser, which contain the aforementioned metallic nanoparticles that act as nucleation centres for metal deposition. This process is purely chemical in nature and is referred to as ‘electroless’ (meaning that no electrode is required). After electroless deposition, electroplating of the tracks can also be performed if extra-thick tracks are required. This involves passing a current through a solution and the existing metal tracks, causing additional metal atoms to be attracted to the tracks, which are incorporated into them. LPKF report that they can achieve through-hole plating of 3D-MIDs using LDS, but they do not specify the process by which this is done. It is possible that they drill through-holes, then use a laser to perform LDS on the siliconchip.com.au exposed surfaces before the electroless and electroplating processes. Chemical process for two-shot injection moulding In the case of a 3D-MID made using two-shot moulding, metallisation is done via a chemical process rather than laser direct structuring. One of the plastics contains a catalyst that is metallisable, while the other does not contain the catalyst. The presence of the catalyst in one of the plastics causes metal deposition on that part when it is immersed in an appropriate chemical bath. Two examples are presented in Fig.20. On the left, the first material to be injected is metallised. On the right, the material that is injected second is metallised. 3D assembly Components have to be placed on 3D-MIDs and IMSE structures for soldering. This is done using 3D ‘pick and place’ machines, which can operate in three dimensions rather than just two as required for conventional PCBs. An example of such a machine is the Yamaha S20 – see siliconchip. au/link/ac3t With IMSE, the main circuit carrier component is mostly made in a single operation, unlike 3D-MID, which requires several operations. Electrically conductive tracks are incorporated at the time of moulding. Discrete electronic components, such as LEDs, can even be incorporated at the same time. IMSEs do have depth, but they tend to be flatter than 3D-MIDs in most applications. IMSEs typically start as 2D films, which may contain a printed design comprising artwork, labels for buttons or conductive or insulating tracks. Then additional plastics processing methods are used to convert them into more complex 3D shapes. The IMSE manufacturing process steps are: 1. The component is designed with appropriate CAD software. An example of one such CAD package is TactoTek IMSE Designer, which is intended for designing IMSE lighting devices for automotive applications (see siliconchip.au/link/ac3p). Another is Altium Designer, which was already mentioned. 2. Screen, inkjet printing or another form of printing is used on a plastic film. Decoration and/or labels are applied to a flat piece of plastic using a printing process; screen printing is the most common. This is followed by an additional printing process to apply electrically conductive tracks, similar to the tracks on a PCB. Special metal-laden inks are used – see Fig.22. 3. Components are placed onto the printed film using pick-and-place equipment. The components are attached with adhesive and electrical connections are made via conductive inks – see Fig.23. 4. The device is thermoformed using heat and an appropriate moulding to form the required 3D shape. WeLDS technology WeLDS is a technology developed by LPKF that combines LDS with laser plastic welding. It creates unique structures by welding 3D-MIDs to other plastic structures – see Fig.21. Australia's electronics magazine Fig.21: an example of WeLDS technology, with a device made by 3D-MID laser welder to another plastic structure. Source: www.lpkf. com/en/welds April 2025  17 Figs.22-25: (1) the tracks are laseretched onto a plastic film; (2) the components are then mounted around the periphery using a pick-and-place machine; (3) thermoforming is done to the part; (4) injection moulding seals the circuitry and gives extra structural rigidity. Source: www. tactotek.com/technology Within these structures, the typical layers of an IMSE part may include: ¬ A film on the top, bottom, or both. ¬ Electronics on the top (or bottom) film, or both. ¬ Injection moulding resin. IMSE can be combined with IMD graphics for, say, a control panel. These are printed on a film which is then placed in the mould cavity and incorporated into the moulded part. A manufacturer in the field, Tacto­ Tek, has a theme of “smart surfaces” to describe their use of IMSE technology. Fig.8 is one example. Also see https:// youtu.be/eGxkby9MBIM Some advantages of IMSE products are said to be a reduced part count, higher durability, reduced assembly time, more simple assembly, weather resistance, reduced weight and thickness compared to other methods. It is also possible to build illumination into the product. Printable inks for IMSE Conductive inks for IMSE contain metal particles such as silver, which is quite expensive. SmartInk from Genes­ Ink (www.genesink.com/smartink) is an example of a silver-containing ink Thermoforming is a process that for IMSE applications. Another such involves heating plastic to its soften- ink is from Dycotec (siliconchip.au/ ing point and then moulding it into a link/ac3u). shape – see Fig.24. Some conductive inks contain Care must be taken in the design graphite or carbon. For transparent stage to ensure that deformation conductors, indium tin oxide (ITO) during the forming process is not so can be used. It is see-through and can great that it causes the printed tracks be ‘printed’ using physical vapour to be excessively deformed and they deposition, electron beam evaporation become non-conductive. This pre- or sputter deposition. cludes shapes with excessively sharp ITO is expensive, so alternatives angles or other areas of high deforma- such as aluminium-doped zinc oxide tion. Care must also be taken so placed (AZO), indium-doped cadmium components remain on flat sections. oxide and carbon-based materials like 5. The thermoformed component graphene and carbon nanotubes are from the previous step is placed in an being explored as substitutes. Carbon-­ injection moulding machine, where it containing inks can also be used for is overmoulded to seal the electronics static dissipation. and circuitry, and to give some strucDielectric inks are also used for tural rigidity – see Fig.25. insulation purposes. Other materials 6. The component is trimmed to used include electrically conductive remove excess material and bring it to adhesives. its final shape ready for use. Due to the high cost of silver, it is Typical examples of structural desirable to find appropriate substioptions with IMSE devices are: tutes. Substitutes that are being inves• A two-film structure with a film tigated are copper, aluminium and on top and bottom, and injection-­ nickel, of which copper is the most moulding resin in between. promising; it is only about 1% of the • A film on the top and injection cost of silver. It has been used to some moulding resin on the bottom. extent. • A film on the bottom and injection A major disadvantage of copper moulding resin on the top. is its tendency to oxidise over time. 18 Silicon Chip Australia's electronics magazine Approaches to improving the oxidation resistance of copper-containing inks include: • Coating copper micro and nanoparticles with various substances. • Using antioxidant additives. • Using copper nanowires. • Making mixtures of copper nanoparticles with other substances like carbon nanotubes. • Sintering copper powder or copper compounds using a laser or flashlamp to make a contiguous copper layer like on a PCB. Non-metallic conductive inks are also possible, such as those made with the conducting polymer poly(3,4-­ ethylenedioxythiophene) mixed with polystyrene sulfonate. This is referred to as PEDOT:PSS. 3D-MID vs IMSE 3D-MID and IMSE have their advantages and disadvantages. 3D-MID tends to be used when miniaturisation, high reliability and a 3D structure is required. IMSE devices tend to be flatter, although still three-dimensional, and are more suited to control panels and other human interface devices, including ‘smart surfaces’. Both technologies have many applications across aerospace, automotive, medical and consumer electronics. There are no hard and fast rules about which technology should be used where. It comes down to cost, designer intent, volume and complexity. SC Companies Celanese (www.celanese.com/ products/micromax) for inks Cicor (siliconchip.au/link/ac3v) Contag (www.contag.eu) Distributed Micro Technology Ltd (www.dmtl.co.uk) Dycotech (siliconchip.au/link/ac3u) DuraTech (www.duratech.com) Eastprint (www.eastprint.com/ in-mold-electronics) Essemtec (https://essemtec.com) GenesInk (www.genesink.com) Harting (https://3d-circuits.com). See their video on 3D-MID at https://youtu.be/DcjGGJlc81I LPKF (www.lpkf.com/en) Lüberg Elektronik (www.lueberg.de) Sun Chemical (siliconchip.au/link/ ac3w) for inks TactoTek (www.tactotek.com) siliconchip.com.au Measuring tools for now and the future DIGITAL READOUT 7” Colour LCD Screen Colour Display Multiple Pre-Set Colours ZERO Programmable Up To 3 Axis One Touch Axis Zero Keys SCAN HERE FOR MORE INFORMATION Multi Language Menu 2-Year Warranty 319 (Q8500) $ SAVE $33 Measuring Box set 70-605 120mm Compact Linear Scale - MX-500-120/5U Touch Point Sensor - TPS-20 • CNC machined for high accuracy • Ground measuring face • Black anodized coating for a protective anti rust coating • Precision laser engraved markings • Compact Scale • Glass scale with 5µm resolution • 3m connection cable • Accuracy within 0.005mm • Ø10mm hardened & ground ball end • LED Light & Beeper Sensor $ 180 (Q8510) $ 79 (M690) $ SAVE $18 SAVE $14.50 SAVE $22 Digital Height Gauge 36-2105 Digital Caliper 31-188 Digital Caliper - Coolant Proof - 31-1851 • 0-300mm / 12” range • Carbide tipped scriber • Metric/imperial conversion • 0.01mm/0.001” resolution • ±0.04mm accuracy • 600mm / 24" • Large clear LCD On/ Off, Metric/Imperial. Zero switches - Zero setting at any position • 150mm / 6” • IP67 coolant proof • Large clear screen display • Metric/Imperial conversions • Four way measurement 550 (Q2105) 450 (Q188) 99 (Q670) 198 (Q1851) $ $ $ $ SAVE $19 SAVE $99 SAVE $89 SAVE $44 View and purchase these items online: www.machineryhouse.com.au/SIC2503 SYDNEY BRISBANE MELBOURNE (03) 9212 4422 (08) 9373 9999 PERTH ADELAIDE 1/2 Windsor Rd, Northmead 625 Boundary Rd, Coopers Plains 4 Abbotts Rd, Dandenong 11 Valentine St, Unit 11/20 Cheltenham Pde Woodville SA 5011 (02) 9890 9111 (07) 3715 2200 Specifications and prices are subject to change without notification. All prices include GST and vild until 27.04.25 Kewdale (08) 9373 9969 01_SC_270325 80 (Q605) Refractometer - 70-670 • Scale 0-18 Brix • Measures coolant to water ratio • Automatic temperature compensation • Differentiates between old and new coolants Mini Projects #024 – by Tim Blythman SILICON CHIP Discrete 555 Timer The 555 is one of the best known ICs; it was designed over 50 years ago but is still in production and use. It has many uses beyond its original intent as a timer. Our circuit closely approximates the operation of the main features of the 555 timer, allowing classic 555 designs to be investigated. W e have seen circuits and even kits that attempt to be faithful to the internal workings of the 555; our intention with this circuit is to see how easy it is to implement the workings of an integrated circuit (IC) using just a few components on a breadboard. It is not a direct replacement for a 555, but it will allow many 555 circuits to be built and investigated. We have favoured simplicity over exactness. Our circuit does not have all the features of even the cheapest 555 chip. We’ve simulated and tested it at 5V, and we know that it works from about 4V to 6V. It should work at higher voltages, too, but we’re specifying some 10V capacitors, so you would need to change that for operation above 9V. It lacks a RESET input and pushpull OUTPUT, but these are not needed in the most common applications. It wouldn’t be hard to add them, but we felt they would detract from the simplicity. You can see our circuit in operation by watching the video at: siliconchip. au/Videos/Discrete+555 comprises a handful of components. Three identical resistors connected in series produce voltages at 1/3 and 2/3 of the supply. There are two comparators and a latch; these are the core components used for timing. In the typical astable configuration (Fig.2), the TRIGGER and THRESHOLD pins are connected to a capacitor, C. The capacitor charges via the two resistors until it reaches 2/3 of the supply voltage, triggering COMPARATOR 1. This activates the latch and thus the DISCHARGE transistor. The capacitor then discharges until its voltage (and thus TRIGGER and THRESHOLD) drops below 1/3 the supply voltage and COMPARATOR 2 is triggered. The latch changes state and the DISCHARGE transistor switches off, allowing the voltage to rise and the cycle to continue indefinitely. The 555 timer The block diagram of a 555 timer (Fig.1) is a good place to start. Even a simple IC like this has its own building blocks. Each of these blocks Fig.1: the 555 IC comprises these internal building blocks. Our version lacks the reset function and output driver, although it includes an indicator LED to show the output state. 20 Australia's electronics magazine Silicon Chip siliconchip.com.au Fig.3: it’s remarkable that the building blocks shown in Fig.1 can be reduced to two or three transistors and a handful of resistors. The real 555 has many more transistors, making it a lot more tolerant of supply voltage variations and other operating conditions. Fig.3 is our circuit, with the blocks marked to align with Fig.1. The three 10kW resistors in series create the 1/3 and 2/3 supply voltage references. The components at right are the extra ‘external’ components needed to set up the circuit as an astable multivibrator. Each of the comparators consists of two transistors and two resistors, with one comparator having an extra transistor to invert its output. In each comparator, the two transistors form a differential pair. All the current through the pair must flow through the top resistor, which Fig.2: just three external components are needed to turn the 555 IC or our circuit into an oscillator. siliconchip.com.au connects to the emitters and is split into separate collector circuits. The current through each collector will thus depend on whether each transistor is conducting. With their emitters at the same voltage and since the emitter-base junctions are effectively silicon diodes, whichever base is at a lower voltage will conduct substantially more of the current. That will switch on that transistor, allowing current to flow through the corresponding collector. Q2’s base is set to 3.3V by the divider. If Q1’s base voltage is lower than that, Q1’s collector will carry all the remaining current coming through the emitter resistor. No current flows through Q2’s collector, and it sits near 0V. If Q1’s base rises above 3.3V, then current flows down Q2’s branch instead, causing the voltage on Q2’s collector to rise due to current through the 10kW resistor. The other differential pair works similarly, although its output is instead fed into an inverter (Q7 and its collector resistor) so that the TRIGGER output goes high when Q3’s base falls below the 1/3 level. Instead of resistors, a real 555 IC uses current sources and current mirrors, allowing the circuit to work better over a wider range of voltages, but resistors are simpler. The latch Transistors Q5 and Q6 plus four resistors form a bistable latch. This Australia's electronics magazine is effectively a form of memory that retains its state unless it receives an external signal to change. If one transistor is on, it pulls the base of the other transistor low, forcing it off. This is positive feedback, reinforcing the current state of the circuit. To change the state of the latch, an external signal supplies base current to switch one of the transistors on, forcing its counterpart to turn off. Here, we use diodes to inject current from each of the differential pairs into either side of the circuit. The last thing needed to use our timer circuit in the classic 555 astable configuration is a DISCHARGE output. This is simply an NPN transistor in the same open collector configuration seen in Fig.1. We’ve also added transistor Q9 to drive LED1 to show the state of the circuit. It also helps to even out the load on Q5 and Q6 so that they behave symmetrically. Astable oscillator All that is needed to create an astable multivibrator (or oscillator) is to add the parts on the right-hand side of Fig.3; these are the same minimal parts needed to turn a 555 IC into an oscillator. They do a job very much the same as in an IC-based circuit. The capacitor starts in a discharged state, meaning that TRIGGER and THRESHOLD are both low. Importantly, the TRIGGER voltage is less than 1/3 supply, so the current flows April 2025  21 through D1, meaning that Q5 is on and Q6 is off. DISCHARGE (Q8) is off and the capacitor can charge through the resistors. Q9 and the LED are on. At 1/3 supply, the TRIGGER comparator stops supplying current to D1, and the latch keeps its current state. At 2/3 supply, the THRESHOLD voltage is passed and current now passes through D2, switching on Q6 and switching off Q5. DISCHARGE switches on too, and the capacitor discharges until 1/3 supply is reached. The cycle then repeats. Parts List – Discrete 555 Timer (JMP024) 1 breadboard or prototyping board [Jaycar PB8820] 4 BC557 100mA PNP transistors (Q1-Q4) [Jaycar ZT2164] 5 BC547 100mA NPN transistors (Q5-Q9) [Jaycar ZT2152] 2 1N4148 75V 200mA small signal diodes (D1, D2) [Jaycar ZR1100] 1 yellow 3mm LED (LED1) [Jaycar ZD0110] 11 10kW ¼W axial leaded resistors [Jaycar RR0596] 1 4.7kW ¼W axial leaded resistor [Jaycar RR0588] 1 2kW ¼W axial leaded resistor [Jaycar RR0579] 4 1kW ¼W axial leaded resistors [Jaycar RR0572] 2 220μF 10V electrolytic capacitors [Jaycar RE6157] 1 5V DC power supply Hookup wire or jumper wires Construction We laid our circuit out on a breadboard, since we expect readers will want to change the circuit to test out its operation. It could be transferred to a prototyping PCB like Jaycar’s Cat HP9570 instead. Our Parts List includes the wiring and the parts needed to use the circuit as an oscillator; Fig.4 shows the layout we used. Q1-Q4 are the PNP transistors; we used BC557s, but any of the BC55x series parts should work. Similarly, Q5-Q9 are BC547 NPN transistors that can be substituted with any BC54x equivalent. The red arrows show the external ‘pins’, with power and ground being supplied through the side power rails. All power links are shown in red, with ground in black. Other internal connections are blue. Note the power links at the top of the breadboard. The green wires and three components at upper left are the added components needed to turn the circuit into an oscillator. The values shown here should cause the LED to flash at a rate of about 1Hz. While building your version, you can also refer to our photos. If you don’t see anything happen when you apply power, check your wiring. You can probe the circuit with a multimeter to see what might be wrong. Testing We started by building our circuit in the LTspice circuit simulator. It is free to use and can be downloaded from siliconchip.au/link/ac2p We published a series of articles about LTspice in 2017 and 2018 (siliconchip.au/Series/317). It’s a great way to test out circuit configurations and values before going to the trouble of plugging components into a breadboard. You can try our simulation file to see how the circuit operates (download from siliconchip. au/Shop/6/1821). Scope 1 shows the output of the simulator. You can see the two comparators briefly activating in turn and toggling the state of the latch. The waveform is oscillating at 1.24Hz. The calculated frequency for a 555 timer in this configuration with these components is 1.32Hz. We suspect the reason our version is a bit slower than expected is that it slightly overshoots the 2/3 supply voltage threshold. If you are going to experiment, we suggest sticking with external resistors similar in value (around 1kW) to the ones that we have used. Conclusion The comparator and latch are very common building blocks in all types of circuits. Here, you can see how they can be combined to create a simple but flexible circuit that can SC do many jobs. Scope 1: our LTspice simulation of the astable multivibrator. The grey and purple traces are the 1/3 and 2/3 supply reference voltages, while TRIGGER and THRESHOLD follow the green trace (since they are connected together). The red and cyan traces are the outputs of the comparators that trigger the latch to change state. 22 Silicon Chip Australia's electronics magazine siliconchip.com.au Fig.4: this is a simplified version of the classic 555 timer IC that you can build from a couple of dozen components. Follow this diagram closely, since many of the components are close together. Observe the type and orientation of the transistors. The PNP transistors on the right have their emitters joined and thus they share a row. The ‘external’ timing parts are those at upper left plus the green wires. ourPCB LOCAL SERVICE <at> OVERSEAS PRICES AUSTRALIA PCB Manufacturing Full Turnkey Assembly Wiring Harnesses Solder Paste Stencils small or large volume orders premium-grade wiring low cost PCB assembly laser-cut and electropolished Instant Online Buying of Prototype PCBs www.ourpcb.com.au siliconchip.com.au Australia's electronics magazine 0417 264 974 April 2025  23 PICO/2/COMPUTER > Design and firmware by Peter Mather > Words and MMBasic by Geoff Graham This computer uses the latest and greatest Raspberry Pi Pico 2. Like the PicoMite, it can be programmed in MMBasic, but it has a lot of extra features like a HDMI video connector, multiple USB sockets for devices like keyboards, mice and game controllers, and an audio output. I n our February issue this year, we introduced the latest version of the PicoMite firmware for the Raspberry Pi Pico 2. It includes advanced features such as support for HDMI video and USB keyboards. In this design, we bring these elements together to provide a high performance ‘boot to BASIC’ computer that is fast and very capable. This type of computer was popular in the 1970s and 1980s (and still is today!), including examples such as the Apple ][, Commodore 64, Tandy TRS-80 and many others. These computers all included a BASIC interpreter and, when the computer was powered up, booted straight to the BASIC command prompt. There, 24 Silicon Chip you could enter programs, test commands and immediately be productive. The Pico 2 Computer is just as easy and fun to use but much, much more powerful! It includes everything that you need to get started. That makes it ideal for learning to program, entertaining/teaching children about computers and just having fun exploring its capabilities. This Computer can also be used for more than just running calculations. You can use it to interact with the physical world, including measuring voltages, detecting switch closures and driving outputs to light LEDs, play music, generate sound effects and more. Australia's electronics magazine This is a ‘reference design’. By this, we mean that it can be used as a template for a computer of your own design with the best of its features. If you want to ‘roll your own’, you are welcome to take elements of this design, modify them as needed and incorporate them in your own creation. You can also simply build this computer as-is, and you will have a capable and high-performance boot-to-­BASIC computer. Features The video output is HDMI-compatible in one of three resolutions: 640 × 480, 1280 × 720 (wide-screen) or 1024 × 768 pixels. At these resolutions, the siliconchip.com.au output is monochrome. However, by using the MODE command, you can select more colours at lower resolutions. The built-in BASIC program editor uses the full resolution yet, by using the TILE command, it will colour the characters for you. For example, it uses cyan for keywords, green for comments etc. This makes for a colourful and intuitive editing experience at the full screen resolution. The USB keyboard input has full support for the function keys, arrow keys etc. Our previous computers used PS/2 keyboards, but they are becoming difficult to find these days, so support for a USB keyboard is a welcome addition. This facility extends to wireless keyboards with a USB dongle, so you do not need to be tethered by a cable. This design includes a four-port USB hub with four USB Type-A sockets so that you can add additional devices – primarily a USB mouse and USB game controllers. The mouse is most useful when using the built in MMBasic program editor, where it gives you an almost GUI-like experience, with the ability to position the insert point and copy and paste – all using the mouse. As with the keyboard, you can also use a wireless mouse. One or more USB game controllers can also be plugged in. Within a BASIC program, you can query the position of the joystick, the state of the buttons etc. So, if you are into writing games, you can create the full arcade (or home games console) experience. More features A highly accurate battery-backed real-time clock is included in the design. This means that the Pico 2 Computer will always know the correct time. You might use this within a program, but it is also useful in that all files created by MMBasic will be stamped with correct creation times. Speaking of files, the Pico 2 Computer includes a microSD card socket with support for cards formatted in FAT16 or FAT32, and capacities of up to 32GB. Files written on these can be read/written by Windows, Apple and Linux computers, so this is an easy way of transferring files to and from larger desktop or portable computers. siliconchip.com.au The Pico 2 Computer is built on a 90 × 100mm PCB. It uses surface-mounting components and can be hand-built or machine-assembled. The only component that must be soldered by hand is the Raspberry Pi Pico 2 module. Built into the PicoMite firmware is an internal drive (called drive A:) that uses the flash memory in the Raspberry Pi Pico 2. This is about 2MiB (ample for normal use), so this is another place to save program and data files, particularly during program development. On the back panel is a stereo audio output connector. This provides a high-quality audio output of about 3V peak-to-peak, suitable for feeding a HiFi system or amplified speakers. Within BASIC programs, you can create various tones and sound effects as well as stream music files in WAV, FLAC, MP3 or MOD formats located on the internal filesystem or SD card. To connect to the outside world, the Pico 2 Computer has 14 input/output (I/O) pins that can be used as digital inputs or outputs. As inputs, they can monitor switches and sensors (humidity, temperature, location and more). As digital outputs, these I/O pins can drive LEDs and powered relays to switch heavy loads. Three of these pins can also measure voltages, so you can monitor signals Australia's electronics magazine from the analog world, such as battery charge levels. The PicoMite firmware We described the new PicoMite firmware for the Raspberry Pi Pico 2 in detail in the February 2025 issue (siliconchip.au/Article/17729). At its core is the MMBasic interpreter, which makes it easy to write programs in the BASIC language. BASIC is an easy-to-learn language so it, and the Pico 2 Computer, are ideal for someone who wants to get into programming and learn the basics. At the same time, it is quite powerful, so you can develop large and complex programs ranging from controlling physical processes to calculating the positions of the planets. It is also an ideal platform for creating graphical computer games, ranging from classics such as Tetris and Chess through to more advanced 3D simulations. The PicoMite firmware includes support for multiple video layers and graphical objects such as sprites. For anyone familiar with the home April 2025  25 computers of the 1970s and 1980s, this will all be recognisable. The difference is that the Pico 2 Computer is much faster and has many more resources that these early computers. At roughly 100 times faster and with 10 times the memory, this computer is something that the programmers of the 1970s and 1980s could only dream about! The PicoMite firmware is fully self-contained. You do not need an operating system or other external programs. It includes its own feature-rich 26 Silicon Chip program editor and drivers for all the I/O devices (SD cards, clock, audio, USB devices, video display etc). Circuit details As shown in Fig.1, at the heart of the Pico 2 Computer is the Raspberry Pi Pico 2 module. It is amazing value; for a little over $8, you get a dual-core 32-bit CPU capable of running at up to 400MHz, including 520KiB of built in RAM. A separate chip on the module provides 4MiB of flash memory for programs and general storage. Australia's electronics magazine This module runs the PicoMite firmware, including the BASIC interpreter, with the rest of the circuit being primarily used to interface it to the video, keyboard and some specialised components. A supervisor device (MAX809R) monitors the 3.3V power rail (VDD) and provides a reset signal to the RP2350A microcontroller to ensure that it is cleanly shut down when the power is removed. It will drive the reset pin of the Pico 2 low within 65µs of the 3.3V power rail falling below siliconchip.com.au circuits and an ingenious scheme for driving the five status LEDs using just three outputs. The circuit includes a 12MHz crystal oscillator that is used by the CH334F to create the accurate timing required by the USB standard. To load MMBasic on the Pico 2, you need to disconnect the hub and directly access the USB interface on the Pico 2. Of course, you can do this before mounting the Pico 2 module but it may become necessary to do this again later (eg, to update the firmware). So jumpers JP1 & JP2 allow you to isolate the hub, and you can use the extra micro-USB connector on the PCB’s edge (CON5). Serial console Fig.1: the Raspberry Pi Pico 2 module is at the core of this design, with the rest of the circuit providing the video, keyboard, microSD and external I/O interfaces. Other features include a real-time clock, an integrated USB four-port hub and a dedicated serial-to-USB bridge for the serial console. 2.63V, and will maintain it low until it is above that threshold for at least 140ms. USB interface The PicoMite firmware uses the USB interface integrated in the RP2350A processor on the Pico 2 to provide support for a USB keyboard. However, in this design, we also wanted to provide for a USB mouse and gamepads in addition to the keyboard. To do this, the Pico 2 Computer includes a CH334F integrated USB 2.0 four-port hub. siliconchip.com.au The connection between the CH334F and the USB interface on the Raspberry Pi Pico 2 is made by soldering through three holes on the PCB to connect to test pads on the underside of the Pico 2 module. These pads provide the USB interface, meaning we do not need to plug anything into the module’s USB connector. The CH334F includes the USB 2.0 driver circuits (called USB PHYs) that connect to the four USB Type-A sockets on the front panel. The CH334F also includes the required protection Australia's electronics magazine Because the USB interface on the Raspberry Pi Pico 2 is used for communicating with various USB devices such as the keyboard, it cannot be used for the serial-over-USB console used by the BASIC interpreter to communicate with a desktop or laptop computer. In a self-contained computer like this one, the serial console is not critically important, as the MMBasic console output will display on your HDMI monitor anyway. However, having the serial console is handy for connecting to a desktop or laptop computer, so the Pico 2 Computer uses a CH340C serial to USB bridge to provide the serial-over-USB console interface. The CH340C converts the TTL asynchronous serial signal from the Pico 2 (on pins GP8 and GP9) to a USB 2.0 signal using the CDC (Communication Device Class) protocol. The CH340 is in an SMD SOP-16 package that includes the crystal and oscillator required for USB timing. This chip (and the CH341, which is similar) is used in many Arduino Nano clones, and the driver for it is included by default in Windows 10/11 and Linux. Also, many macOS builds include the driver. Video and audio The HDMI connector is driven by output pins on the Raspberry Pi Pico 2 via 220W resistors. The Pico 2 and the HDMI connector are positioned close to each other to reduce the track lengths and the chances of crosstalk and interference. The stereo audio output is generated April 2025  27 Power switch 14 GPIO (General Purpose I/O) + 3.3V & 5V pins Reset switch Stereo Audio 1V RMS HDMI Video (up to 1280 × 720) USB-C power input and serial console Raspberry Pi Pico 2 with BASIC interpreter Built-in editor Real-time clock using a CR2032 cell 180KiB of program space 228KiB of RAM Four-port USB hub IC MicroSD card up to 32GB Micro USB for firmware loading 4 × USB ports The Pico 2 Computer is a fully featured reference design with HDMI video, four USB ports for keyboards, mice and game controllers, a battery-backed real-time clock, microSD card socket and 14 externally available I/O pins. by the PicoMite firmware using pins GP10 and GP11. It is a PWM (pulsewidth modulated) signal that passes through a multi-pole low-pass filter to remove the carrier frequency. This filter, along with the low-noise regulator used to generate the 3.3V rails, results in a low noise audio signal capable of reproducing tones, sound effects and music with good fidelity from 10Hz to 15kHz. The maximum audio output level is 3V peak-to-peak (approximately 1V RMS) and is intended to be fed to amplified speakers. The amplifier used must have a capacitor-coupled input (most do), as the output signal has a DC offset. Timekeeping is provided by a DS3231 real-time clock (RTC), which is an extremely accurate timekeeper with an integrated temperature-­ compensated crystal oscillator (TCXO). It will typically keep the time accurate within a few seconds per month. It also includes a comparator circuit that monitors the status of the power supply and will automatically 28 Silicon Chip switch to the backup battery to keep the clock running when power is removed. The battery used for this is a non-­rechargeable 210mAh 3V lithium coin cell (CR2032), which should be good for many years of use. External I/O There are 14 input/output pins on the rear panel of the Computer that connect to pins on the Pico 2, which can be used as digital inputs or digital outputs. Some of these can also be used as PWM outputs as well as I2C, SPI and asynchronous serial communications channels. Three of these pins can be used as analog inputs, with the ability to measure voltage in the range of 0-3.3V. All these features are configured by the BASIC program running on the computer. The last component of note is the microSD card socket, which connects to another set of I/O pins on the Pico 2. The PicoMite firmware uses the SPI protocol to talk to the card and this is not influenced by the card type, so all Australia's electronics magazine types (Class 4, 10, UHS-1 etc) with a capacity of up to 32GB are supported. Power supply The power requirement for the Pico 2 Computer is 5V (typically drawing 170mA), which can be supplied via a normal USB charger with a USB Type-C plug. This 5V supply is used by the physical USB ports, but the rest of the computer runs from 3.3V, and this is provided by an AMS1117 low-dropout three-terminal linear regulator. There are many other compatible regulators, such as the LD1117, NCP1117, STC1117 etc. They may have varying specifications like the maximum input voltage, but as the input supply is 5V in this circuit, any of those types would be suitable. The Pico 2 also has an onboard regulator capable of supplying the required 3.3V, but this is a switching regulator and the electrical noise generated by its operation causes noise on the audio output. It also degrades the ability of the ADC inputs to measure voltages accurately. That is why this design siliconchip.com.au You can also find the most up-to-date information on the Pico 2 Computer (including design files and firmware) at https:// geoffg.net/ picomitevga. html uses the dedicated linear regulator and the onboard switching regulator is disabled. Purchasing an assembled board While the Pico 2 Computer can be built by hand, it takes some dexterity. There are many small surface-­ mounting components, including the USB hub, which comes in a quad flat no-lead (QFN) package. There are also some passives as small as 1.2 × 0.6mm. If you have the skill, it is feasible to hand-solder these components. However, this project is primarily designed for automated assembly. We therefore won’t give any detailed assembly instructions. If you have the required skills, you should be able to use the overlay diagram (Fig.2) to build your board. JLCPCB is a major PCB fabricator based in China. Through their LCSC supply arm, they can even supply the components and solder them to the PCB they make using solder paste applicators, pick-and-place machines Fig.2: here is the and reflow ovens. PCB overlay in At the time of writing, JLCPCB will case you want make the PCB, supply, mount and to assemble the solder all the components except the board yourself. Pico 2 for about $150 for two boards This also shows (plus three spare PCBs). Over time, how to orientate this price may vary with exchange the Pico 2 even if you’re using rates and other factors, but it is still a a pre-built good price for an almost fully assemboard. If you’re bled computer. adding the parts This assembly even includes large manually, take components, such as the conneccare with the tors and switches that must be hand-­ orientations of soldered. The only assembly required the diodes, LEDs, by you is to solder the Raspberry Pi ICs and crystal. Pico 2 and load the firmware. Then There are three you are ready to go. You could remove different sizes of resistors some of the larger components from and four of the BOM (Bill of Materials) given to capacitors; the JLCPCB and save some money by solsmallest in both dering them yourself. cases (1.2 × The process of ordering the assem0.6mm) can be bled boards is simple. First, download hand-soldered, three files from the Silicon Chip webbut not easily. site. These are “Pico 2 Computer Gerbers.zip”, which contains the design files for the PCB, “Pico 2 Computer BOM.xlsx”, which is the Bill of Materials, and “Pico 2 Computer CPL.xlsx”, which has the types and positions of the components on the PCB. On the JLCPCB website (https:// jlcpcb.com), click on the Instant Quote button and drag the “Pico 2 Computer The front of the Pico 2 Computer mounted in its small matching instrument Gerbers.zip” file onto the blue button case. siliconchip.com.au Australia's electronics magazine April 2025  29 labelled “Add Gerber File”. JLCPCB will then read the files and display an image of the front and back of the PCB. The website will also fill in the defaults for the PCB, such as thickness, colour etc. You can leave these as suggested. Scroll to the bottom of the page and select “PCB Assembly”. This will display more options, which you can leave at their default, other than selecting how many boards that you want them to fully assemble (I recommend two). Then click on the “Next” button on the right and the website will display a new page showing a large image of the board. Then click on the “Next” button again. On the next page, drag and drop the “Pico 2 Computer BOM.xlsx” file onto the button labelled “Add BOM File”, and drag and drop the “Pico 2 Computer CPL.xlsx” onto the “Add CPL File” button. Then click on the “Process BOM and CPL” button. The website will then show a list of all the parts required, the quantity that JLCPCB has in stock and their associated prices. All the components on the Bill of Materials are JLCPCB catalog items, and they should all be in stock. If, for some reason, a component is not available, you have the choice of leaving it out and sourcing it yourself. Alternatively, you could search JLCPCB for a substitute, perhaps one with a slightly different specification. At this stage, you can also choose to omit components that you wish to hand-solder to save cost. At the bottom of the page, click “NEXT” and you will be taken to a page that shows an image of the completed board. Clicking “NEXT” again will take you to the final quote detailing the total price. If you are happy, click “SAVE TO CART” and then proceed to give them your address and pay. Mounting the Pico 2 The one component that JLCPCB does not include in the assembly is the Raspberry Pi Pico 2 module. This is surface-mounted flat on the PCB. A special characteristic involved in this design is that you need to solder through three holes in the PCB to connect the USB pads on the underside of the Raspberry Pi Pico 2 module to the PCB. For that to work, you need to closely follow the instructions below. First, accurately position the Pico 2 module on its pads and, while holding it very flat on the PCB, tack-solder one corner pad. Check the alignment and, if it is still correct and the module is still flat on the PCB, tack-solder the opposite pin. With the Pico 2 module securely fastened, turn over the PCB and locate the three solder pads with plated through-holes identified as A, B and C in Photo 1. Apply plenty of liquid or paste flux in these holes and melt fine-gauge solder wire into them. The solder should flow through the holes and adhere to the three gold-plated pads on the underside of the Raspberry Pi Pico 2 module. While you are doing this, identify a similar plated through-hole designated D in Photo 1 and similarly apply flux and run solder into this hole. This connects to a heatsink pad on the underside of the CH334F USB 2.0 four-port hub. Soldering to this pad will assist in keeping that chip cool. Then work down the Raspberry Pi Pico 2 and solder the remaining solder pads. Finally, return to the first two pads that were tack-soldered and re-solder them securely. Figs.3 & 4: these are the panel cutouts required for the end panels of the instrument case. Pre-made panels are available that already have these holes neatly made and labels printed on them. All dimensions are in millimetres. Figs.5 & 6: the artwork for the end panels of the instrument case at actual size. You can also download these as a PDF from the Silicon Chip website (siliconchip.com.au/ Shop/11/1834). See siliconchip.au/ Help/FrontPanels for details on producing and attaching labels. 30 Silicon Chip Australia's electronics magazine siliconchip.com.au Finishing assembly To finish the assembly, you need to place jumpers between pins 1 & 2 and pins 3 & 4 on the four jumper pins near the CH334F four-port hub chip (IC20). These connect the hub chip to the USB RP2530A processor and only need to be removed when you are loading or upgrading the PicoMite firmware. Jumpers also need to be placed on the group marked SELECT SD PINS. For normal operation, place jumpers to connect GP26, GP27 and GP28 to their respective centre pins. These can be changed to GP2, GP3 and GP4 if you want to have the analog input pins usable on the external I/O port. The completed PCB can be used without an enclosure. In that case, you can attach plastic tapped spacers or standoffs to the four mounting holes to stop it from scratching your desk. However, the PCB is designed to fit in a Multicomp MCRM2015S enclosure available from element14/Farnell. The compatible Hammond RM2015S is available from DigiKey, Mouser etc. If you’re putting the board in one of these cases, you will need to make the cutouts in the front and rear panels as detailed in Figs.3 & 4, then print and apply the artwork depicted in Figs.5 & 6. Alternatively, you could purchase PCB-based front and rear panels from the Silicon Chip Online Shop and save yourself the effort of making all those holes (this also guarantees neatness). The labels will be printed on those panels, although they will only be labelled in white. These panels will also be included our kits (along with the pre-assembled PCB). Photo 1: the A, B & C solder pads connect to associated gold-plated pads on the underside of the Pico 2. Hole D connects to a heatsink pad on the underside of the CH334F USB 2.0 four-port hub. All four should be fluxed and solder run into the hole to make the connections. Loading the firmware With the hardware assembled, you can load the PicoMite firmware. For this, you need the firmware file “PicoMiteHDMIUSBV6.00.01.uf2”, which is included in the PicoMite firmware zip file downloadable from the Silicon Chip website (siliconchip. au/Shop/6/833) or the author’s website at http://geoffg.net/picomite.html (scroll to the bottom of the page). To load the firmware, you need to remove the two jumpers previously placed on the header near the CH334F four port hub chip. Then, while holding down the white button marked BOOTSEL on top of the Raspberry Pi Pico 2 module, plug your desktop or laptop computer into the connector siliconchip.com.au The Pico 2 Computer can be used without a case. With feet in the corners, it is at home driving a HDMI monitor with a USB keyboard, mouse & game controllers. The rear of the Pico 2 Computer mounted in its small matching instrument case. Australia's electronics magazine April 2025  31 For x = 0 To 16 For y = 0 To 16 For i = 0 To 16 Read d(x, y, i) ‘ load the shortest path database Next i Next y Next x StartGame: If TestMode <> 1 Then StartScreen Score = 0 NbrMen = 3 NextOrange = OrangeLevel OrangeX = 0 : OrangeY = 0 BonusPacMan = BonusLevel Level = 0 NewLevel: LoadData LostALife: ‘ scramble the ghost colours For i = 1 To 10 j = Int(Rnd * 4) : k = Int(Rnd * 4) l = GhostColour(j) : GhostColour(j) = GhostColour(k) Next i ‘ draw the ghosts on the screen and save them as a BLIT object F1: Save F2:Run F3:Find F4:Mark F5:Paste Ln: 27 Col: 55 INS Screen 1: the built-in editor is colour coded with cyan for keywords, green for comments etc. It includes a search facility, a clipboard for copy and paste, automatic indenting and more. If an error occurs, the editor will start with the cursor placed on the program line that caused the error. Screen 2: the Mandelbrot set (a fractal) was a favourite test for the home computer of the 1970s through to the 1990s. In the early days, it took some 24 hours to calculate it! The Pico 2 Computer drew this comparatively highresolution version in a little over four minutes. Coin Cell Precautions marked PROG on the front of the Pico 2 Computer’s PCB. When you do this, the Raspberry Pi Pico 2 should connect to your computer and create a virtual drive, as if you had plugged in a USB memory stick (you can ignore any files on this ‘drive’). Then copy the firmware file (with the extension .uf2) to this virtual drive. Once completed, the Pico 2 will restart and the LED on it will blink slowly, indicating that the PicoMite firmware is running. While the virtual drive created by the Raspberry Pi Pico looks like a USB memory stick, it is not; the firmware file will vanish once copied, and if you try copying any other type of file to it, that file will be ignored. If you later upgrade the firmware, you should be aware that this operation may erase all the flash memory, including the current program and any files in drive A:. So ensure that you back up all of your data before upgrading. Final setup Replace the pins on the jumper group near the CH334F chip and plug in a USB keyboard and HDMI monitor. Apply power to the USB Type-C connector on the rear edge of the PCB and depress the ON/OFF switch beside it to switch on the computer. You should then be greeted with the firmware’s copyright notice on the HDMI monitor. With the firmware loaded and the monitor and keyboard connected, you should see the command prompt (a greater than symbol, ‘>’) on the monitor. At this point, you can enter commands, run programs etc. However, before you jump in, two additional steps should be completed. The first is to enter the command OPTION RESET HDMIUSB. This will set the firmware options to suit this design and will save you a lot of time from having to enter each option individually. Following this, if you want to use the alternative SD card connection pins, enter the following commands at the command prompt: OPTION SDCARD DISABLE OPTION SDCARD GP22, GP2, GP3, GP4 If not mounting the Pico 2 Computer in a case, care should be taken so that the device is not left anywhere that children could get hold of it. Coin cells are very dangerous to children if they swallow them, and some will do so given the opportunity. Make sure that can't happen! The final action is to set the date and time in the real-time clock. The command to do this is entered at the command prompt and is: Australia's electronics magazine siliconchip.com.au 32 Silicon Chip RTC SETTIME year, month, day, hour, minute, second Parts List – Pico 2 Computer While most computers will already have drivers for the chip used for the console, if you find you do need a driver, help is available at https:// sparks.gogo.co.nz/ch340.html Note that the PicoMite firmware sets the console to 115,200 baud, so the terminal emulator running on your desktop computer will need to be set to this speed. In the meantime, have fun with your new computer! SC 1 double-sided PCB coded 07104251, 90 × 100mm 1 pair of front & rear panels made from FR4 PCB material with a black solder mask and white silkscreen printing (optional) [Silicon Chip SC7453] 1 Raspberry Pi Pico 2 (RP2350A) without header pins (MOD1) 1 Multicomp MCRM2015S or Hammond RM2015S instrument case (optional) OR 4 M3-tapped Nylon spacers and M3 × 6mm panhead machine screws (for feet) 1 CR2032 3V lithium coin cell (BAT1) 1 30V 750mA resettable polyfuse M3216 (PTC1) [BHFuse BSMD1206-075-30V] 1 latching right-angle PCB-mount pushbutton (S13) [XKB Connectivity XKB5858-Z-E] 1 right-angle tactile pushbutton switch with 6mm actuator (S15) [HCTL TC-6615-7.5-260G] 1 12MHz 20pF 10ppm 4-pin SMD crystal, 3.2 × 2.5mm (X1) [YXC X322512MSB4SI] Connectors 1 CR2032 cell holder (BAT1) [Myoung BS-04-A1BJ005] 1 HDMI socket (CON1) [HCTL HDMI-01] 1 USB-C socket (CON2) [Kinghelm KH-TYPE-C-16P] 2 right-angle horizontal stacked USB Type-A sockets (CON3, CON4) [Shou Han AF SS-JB17.6] 1 USB micro-B socket (CON5) [Shou Han MicroXNJ] 1 microSD card socket (CON6) [Shou Han TF PUSH] 1 SMD stereo audio jack socket (CON7) [Shou Han PJ-313 5JCJ] 1 2×12-pin right-angle 2.54mm-pitch header (CON23) [HanElectricity 2541WR-2x12P] 1 2×2-pin 2.54mm-pitch header (JP1, JP2) [JST RF-H042TD-1190(LF)(SN)] 1 3×3-pin 2.54mm-pitch header (LK1-LK3) [HCTL PZ254-3-03-Z-2.5-G0] 5 jumper shunts (JP1-JP2, LK1-LK3) Semiconductors 1 CH340C serial/USB bridge, SOIC-16 (IC7) 1 DS3231MZ real-time clock & calendar, SOIC-8 (IC19) 1 CH334F quad USB hub, QFN-24 (IC20) 1 MAX809R reset supervisor IC, SOT-23-3 (IC24) 1 AMS1117-3.3 or equivalent 3.3V low-dropout linear regulator, SOT-223-3 (REG1) 1 MDD2301 P-channel Mosfet, SOT-23-3 (Q1) 1 red SMD LED, M1608/0603 size (LED2) [KT-0603R] 5 green SMD LEDs, M2012/0805 size (LED3-LED7) [KT-0805G] 2 SS14 40V 1A schottky diodes, SMA package (D1, D2) Inductors & ferrite beads 1 M2012/0805 multi-layer ferrite bead (FB12) [Murata BLM21PG221SN1D] 2 10μH 15mA 1.15W M1608/0805 SMD inductors (L22, L23) [Sunlord SDFL2012S100KTF] 2 4.7mH 110mA 32.5W 5×5mm SMD inductors (L26, L27) [YJYCoin YNR5040-472M] Capacitors 2 220μF 10V D-case solid tantalum electrolytic [Kyocera AVX TAJD227K010RNJ] 3 10μF 50V X5R M3216/1206 ceramic [Samsung CL31A106KBHNNNE] 7 100nF 16V M1206/0402 X7R ceramic [Samsung CL05B104KO5NNNC] 6 33nF 50V M2012/0805 X7R ceramic [FH 0805B333K500NT] 2 2.2nF 50V M2012/0805 NP0/C0G ceramic [Samsung CL21C222JBFNNNE] 2 470pF 50V M1608/0603 X7R ceramic [FH 0603B471K500NT] Resistors (all SMD 1%) 1 1MW (M1206/0402 size) Pico/2/Computer Pre-Made 1 15kW (M1608/0603 size) Board (SC7468; $120 + post): 1 12kW (M1608/0603 size) 2 10kW (M2012/0805 size) Includes an assembled PCB, 2 5.1kW (M1206/0402 size) Raspberry Pi Pico 2 (which 1 4.7kW (M2012/0805 size) you need to attach to the PCB 1 470W (M1608/0603 size) yourself) and front & rear panels. 2 220W (M2012/0805 size) We currently plan to only supply 11 220W (M1608/0603 size) a limited amount of these kits. 1 2.2W (M2012/0805 size) siliconchip.com.au Australia's electronics magazine Here, ‘year’ is two or four digits and ‘hour’ is in 24 hour notation. Don’t forget to insert a CR2032 cell in the holder so it will keep time when the power is switched off. Using MMBasic On startup, MMBasic will issue the command prompt and wait for you to enter something. It will also return to the command prompt if your program ends or encounters an error. When the command prompt is shown, you have a wide range of commands that you can enter and execute. For example, you can list the program held in memory (LIST) or edit it (EDIT), or perhaps check the memory usage (MEMORY). The command RUN instructs MMBasic to run the program currently held in program memory. All of these and more are described in detail in the PicoMite User Manual, which is included in the firmware download package. Almost any command can be entered at the command prompt, and this is a good way to test a command to see how it works. A simple example is the PRINT command, which will simply print the result of a calculation. You can try this by entering the following at the command prompt: PRINT 1/7 MMBasic will print out the result of dividing 1 by 7 (ie, the number 0.1428571429) before returning to the command prompt. If you are new to the BASIC programming language, refer to Appendix I at the back of the PicoMite User Manual. This is a comprehensive tutorial on the language, which will take you through the fundamentals in an easyto-read format with lots of examples. Using the serial console April 2025  33 The Future of our Power Grid The first article in this series last month described how our electrical grid is changing, the pros and cons of the various types of generators, costs and Demand Response. This second and final instalment finishes the discussion by covering inverters and grid stability. L ast month, I explained how as coal and gas power stations reach their end-oflife, they are increasingly being replaced by other generation methods like wind and solar power. However, that transition is not without its challenges due to the way that generation varies over time, with changes in the weather and the day/night cycle. Thankfully, this transition is slow, which is allowing the deployment of various techniques and technologies to overcome those limitations. Energy storage and Demand Response were covered in that first article, but now we come to the nitty-gritty, such as the ways that solar and wind generators are connected to the grid to better match demand and improve grid stability. Solar inverters Inverters for photovoltaic panels take a DC supply from the solar array and convert it to AC to feed the grid. They typically use Insulated Gate Bipolar Transistors (IGBT) arranged in a three-phase H-bridge topology (see Figs.13, 14 & 15). The IGBT is effectively a small Mosfet and a large bipolar junction transistor (BJT) combined on a single die. By combining the two transistor types, the IGBT benefits from the advantages of both technologies; the BJT is well-suited to high-power applications due to its favourable output characteristics, and the Mosfet is a convenient way to provide base drive to the BJT given its high gate impedance. Using pulse-width modulation (PWM), a three-phase AC waveform can be synthesised from the input DC, similar to the operation of our Mk2 Variable Speed Drive for Induction Part 2 by Brandon Speedie Motors, published in the November & December 2024 issues (siliconchip. com.au/Series/430). Typically, the chopper frequency is in the order of 50kHz or so. It is filtered out by an LC network (usually a ‘pi’ or ‘T’ filter) on the output of the inverter to form a smooth sinusoidal waveform. Utility-scale solar farms receive further filtering from the inductance in their grid-connected transformers, which step up the low voltage output from the inverter to the high voltage of the transmission network. This synthesised AC waveform needs to be precisely controlled to synchronise with the grid. This is achieved by sampling the grid voltage to form a phase-locked loop, which becomes a reference waveform. By varying the amplitude and phase of the synthesised waveform with respect Figs.11 & 12: the topology of an AC-coupled hybrid solar and battery generator is shown in the left diagram. The alternative configuration of a DC-coupled hybrid solar and battery generator is shown at right. For the DC-coupled system, with sufficient irradiance, power can be exported to the grid and charge the batteries simultaneously without having to oversize the inverter. Original source: https://blog.fluenceenergy.com/energy-storage-ac-dc-coupled-solar 34 Silicon Chip Australia's electronics magazine siliconchip.com.au to this reference, the output voltage and current can be controlled with precision. This control is referred to as ‘grid following’, as the inverter is tracking the grid waveform and operating as a current source. The other type of inverter control is called ‘grid forming’, meaning the inverter operates as a voltage source and largely ignores the existing grid waveform. In normal operation, the inverter controls its output power to optimise the operating point of the solar array. This is known as maximum power point tracking (MPPT), which involves holding the array DC voltage at the optimum current for the solar panel to generate its maximum power (see Fig.16). This position is constantly changing with variations in irradiance and temperature, so the MPP tracker works through trial-and-error to dither the DC voltage up or down to search for increased power. Fig.13: a typical IGBT die structure. Original source: https://w.wiki/Bqfd Fig.14: an equivalent circuit of the Insulated Gate Bipolar Transistor (IGBT). It has a BJT and Mosfet connected together on a single silicon die. Original source: https://techweb.rohm.com/ product/power-device/igbt/11640/ Battery inverters Similarly to their solar counterparts, battery inverters take a DC voltage from the cells and convert it to an AC voltage for the grid. In fact, many solar inverter OEMs service the battery market with identical hardware. The difference is in the control software; the MPP tracker is replaced by algorithms to gracefully charge or discharge the cells with minimal degradation. Battery health is mainly a function of temperature and state of charge (SOC), so current limits are reduced at extremes of temperature, or when the cells are fully charged or discharged. Fig.15: an inverter circuit showing output ‘T’ filter (an LCL network) and the additional inductance from the grid-tied step-up transformer. The six IGBTs synthesise a three-phase AC waveform using PWM. Original source: https:// imperix.com/doc/implementation/active-damping-of-lcl-filters Battery-solar hybrids Increasingly, batteries are being built alongside solar photovoltaic systems. They are a good combination, as the battery not only avoids paying for grid electricity but also network fees. Most solar-battery hybrids currently in operation on the grid are ‘AC coupled’, meaning that they are joined on the output side of their respective inverters (see Fig.11). A new technology gaining popularity is the ‘DC coupled’ hybrid. Rather than the batteries connecting directly to a dedicated inverter, they instead interface to the solar array through a DC-DC converter. The inverter then converts both battery and solar power to AC for the grid (see Fig.12). siliconchip.com.au Fig.16: the output characteristics of a solar panel for different values of irradiance. A connected inverter constantly searches for the optimum operating point in a process known as maximum power point tracking (MPPT). Source: www.researchgate.net/figure/fig3_324179520 Australia's electronics magazine April 2025  35 Fig.17: a Doubly Fed Induction Machine (or Generator) used to generate power from a wind turbine. The stator is directly connected to the grid, while the rotor is fed from a back-to-back inverter. The DFIM therefore decouples the turbine rotational speed from the grid frequency, allowing the control system freedom to optimise for maximum power. Original source: www.mdpi.com/energies/energies-15-03327/article_deploy/html/images/energies-15-03327-g001.png The main benefit of this topology is removing the inverter as a bottleneck to power flows, as most solar systems match an oversized array to their inverter. This is known as the DC/ AC ratio; it is usually around 1.3:1, to balance the cost of the inverter against increased revenues from higher power handling. On residential systems, this leads to the ubiquitous 6.6kW array matched to a 5kW inverter. The drawback to such a ratio is that when there is sufficient irradiance, potential power generation is wasted as the inverter is already at its limit. With an AC-coupled hybrid, this bottleneck also limits the battery charging; any power from the solar array has to pass through the grid-­connected solar inverter before it comes back through the battery inverter and into the pack. On a DC-coupled system, this limit is alleviated. Assuming sufficient irradiance, the inverter can be exporting at full power, and energy that would otherwise be lost from the solar array is used to directly charge the batteries, giving a superior yield for a given solar array. DC-coupling can also help a system remain below a given size for regulatory reasons. Grid-scale systems with less than 5MW of inverters have a simpler grid connection process, and residential systems are capped at 5kW of export. 36 Silicon Chip It is the residential sector in particular that will see an increased uptake of DC-coupled ‘hybrid inverters’ over the coming years. Wind turbine inverters Early turbine designs simply connect an alternator directly to the grid, but this limits the rotor to a fixed operating speed (the grid frequency), which is not necessarily the optimum speed for maximum power. A more modern design for small wind turbines uses a rectifier to convert the alternator’s AC output to DC, then an ordinary solar inverter to convert it back to AC to feed the grid. This way, the inverter has freedom to use its MPP tracker to find the best operating point, which improves yield despite the additional losses from the conversion process. Grid-scale wind turbines use a different inverter-based technology known as the doubly fed induction machine (DFIM). The stator is directly coupled to the grid, while the rotor is energised by a back-to-back inverter (see Fig.17). Thus, the rotor can be fed with an arbitrary waveform in much the same way as a solar inverter. By varying the voltage and phase, the power coming out of the stator is tightly controlled. Most critically of all, the rotor can be excited with a Australia's electronics magazine fixed frequency to match the grid. The stator output will always produce this same frequency, despite constant variation in the turbine speed due to wind fluctuations. This allows the control system to optimise the rotational speed of the turbine for maximum power, in a similar way to MPPT for solar panels. Grid stability – voltage control Network operators must keep tight control over grid voltage to prevent damage to connected assets and the network infrastructure itself. This voltage is only permitted to vary in a very narrow range. In Australia, that’s 230V AC +10%, -6% for a single-phase supply (ie, 216V to 253V AC). There are two main tools that can be used to maintain these limits: transformer tap changers and reactive power control. Transformer tap changers simply select between a series of closely spaced taps on the substation stepdown transformer. These taps subtly change the transformer ratio that is linking the high voltage transmission network with the low voltage distribution network, therefore providing control of the output voltage (see Fig.18). The other method of voltage control is using reactive power. Electric motors are by far the most common siliconchip.com.au Fig.18: a transformer tap changer can regulate the grid voltage by altering the ratio between the transmission and distribution networks. Original source: www.researchgate.net/figure/ fig1_224188399 load on the grid, making up more than 90% of total electricity demand in some regions. As they are strongly inductive, the grid operates with a lagging power factor; current lags voltage. In an inductive grid, the voltage is lower for a given power consumption than if that load was purely resistive. Capacitance can be used to compensate, which is commonly referred to as power factor correction (PFC). At its simplest, PFC involves switching banks of capacitors in and out of circuit – see Fig 19. Adding capacitance will cause the grid voltage to increase, and removing it will cause voltage to decrease. This Fig.19: a traditional capacitor bank used for power factor correction (PFC). Contactors K1 through K3 etc can be controlled to switch in a variable amount of capacitance, contributing reactive power to the grid and thus controlling voltage. Original source: https://electrical-engineering-portal.com/buildingcapacitor-bank-reactive-power-compensation-panel simple method is very commonly used by grid operators for voltage control, but there are some newer technologies that offer superior performance. The Static VAR Compensator (SVC) works in a similar way to the capacitor banks mentioned above, but rather than using mechanical relays, a power semiconductor such as a thyristor is employed, as shown in Fig.20. The thyristor can switch the capacitors in and out of circuit faster than mechanical relays and won’t wear out. This technology is widely used for reactive power control at gridscale generators and substations. The Static Synchronous Compensator (STATCOM) offers further performance improvements. Rather than a thyristor, the STATCOM arranges IGBTs in a H-bridge topology, with capacitance across a DC bus, as per Fig.21. The H-bridge can synthesise an AC waveform with a fully controllable phase shift, providing very tight and fast control of reactive power. STATCOMs are popular at substations where advanced voltage control is required, such as rural areas where a feeder may have to cover a long distance, or where SWER (Single Wire Earth Return) lines are in use. The STATCOM shares many Fig.20 (above): a simplified schematic of a Static Var Compensator. Similar to the PFC unit from Fig.19, banks of capacitors are switched into circuit as needed. Original source: www. researchgate.net/figure/fig1_308944567 Fig.21 (right): a simplified schematic of a Static Synchronous Compensator or STATCOM. The H-bridge produces an arbitrary waveform, which in most cases is generated with the current leading the voltage (ie, capacitance). Note the similarity to the inverter in Fig.13. Original source: https://doi.org/10.1007/s42452-020-03315-8 siliconchip.com.au Australia's electronics magazine April 2025  37 Fig.22: An example of spinning reserve in South Australia. During this period (October 18th to 21st), over 100% of grid demand is being met by rooftop solar. Most other generators are not needed and have switched off, except for a small amount of wind and utility solar, and notably some gas. It is uneconomic to run a gas generator for energy during this time; its benefit is providing grid stability through the angular momentum of its turbine and alternator. Source: https://explore.openelectricity.org.au similarities with the inverters discussed in the earlier sections; the main difference is that the DC bus only has capacitance connected in the STATCOM, rather than solar panels or a battery. Inverters are therefore a great way to control reactive power, and widely used at the utility scale for voltage control. Fig.23 shows a real-world example of a solar farm that operates with a power factor of 0.85. As it increases its output power, the grid voltage decreases through the action of the reactive power it contributes. In this way, IBRs (inverter-based resources) 38 Silicon Chip will play an important role in regulating grid voltage in coming years. A segment with good potential is rooftop solar, which currently provides almost no reactive power from its 20GW of installed capacity. A simple settings change could enable up to 15GVAr of support for free, which is plenty to tightly control voltage across the whole eastern seaboard and also ease network constraints. Grid stability – frequency (inertia) Our existing grid relies heavily on the angular momentum of rotating Australia's electronics magazine machines for frequency stability. This ‘spinning reserve’ works by resisting brief frequency excursions that might destabilise a power system. In most Australian states, this inertia is provided by the large alternators of coalfired power stations, and to a lesser extent gas and hydro. As these machines are electromechanically coupled directly to the grid, they provide momentum that works to maintain a frequency of 50Hz. Any increase in frequency (grid oversupply) will effectively turn the alternator into a motor. It will begin to speed up as it consumes power from the grid, resisting further instability. A sudden reduction in frequency (undersupply) works similarly; the alternator dumps extra power into the grid as it decelerates. Alternators work well in this role as they can produce or consume many times their rated power for short periods, although their response is governed by the electromechanical properties of the system and is therefore uncontrolled. The AEMO carefully tracks ‘spinning reserve’ to make sure the power system has adequate strength to resist any sudden shocks to the system, such siliconchip.com.au as a large generator or load tripping off-line. This is particularly evident in South Australia – see Fig 22. In this example, 100% of grid demand is being met by rooftop solar. All other generators are not needed so have turned off, aside from a small amount of utility wind and solar and a minimal amount of gas. The gas generators will not be making money on their energy production during this time, but they will be receiving payment for providing grid stability. The Torrens Island steam gas generator commonly provides this service, given its central location in Adelaide. It operates for long periods at 40MW, a fraction of its full nameplate capacity. Trials are underway to investigate the feasibility of repurposed coal generators for spinning reserve. It is possible to refurbish an old coal unit as a ‘synchronous condenser’, although early indications are that it will be more expensive than other solutions. Synchronous condensers are effectively large spinning masses with grid-coupled alternators. In normal operation, they draw a small amount of power from the grid to maintain their speed. Should a frequency excursion occur, they absorb or inject power to the grid in the same way as other spinning reserve. Their configuration is essentially identical to the one shown for wind turbines in Fig.17 except that, instead of the motor/alternator being connected to a turbine, it is connected to a rotating mass. These machines are increasingly popular for strengthening weak networks and can also be used for voltage regulation through reactive power control. Inverters can also be used to create so-called ‘synthetic inertia’. The IBR can be configured to monitor grid frequency and rapidly absorb or inject power should a frequency excursion occur. ‘Grid-forming’ batteries are wellsuited to this task given their fast response time, precise output control and ability to work bidirectionally. Successful trials have also been completed using wind turbines for frequency regulation – see Figs.24 & 25. It is estimated that a ratio of 15% ‘grid forming’ to 85% ‘grid following’ inverters is optimal to replace spinning reserve. siliconchip.com.au Fig.23: an output plot of a real-world solar farm used for voltage control. During a period of oversupply, the generator ramps down its output power (red). The grid voltage (pink, purple, green) increases. Some time later, the solar farm ramps up to full power, lowering the grid voltage through the action of its reactive power control. Fig.24: the output (orange) of the Hornsdale wind farm following a setpoint (AGC, black) to regulate grid frequency (grey). Source: Hornsdale FCAS Trial, p24 Fig.25: the output of a traditional synchronous generator across the same period as Fig.24. It underperforms compared to the wind farm given its slower response. Source: Hornsdale FCAS Trial, p25 Australia's electronics magazine April 2025  39 Grid stability – redundancy The grid works on the N-1 principle. That is, there must always be sufficient standby capacity that a trip on any single generator or transmission line will not lead to a blackout. This sometimes dictates some strange grid operations, such as curtailing generators or running transmission lines from areas of low supply towards areas of high supply. As the generation mix changes, these constraints will also change. Wind and solar generators are more decentralised than our existing coal fleet, and typically smaller. This gives a lower concentration risk for any single generator failure but increases operational complexity. Advanced software called distributed energy resource management systems (DERMS) is beginning to be rolled out in many networks. It provides improved visibility and control over grid constraints. These modern control systems are central to the energy transition, managing distributed assets and retaining N-1 redundancy. Grid stability – negative demand The combination of rooftop solar and coal is leading to an interesting problem for network operators. In the middle of the day, it is common for the rooftop system to be supplying its local load and also exporting to the grid. This is leading to periods of low demand, and in future even negative demand (see Fig.26). During these periods, fast ­responding grid-scale assets turn off for economic reasons, but coal generators remain active due to operational constraints, and rooftop solar remains active as it is usually paid a fixed rate ‘feed-in tariff’. It is common at these times for the distribution transformers to be running in reverse, supplying power back onto the high voltage transmission network. This is problematic, as many distribution transformers need to derate their power capacity for reverse flows. This is not a limitation of the transformer itself, but rather the tap changers, which usually employ “asymmetrical switching” to reduce the amount of power it must withstand during the middle of a change. This is advantageous when power is in the normal direction, but for reverse flows, the asymmetrical switching exposes the tap changer to increased power, severely limiting the reverse power capability of the transformer. Many networks are currently investigating and implementing upgrades to better handle this condition. A common solution is simply to inhibit the tap changer when reverse power exceeds its rating. The network operator won’t be able to use the tap changer during this period for voltage control, but they can use reactive power as discussed earlier. Another solution is to incentivise more load into the grid during daylight hours. So-called ‘solar soaker’ tariffs are being trialled, which offer free usage of the network between 10am and 3pm, but a higher rate between 5pm and 8pm. Increased electric vehicle proliferation should also help negative demand, as car chargers have higher usage during daylight hours. Conclusion Modern power electronics are playing a central role in the energy transition. Active stability techniques like Demand Response, IBRs and grid-forming batteries/inverters will replace most of the spinning reserve over the coming decades. Periods of negative demand may lead to lowcost or even free EV charging during sunny days, to make use of ample solar power, and incentivise further investSC ment in battery storage. Fig.26: minimum demand projections for the eastern seaboard grid. South Australia will likely experience negative overall demand this spring, with other states to follow in the coming years. Minimum operational demand is sometimes called ‘base load’. Source: AEMO ESOO 2024, p41 40 Silicon Chip Australia's electronics magazine siliconchip.com.au FROM PROJECTS TO PICNICS, EASTER STARTS HERE! MARK YOUR CALENDAR: Wed 2nd to Mon 21st of April, 2025 ALL PRICES SHOWN IN AUD NOW ONLY $ 109 WILL START ALMOST ANYTHING! 21 TORCH & SOS BEACON DIGITAL CONTROL PANEL 10W WIRELESS QI CHARGER SAVE 55% 12,000MAH POWER BANK NOW ONLY $ 19 WHILE STOCKS LAST 95 . SAVE 30% NEVER LOSE YOUR VALUABLES AGAIN! Smart Tag for Apple ® Devices LA4222 *Smartphone not included $ 2 X USB PORTS 199 15L Portable Fridge/Freezer SAVE 100 $ 850A 12V Compact Jump Starter, Power Bank & Wireless Qi Charger GH1623 SEE NEXT PAGE FOR MORE MB3764 YOUR PERSONAL WEATHER EXPERT WHILE STOCKS LAST $ 89 SAVE $30 XC0434 1080P HD CAMERA WITH FIRST PERSON VIEW NOW FROM $ MAKE THE HOLIDAYS FLY BY! R/C Foldable Mini Drone with 1080p Camera 97 SAVE<at>30% GT4900 *Smartphone not included. FEATURES SHORT CIRCUIT, OVER CURRENT & REVERSE POLARITY PROTECTION Wireless Weather Stations XC0412 / XC0432 / XC0434 XC 0412 FAST PRINT SPEED UP TO 600MM/S PRINT SIZE UP TO 250 X 250 X 260MM 4 COLOUR FILAMENT BOX WITH DRYING CAPABILITY $ ONLY 399 100Ah 12.8V Lithium Cycle Battery GREAT PRICE Deep SB2260 ^ALL PRICES SHOWN IN AUD - VISIT JAYCAR.CO.NZ FOR NZ PRICES $ ONLY 749 MULTI-COLOUR PRINTING. ENDLESS CREATIVITY GREAT PRICE TL5200 Kobra 3 Combo Filament 3D Printer 1 SAVE UP TO 30% ON SELECTED BRASS MONKEY FRIDGE/FREEZERS DIGITAL CONTROL PANEL WHILE STOCKS LAST WHILE STOCKS LAST SAVE 60 SAVE $110 $ 139 $ $ 9 28 9L Ultra-Compact 22L Compact GH1627 Limited stock, check stores for stock GH2020 419 NOW FROM $ 52 SAVE<at> 200 ICE TRAY COMPARTMENT GH1624 DUAL ZONE NOW FROM $ 239 $ 299 SAVE $150 FITS OPTIONAL LITHIUM BATTERY Battery sold separately LARGE WHEELS FOR EASY TRANSPORT 35L & 45L Dual Zone with Battery Compartment REMOVABLE DIVIDER: FRIDGE, FREEZER, OR BOTH, YOU DECIDE! GH2020 / GH2022 DUAL ZONE DIGITAL CONTROLS DUAL ZONE SEPARATE LIDS WITH DUAL HINGES 106 CONVERTIBLE SINGLE / DUAL ZONE 52 FITS OPTIONAL LITHIUM BATTERY Battery sold separately $ GH1640 NOW ONLY 589 SAVE $260 75L Dual Zone with Battery Compartment GH2080 NOW FROM $ 75 SAVE 30% 2 5.2Ah & 7.8Ah 12.8V Lithium Batteries for Fridges GH2049 / GH2051 5.2 AH 7.8 AH WHEELS & HANDLES KEEP THINGS COOL THIS EASTER! 36L, 50L & 60L Convertible Dual Zone GH1640 / GH1642 / GH1644 GH2049 GH2051 ^ALL PRICES SHOWN IN AUD - VISIT JAYCAR.CO.NZ FOR NZ PRICES SERIOUS ABOUT 3D PRINTING? SO ARE WE! $ ONLY 299 $ 10K PRECISION WITHOUT THE PREMIUM PRICE ONLY 699 14K HIGH-PRECISION OPTICAL SYSTEM FOR ULTRA-FINE DETAILS PRINT SPEED UP TO 70MM/H PRINT SPEED UP TO 90MM/H PRINT SIZE UP TO 153.4 X 87 X 165MM PRINT SIZE UP TO 223 X 126 X 230MM AUTOMATIC PRECISE LEVELLING Photon Mono 4 Resin 3D Printer Photon Mono M7 Resin 3D Printer TL5254 TL5250 SEE OUR EXTENSIVE RANGE OF 3D CONSUMABLES IN-STORE OR ONLINE 14 COLOURS AVAILABLE 4 COLOURS AVAILABLE 11 COLOURS AVAILABLE 10 COLOURS AVAILABLE CLEAR HIGHER RESOLUTION & PRECISION SUPPORTS PRINT SPEEDS UP TO 600MM/S CLEAR 100% BIO-DEGRADABLE HIGH STRENGTH & TOUGHNESS LOW ODOUR HIGH PRECISION FDA FOOD SAFETY APPROVED $ ONLY 44 95 . $ EA TL4454 - TL4781 $ 4495 . 1kg 1.75mm PLA+ Filaments FROM ONLY ONLY EA 1kg 1.75mm Hyper PLA Filaments TL4790 - TL4799 29 95 $ . ONLY 5995 . 500g & 1kg PLA Resins TL4433 - TL4540 EA 1kg PLA Pro Resins TL4440 - TL4770 WIRELESS SCANNING WITH 0.1MM ACCURACY $ NOW ONLY 69 95 . SAVE 10% 18 PCS 3D Printer Tool Box Kit TL4594 NOW ONLY $ 239 SAVE $30 MAXIMISE PERFORMANCE & CONTROL UP TO 4 PRINTERS SIMULTANEOUSLY Sonic Pad TL4800 Limited stock, check stores for stock $ ONLY 699 CR-Scan Ferret Pro 3D Scanner TL4418 *Smartphone not included TERMS & CONDITIONS: Prices valid from 2/4/25 to 21/4/25. Stock may be limited on sale items. No rain checks. Savings on Original RRP (ORRP). Prices are correct at time of publication but are subject to change. Jaycar reserves the right to change prices if and when required. All prices shown in AUD. For full Price Beat Promise T&Cs see website. For full gift card T&Cs see website. For 1 Hour Click & Collect T&Cs see website. SUPPLY CHAIN DISRUPTION: We apologise for factors out of our control which may result in some items not being available on the advertised on-sale date of the catalogue. For full T&Cs see website. + For details and terms on payment options see website. 73 STAY PROTECTED, SPEND LESS CATCH NATURAL WILDLIFE IN ACTION 4K NO-GLOW INFRARED LEDS FOR COVERT NIGHT VISION UP TO 20M NOW ONLY 299 $ SAVE $ 100 ACCESS VIA SMARTPHONE OVER 4G INTRO OFFER $ 1080p Outdoor Trail Camera with 4G 119 NOW ONLY $ * Smartphone not included SAVE $40 NEW SUN POWERED PEACE OF MIND 4K Outdoor Trail Camera SAVE $ 20 QC8067 139 FEATURES REMOTE SMARTPHONE VIEWING 1080p Wi-Fi Battery Powered Camera & Solar Panel QC8063 BUILT-IN LEDS FOR NIGHT VISION UP TO 30M QC3914 2 YEAR WARRANTY 4K 1TB STORAGE 2TB STORAGE NOW FROM $ 799 QV5530 8 Channel 4K NVR Kits SAVE<at>$300 3 Cameras QV5530 6 Cameras QV5532 $ ALL NVR & DVR KITS FEATURE REMOTE SMARTPHONE VIEWING * ONLY 199 4 Channel DVR Kit with 2 x 1080p Cameras QV5040 Smartphone not included WIRELESS FOR EASY INSTALLATION 1TB 4K STORAGE NOW ONLY $ 799 1TB STORAGE SAVE $200 NOW FROM $ 8 Channel Wireless NVR Kit with 4 x 4K Cameras 389 SAVE<at>$150 QV5622 QV5060 DVR Kits with 4 x Cameras QV5060 / QV5064 / QV5066 HOME OR AWAY, WE’VE GOT YOUR SECURITY COVERED NOW ONLY 9 $ 95 . SAVE 30% 130dB Personal Alarm with Torch LA5111 4 NOW ONLY $ 14 $ . SAVE 20% Surveillance Warning Sign LA5115 NOW ONLY 95 3695 . SAVE 15% NOW ONLY $ 24 95 . SAVE 15% Dummy Camera LA5325 Wireless Driveway and Entry PIR Alert Kit LA5178 ^ALL PRICES SHOWN IN AUD - VISIT JAYCAR.CO.NZ FOR NZ PRICES Mini Projects #025 – by Tim Blythman SILICON CHIP Weather Monitor Our WiFi Weather Logger Mini Project from last December has been handy for keeping track of temperature and humidity. However, it isn’t always convenient to use a computer to check on its status. This Mini Project that makes it easy to monitor the Logger’s temperature and humidity without needing to open up a web browser. T he WiFi Weather Logger Mini Project (December 2024; siliconchip. au/Article/17315) is a compact unit consisting of a WiFi Mini processor module, a DHT11 temperature and humidity sensor and board that interfaces to a microSD card. It monitors the temperature and humidity by reading the sensor and makes that data available as a web page via WiFi. It also records the data to the card, and that data can be downloaded as a simple yet flexible CSV (comma separated variable) file. CSV files can be opened by many spreadsheet programs, so it is easy to analyse the data and create charts and the like. We often found ourselves using the interface to simply check the temperature and humidity. That means opening up the web page on a computer or smartphone screen and viewing the data. It is not difficult, but we realised that it would not take much hardware to create a simple, standalone monitor. The result is this Weather Monitor. It consists of an Arduino Uno R4 WiFi board attached to a liquid crystal display (LCD) shield to provide a clear and simple readout of the current temperature and humidity. Since the Weather Logger keeps track of time, the Weather Monitor can also display that time. Handily, Hardware The hardware is little more than an Uno R4 WiFi with an XC4630 Colour LCD Shield plugged into it. You’ll also need a suitable USB-C cable to power the Monitor, since the Uno R4 WiFi has a USB-C socket. Screen 2: you need to edit the WiFi network security settings (SSID and password) and set the IP address of the Logger so that the Monitor can communicate with it. You can also change the colour scheme at the same time. Screen 1: the WiFi Weather Logger serves up a web page that looks like this. By looking for key character sequences, the Monitor can extract the information to reformat it for its screen. siliconchip.com.au the Monitor also displays the status of the Logger, so you can quickly see if it is having problems, such as an SD card problem or lost WiFi connection. The Uno R4 WiFi is much more powerful than the older R3 Uno, so this simple combination of hardware could have many other handy uses, as a simple internet-connected display device. Australia's electronics magazine April 2025  45 The LCD Shield just plugs into the headers for the Uno R4 WiFi. The stack is around 20mm deep, and the two boards are about the same width. Thus, the assembled stack will rest neatly on its edge, as you can see in our photo. The orientation we have chosen allows the USB cable to attach near the Monitor’s lower edge (on the righthand side), meaning the cable does not hang awkwardly. This can be changed if you like. Software Web pages are requested and received using the hypertext transfer protocol (HTTP). The WiFi Weather Logger works as an HTTP server, offering web pages via its WiFi interface. Thus, the Weather Monitor needs to use an HTTP client to access those web pages. Interestingly, it’s a lot easier to write an HTTP server program than an HTTP client with the Arduino IDE. That’s because there are Arduino libraries for creating HTTP servers, included with most WiFi-capable board profiles. So implementing an HTTP server is easy to do with the Arduino IDE. Downloadable HTTP client libraries exist, but a client must handle all the possible server options, while a server gets to choose which options it offers. Also, a client may have to deal with a large amount of data if the server delivers a large file. Because of this complexity, our Monitor software has been specifically written to handle the server protocols used by the Logger. It should work fine if you want to customise the software to work with other Arduino HTTP servers, but it may not work with other server types. The software connects to a pre-­ programmed WiFi network and attempts to download the web page from the Logger using its IP address. This means you should ideally set your router to allocate the Logger a fixed IP address. This might be called something like address reservation in your router settings. Otherwise, you might find your router allocates a different IP address to the Logger at some point and the Monitor can’t find it. This could happen if, say, the power goes off, or the router is reset. The Logger and Monitor must be on the same network, such as the same home WiFi network. There are ways to make the Logger accessible from the wider internet, but these are well beyond the scope of this article as they require a good knowledge of network security principles. When the Monitor downloads the web page, it scans the contents for key sequences to locate necessary information. For example, once it finds the text “Temp: ”, it knows that the temperature reported by the Logger will follow. Screen 1 shows a typical page served up by the Logger. The web page is a HTML (hypertext markup language) document that Parts List – Weather Monitor (JMP025) 1 assembled WiFi Weather Logger (see our December 2024 issue) 1 Arduino R4 Uno WiFi main board [Jaycar XC9211] 1 2.5in colour LCD shield [Jaycar XC4630] 1 USB-C cable to suit the Arduino Uno R4 WiFi [Jaycar WC7900] Screen 3: the Serial Monitor output should look like this for a normal startup. We found that our unit took up to 20 seconds to connect to a WiFi network (when it waits to receive the IP address), plus another five seconds to access the web page. contains so-called tags that are not visible in a web browser, although they change how the web page is rendered on the screen. For example, the title is extracted by looking for the “<title>” tag. Apart from the raw information on the page, the Monitor also records a timestamp (using its internal milliseconds counter) of when the web page was retrieved. From that information, it can show an updating time display without having to continually load data from the Logger. The Monitor tries to load a page from the Logger every minute, and updates its displayed information when it does so; the time and date display is updated continuously. If there have been no successful updates for an hour, the screen reports this instead of repeating stale data. Assembly for this project is literally plug and play! The shield simply plugs into the main board and then it is ready for programming. 46 Silicon Chip Australia's electronics magazine siliconchip.com.au Screen 4: while starting up, you will see a screen like this on the LCD. It also might appear if the Logger is offline and the Monitor can’t communicate with it. While the display shield has a touch sensor, we do not use it. It’s easy enough to display all the available information, so there are no selections that need to be made. The WiFi interface is also blocking, meaning that whenever it is busy, we wouldn’t be able to check for a touch input anyway. Uno R4 board profile Assuming you have completed construction by plugging the LCD shield into the Uno R4 WiFi board, you can start the programming process. You will need the Arduino IDE with the Uno R4 Boards profile installed. The sketch itself includes some library files, but there are no external libraries needed. The Arduino IDE can be downloaded from www.arduino.cc/ en/software To install the Uno R4 Board profile, open the Boards Manager, search for “R4” and install the “Arduino Uno R4 Boards by Arduino” option. We used version 1.3.2; other versions should work fine, but if you have problems with the sketch compiling, try changing to this version. Programming There are some customisations that you will need to apply to the sketch before uploading it. Open the WEATHER_LOGGER_MONITOR sketch and switch to the WEATHER_ LOGGER_MONITOR.ino tab, as shown in Screen 2. Lines 2, 3 and 4 contain the WiFi network settings and IP address of the Weather Logger (which you should have built previously). Change these siliconchip.com.au Screen 5: if you see this screen then all is well. The remote temperature, humidity and time are reported; the time is kept current using an internal timer on the Uno R4 WiFi. to suit your network and the Weather Logger that is connected to it. You can set the display’s colour scheme on lines 14 and 15. FGC is the foreground (text) colour, while BGC is the background. BLACK, WHITE, RED, BLUE, GREEN, CYAN, MAGENTA, YELLOW and GREY are available #defines that are set in the XC4630d.c file; you can also use 16-bit RRRRRGGGGGGBBBBB format colour values. Lines 8-11 of the XC4630d.c file also set the version of the XC4630 LCD shield; the most recent V4 is enabled (by not being commented out with “//”). If you find this version does not work, you can try the other versions. You can flip the orientation of the display by changing the command XC4630_rotate(4) to XC4630_rotate(2) in the setup() function. After these changes, the sketch is ready to upload. Make sure to choose the Uno R4 WiFi board and its correct serial port (eg, COM port on Windows). Open the serial monitor to 115,200 baud and be ready to view the startup messages as seen in Screen 3. If the Monitor does not connect to the WiFi network within 30 seconds, it will reset itself to try again. We found that even with the correct settings, the Monitor sometimes failed to access the Logger’s web page. You can either wait 60 seconds for it to try again or use the reset button on the Uno R4 WiFi. The LCD should show information similar to that in Screens 4 & 5, depending on its phase of operation. The software can be downloaded from siliconchip.au/Shop/6/1836 Australia's electronics magazine You can see a video of the Weather Monitor booting up and then displaying the weather & time at siliconchip. au/Videos/Weather+Monitor Summary If you see something like Screen 5, then everything is working as expected. The status line at the bottom of the page will report when the Monitor is trying to download fresh data, while the clock may occasionally stall for a few seconds due to the blocking nature of the WiFi interface. We considered adding charts to the Monitor by having it download the CSV files from the Logger, but the amount of RAM on the Uno R4 WiFi is not enough to allow this. If you are a dab hand with Arduino, you might like to try customising the Logger or Monitor sketches to report different data. Changing the title of the Logger’s web page, for example, should change the title displayed on SC the Monitor. This Weather Monitor uses a wireless connection to our previous WiFi Weather Logger to collect data. The Logger is built using just an ESP8266, DHT11 and microSD card shields. April 2025  47 Antenna Analysis and Optimisation Over the last two articles, we have explained how antenna matching and VSWR work and given instruction on using the free “Smith” software to design antenna matching networks. This final part in the series explains how to determine the bandwidth of an antenna matching network. Part 3 by Roderick Wall, VK3YC A fter reading the article last month, you should know how to use the free “Smith” software to design an antenna matching network. This will allow you to bring most antennas to resonance and achieve a VSWR close to the ideal of 1:1 at a specific frequency. Of course, radio transmitters and receivers often need to operate over a range of frequencies. You need to be able to design the matching network with enough bandwidth to pass signals over the range of interest. We will now look at using another free software package to achieve that. Smith charts also have ‘constant-Q curves’ that can be used to control the bandwidth of a matching network. For this, we will use the Iowa Hills Smith Chart software; like the Smith program we used last month, it is also a free download. This software used to be available from https://iowahills.com but that website is unfortunately gone. Luckily, someone kept a copy of it, so you can download a zip of the whole website from siliconchip.au/link/ac0y Within that zip, navigate to the subdirectory “cb.wunderkis.de\wk-pub\ www.iowahills.com\Downloads” and you will find a file named “Iowa Hills Smith Chart.zip”. Extract that, then unzip it, and you will be able to run the executable. We tested it on Windows Screen 9: this C/L/C/L matching network comes close to the black Q = 1 curve. 48 Silicon Chip 10 and 11, and it worked fine on both versions. Screens 9-12 show two different matching networks that match a (118 – j99.5)W load to a 50W source. One using a Q = 1 curve (Screens 9 & 10) and the other using a Q = 6 curve (Screens 11 & 12). The matching networks were kept within the constant-Q curves as shown. The Return Loss graph (Screen 10) shows that the bandwidth for the Q = 1 network in Screen 9 is 4MHz wide for a VSWR of 1.22:1. On the other hand, Screen 12 shows that the bandwidth for the Q = 6 network (Screen 11) is 1MHz. To achieve the wider bandwidth for the Q = 1 network, four Screen 10: the frequency response plot of the matching network/antenna combination shown in Screen 9. Australia's electronics magazine siliconchip.com.au components were used, while the Q = 6 network used just three components. The antenna Q sets the lowest possible Q and the widest possible bandwidth. If the antenna Q is high, the bandwidth will be narrow regardless of the matching network. The Q = 6 matching network demonstrates that the bandwidth can be controlled to make it narrow if required. These plots were modelled using ideal capacitors and inductors, although the Iowa Hills Smith Chart software can also model non-ideal components. The return loss graph in Screen 12 indicates the bandwidth is around 1.25MHz for a VSWR of 1.22:1. Using the Iowa Hills software This software works similarly to Fritz’s Smith chart software but it has its own quirks and procedures. Let’s go through the steps required to reproduce Screens 9 through 12. Launch SmithChart.exe and make the window larger if you’d like to. Then, in the upper-left corner, change the frequency (F0) to 28.4 (MHz) and the Span to 5 (MHz). Go to the “Set Load” menu and choose the “Load, Source, and Parasitics” menu option. In the dialog box that pops up, change the Load Impedance to 118 real and -99.5 imaginary numbers giving (118 – j99.5)W, click Apply, then click Close. You will then see the red (antenna) and blue (source) points in the correct positions on the Smith chart. Next, we add the Q = 1 curve by selecting the “Q and VSWR Circles” option, also under the “Set Load” menu. In the right-hand column under Q, change the first 0 to 1 (for Q = 1), click Apply and then click Close. Now we can build our matching network. We insert components starting at the Load end, so open the “Shunt” menu and click “Inductor”. Click on the upper black Q = 1 curve, and you will see that the inductor inserted in the lower-left corner of the screen has a value of 620nH. Next, click “Series” and then “Capacitor”, then click on the horizontal blue line running across the middle of the chart to add a 56pF capacitor. Repeat those two steps to add a 560nH shunt inductor and a 110pF series capacitor to reach the blue dot in the middle of the chart. You should then have a display that matches Screen 9. Next, click on the Return Loss radio button at centre left of the screen and you will be greeted by a plot that matches Screen 10. The steps to reproduce Screen 11 & Screen 12 are similar to the above except that you add the Q = 6 curve Screen 11: this C/C/L matching network touches the black Q = 6 curve so it has a smaller bandwidth than Screen 9. siliconchip.com.au instead of Q = 1 and then you add a shunt inductor, followed by a series capacitor and finally, a shunt capacitor. Refer to Screen 11 to see where to click to get the same values of 200nH, 270pF & 240pF, respectively. After returning to the Smith chart, you can right-click on it four times to remove the components you added, change the Q curve via the “Q and VSWR Circles” menu option and then proceed to add the new matching components. Screen 13 shows the result of clicking the “Sweep SC” radio button on the left after setting the span to 1.25 (MHz) for the example in Screens 11 & 12. This Sweep value is the same as the bandwidth, and the black line on the chart confirms that the 1.25MHz sweep fits inside the constant VSWR 1.22 circle. In this example, hitting the Q = 6 curve while matching keeps the bandwidth narrow at 1.25MHz. You can use this approach to limit the bandwidth in your matching networks to just what is required for better selectivity. Transmission lines Smith charts can be used to determine what the impedance is at each end of a transmission line and to show how transmission lines transform impedance. Screen 12: the frequency response plot of the matching network/antenna combination shown in Screen 11. Australia's electronics magazine April 2025  49 ◀ Screen 13: by enabling the Sweep feature, we get the black line that shows how the VSWR varies over the frequency range of interest. Screen 1: click this Keyboard button in the Smith V4.1 software to type in the complex antenna impedance. The Velocity Factor (VF), also called wave propagation of velocity, is the ratio of the speed of an electromagnetic wave through a transmission line to that in a vacuum. Velocity Factor equals the reciprocal of the square root of the dielectric constant K (relative permittivity εr) of the transmission line. Use the following formulas to convert between VF and K: VF = 1 ÷ √K K = 1 ÷ VF2 For this analysis, we will return to Fritz’s Smith chart software that we introduced last month. Start a new chart and insert a point at (50 + j65.65)W at 28.4MHz using the Keyboard button shown in Screen 1 (reproduced from last month). Next, insert a transmission line by clicking the insert transmission line button shown in Screen 3. It’s the fourth from the left in that image. Leave the transmission line impedance as 50W and set the Dielectric Constant (εr) to 2.2956, which corresponds to a VF of 66%. This can be changed later if required for testing different transmission lines by clicking on the transmission line in the schematic window. The schematic window shows both the electrical length and the mechanical length for the transmission line, as you can see in Screen 14. When you click OK, you will need to set the length of the transmission line. Move the mouse to intersect with the 20m blue circle at lower left, visible in Screen 14, and click there. You will see that the transmission line length is set to 0.2495λ, effectively 1/4 of the wavelength. When both ends of a 50W transmission line are terminated with (50 + j0)W, the line will not transform the impedance and it can be virtually any length. However, in this example, the antenna is (50 + j65.65)W so the 50W transmission line will transform the antenna impedance to about (18.415 – j24.283)W. The impedance at the transmitter source end depends on the length of the transmission line. Note that the VSWR did not change from 3.4:1. VSWR is the same at both ends for any length of transmission line; the transmission line runs around the constant VSWR circle. Try different impedance transmission lines in Smith and see what happens. In this example, you can add a parallel inductor of around 214nH to bring the VSWR to 1:1. This demonstrates how a transmission line can be part of an impedance-matching circuit. Note how the matching component is at the transmitter end of the transmission line and not at the antenna end, so there will be power travelling in both directions along the transmission line as in Fig.10 (from part one). In this example, the 1/4-wavelength transmission line moves the antenna from one matching circle to another. This example demonstrates why, when you are analysing an antenna or antenna element, the coaxial cable between it and the antenna analyser should be short as possible for accurate measurements. If an antenna analyser were connected to the transmitter end of the transmission line in this case, it would give a reading of (18.356 – j24.101)W and not the antenna impedance of (50 + j65.65)W. Screen 3: this toolbar lets you insert different elements into the circuit you want to test. This image and Screen 1 have been reproduced from last month’s article. 50 Silicon Chip Australia's electronics magazine siliconchip.com.au Screen 14: using a 1/4-wavelength transmission line and a parallel inductor for antenna matching. Screen 15: using a short transmission line and a series capacitor for antenna matching. siliconchip.com.au Australia's electronics magazine April 2025  51 If a transmission line has the same impedance as the system and is 1/2-wavelength long or a multiple of 1/2-wavelength, the complex impedance will be the same at both ends of the transmission line. That’s another way of saying that a 1/2-wavelength transmission line goes in a complete circle on the Smith chart. If the transmission line is made exactly half a wavelength long, Smith will not show it because it is effectively just a point. Try different impedance transmission lines in Smith and see what happens. Screen 15 shows a transmission line being used to get onto a matching circle, with a series capacitor to complete the match. Transmission lines of different impedances and lengths are often used with antennas for matching impedances. Screen 16 shows a 50W open stub (OS) being used to complete a match. A parallel capacitor or shorted stub (SS) could have been used instead of the open stub. Fig.16 is a two-metre Zepp J-pole antenna. It uses a transmission line to match a half-wavelength radiator element. The radiator is just under half a wave parallel feedline for tuning. This concept evolved into the Zepp J-pole antenna. Exercises for the reader Fig.16: the Zepp antenna is a clever configuration (also known as a J-pole antenna) providing inherent transmission line matching. wavelength long, while the matching transmission matching line is around a quarter wavelength. The Zepp antenna was invented by Hans Beggerow for use on the German Zeppelin airships. Trailed behind the airship, it consisted of a half-wavelength long radiator with a quarter Antenna tuners sometimes use the high-pass T configuration, with series variable capacitors at each end and a parallel inductor (to ground) in the middle. By varying the capacitances, this allows them to get a decent match with a wide range of antennas. You can simulate this in Smith and experiment with component values to get various antennas to a VSWR of 1:1. Other antenna tuners use a low-pass pi configuration, which has a parallel variable capacitor at either end and a series inductor in the middle. You can experiment with that configuration too. Arguably, the low-pass configuration is better since it will filter out unwanted harmonics that may cause interference, whereas the high-pass T configuration will pass the harmonics through. As a final exercise for the reader, produce a Smith chart showing a 305W transmission line being used to match a Zepp antenna with a complex impedance of (1889 – j0.0212)W to a SC 50W system. Screen 16: a series transmission line along with an open stub transmission line can also be used for matching. 52 Silicon Chip Australia's electronics magazine siliconchip.com.au altronics.com.au POWER UP EASTER! Includes travel adapters Big deals for a BIG Easter break. 99 SL4576W 100Ah Up to 135aH capacity. Just 75mm thick! SAVE $350! 719 $ 159 $ A0319A 20W 10000mAH $ A0323 65W 15000mAH Recharge anywhere you go! Introducing our new global travel power banks. Available in either 20W or 65W power delivery with in-built mag safe compatible wireless charging they come with travel adaptors for travel all over the world. All supplied in a handy zip up case. SAVE $450 ONLY WHILE STOCKS LAST! 849 $ SL4580W 135Ah M 8194A Powerhouse LiFePO4 Slimline Lithium Batteries Ultra slim 75mm profile with full current discharge capability and a 5 year warranty. Space at a premium in your camper, caravan or 4WD? These compact batteries are perfect for remote power solutions without taking up precious cargo space for your gear. Pre-fitted with Anderson input and output connections and battery capacity gauge. 100Ah: 600 x 275 x 75mm. 135Ah: 600 x 410 x 75mm. Caravanning essential! NEW! The power to get you on the road again! 15W M 8521B Fast Charge NEW! 199 $ Lithium Jumpstarter for 12V battery vehicles. 20000mAh rated battery provides up to 1000A peak output when cranking. Fitted with a 60W USB C power delivery charger (like a giant battery bank!). SAVE 25% X 2920 4 for $ 80 59 $ SAVE $35 140 $ Q 1300 Tyre Pressure Monitor System This solar powered TPMS unit sits on your dash and provides wireless monitoring of your tyre pressures. Provides high/low pressure alarms, leak detection and temp monitoring. Signal booster option Q1302 $95. SAVE 13% Now with LiFePo4 support Handy Battery Maintainer Hassle free maintenance charging for vehicle & LiFePO4 lithium batteries. 6/12V 1.5A output. 50 $ D 2208A Wireless Charger Phone Holder Save time in the car with this handy motorised windscreen (or air vent) phone mount. 15W LED Vehicle Light Rugged aluminium casing with ultra-bright 15W light output for under wheel arches on 4WDs, exterior lighting on campers & caravans etc. IP68 rated. 12-24V DC input. Your electronics supplier since 1976. Shop in-store at one of our 11 locations around Australia: WA » PERTH » JOONDALUP » CANNINGTON » MIDLAND » MYAREE » BALCATTA VIC » SPRINGVALE » AIRPORT WEST QLD » VIRGINIA NSW » AUBURN SA » PROSPECT Sale ends April 30th 2025. Build It Yourself Electronics Centre® WORKBENCH DEALS SAVE $70 SAVE 22% $ $ 239 62 T 2496 T 2065 Desolder parts in seconds! Don’t forget the gas! T 2451 $9.50 per can. SAVE $50 SAVE $30 $ 169 $ T 2651 Tool Kit T 2650 Iron Only. 119 Solderpro 180 Portable Gas Tool High Output Blow Torch Micron® Vacuum Desoldering Station Super hot 1350°C flame with high output nozzle. Handheld or self standing design for tasks such as heatshrinking, model making, silver soldering! Designed to desolder through hole componentry, removing molten solder quickly and easily from solder pads and components. In-handle reservoir is easily removed and cleaned. Includes three desoldering tip, nozzle cleaner and filter pads. 160°-480°C. T 1528A Folding Auto Ranging Multimeter RJ45 Pass Thru Crimper Spade, Ring & Lug Crimper Anderson/Lug Crimper Switch to Pass Thru RJ45 modular crimps and save time! Crimps and cuts in one ratchet action and works with industry standard connectors. Crimps all standard “kwik” connectors such as ring terminals, bullets and spades. Easy to identify red, blue, yellow jaws. A 325mm long heavy duty lug crimper for Anderson connects and ring lugs. Suits cabling from 8AWG to 2AWG. Great for DC power systems. 99 79 $ Micron® USB Lithium Rotary Tool Set NEW! $ SAVE $20 $ 90 $ Includes carry case! Provides in depth functionality for technicians. Plus features such as dwell angle, RPM engine speed and duty cycle it can also be used for car servicing. Folding design stays put on any surface while testing, making it great for auto electricial work. T 1531A SAVE 23% 35 39 Combines a ratchet wire stripper, cutting blade & kwik crimper. Suits 10-24 AWG cable. T 1552A T 1566A SAVE $25 $ Wire Stripper & Kwik Crimper Ideal for heavy duty lugs All metal with ratchet action Superb build quality! SAVE 20% 185W of power for both blow torch and soldering work. Powered by refillable butane cartridges (2 included). Provides 500°C soldering & 1300°C blow torch. Kit includes tips, solder sucker, flux, cutters & solder. Q 1058 SCAN TO FOLLOW US! Stay up to date on latest releases, exclusive specials and news on our socials. Drills, cuts, sharpens, cleans, polishes and engraves most surfaces, this rotary tool is ideal for enthusiasts, hobby makers, or just odd jobs around the house. 5 speeds from 5000 to 25000RPM. USB C recharge with 60 mins operation. 42 accessories included. SAVE 20% 55 $ T 2125 Great for crafts and hobbies too! Like our service? Review your store on Google. Every review helps us serve you better. TECH GADGETS Easy to set up anywhere! SAVE 15% A 3615 50 $ SAVE $40 149 $ D 2325A* Mini Wi-Fi LED Projector Great for movie nights with friends and family! This compact projector offers excellent LED picture quality with 800x480p resolution for screens up to 4m (170”) wide! Very simple to set up with adjustable focus & projection distance (1-5m). HDMI input or Wi-Fi screen mirroring for playback directly from your device. SAVE $19 SAVE $30 60 69 $ $ C 9021B NEW! 129 $ M 8870 Great do-it-all travel charger 140W USB PD Charger Wireless Charging Stand Provides three USB C power delivery (3.1/PPS) ports and one USB A QC port in a compact charger. Equipped with GaN chip for fast, efficient charging. Handy bedside charger - folds up for travel use. A powerful all in one charging stand for you to keep your watch, earbuds and phone charged up and ready to go when you need it. Great value! D 2367 NEW! D 2359C Open Ear BT Earphones C 9038A Latest design in over ear wireless stereo (OWS) headphones - drivers sit on your ear rather than in your ear canal for all day comfort. Includes charging case for up to 8 hours use per charge. $ 49 MacBook Docking Hub $ Value ANC Headphones Brand name noise cancelling performance for a fraction of the price! The Futuretour headphones offer a stylish stealth black finish and excellent noise isolation (up to 20dB reduction) for work and travel. 99 NEW! 100W PD passthrough with HDMI, SD & more. Plugs into MacBook Pro/Air USB C connections and provides a full suite of peripheral connections including card reader. 100W passthrough charging. USB C Ethernet Adapter Handy Card Reader 9 in 1 USB C Multi Hub Gigabit speeds. USB A & USB C connection. Suits SD & Micro SD. D 2331 D 0433A NEW! NEW! 39.95 17.95 $ D 2035* SAVE 14% 30 SAVE 20% 60 $ $ Mini BT Speaker & Wireless Charger A MagSafe compatible charging pad for iPhone and Apple Watch, plus a Bluetooth speaker with lighting. $ USB C to Dual HDMI X 3227* Handy way to connect two 4K monitors. Wi-Fi Strip Lighting Kit This kit includes 5m of RGB strip lighting, power supply and controller unit allowing you to create colourful lighting effects around your home. Works with Alexa and Google Assistant. 60 LEDs per metre. SAVE 18% 4 Way USB C Hub 22 $ With on/off switches. D 2357B NEW! 49 $ D 2366 NO STRESS 30 DAY RETURNS! GOT A QUESTION? Not satisfied or not suitable? No worries! Return it in original condition within 30 days and get a refund. Ask us! Email us any time at: customerservice<at>altronics.com.au Conditions apply - see website. STAY POWERED UP. M 8060 300W 20% M 8050 150W 130 20% $ OFF THIS MONTH! 47 $ OFF THIS MONTH! M 8062 600W 229 $ M 8051 300W 63 $ M 8064 1000W M 8054 600W 339 86 $ $ M 8065 1500W M 8056 1000W 499 165 $ Pure Sine Wave $ Modified Sine Wave M 8066 2000W BlackMax DC to AC Power Inverters. Housed in a rugged aluminium extrusion, this range delivers robust reliability and unwavering performance - even under severe operating conditions. For peace of mind all models have been certified to Australian Standard AS/NZS 4763.2011. Ideal for tricky loads, such as laptops, TVs & game consoles. Perfect for 4WDs, campers, caravans & trade vans. 664 $ M 8067 2500W 949 $ Simple portable power for any car. Stacked with features for a great price! SAVE 22% N 2019D 60 $ 239 $ The same top notch quality and safety features as our popular Black Max inverter series (left), with a modified sine wave design to bring 240V power to any vehicle at a fantastic price. 240V outlet runs most simple appliances such as power tools, pumps, lights, fans and heater elements. SCOOP BUY! Grab a bargain. Lithium Ready Solar Charge Controller Suitable for permanent 12/24V systems with either lead acid or lithium chemistry batteries. Supports Li-NiCoMn & LiFePO4. 30A max charge current. LCD provides easy to read system status. 4 x USB charge ports. M 8057 1500W SAVE $246 99 $ N 2089 40A Powerhouse® DC-DC Battery Charger This dual input design connects to a solar panel and your cars alternator (12 or 24V) to provide charging for secondary batteries such as those used in campers, caravans and trades service vans/trailers. Suitable for Lead Acid, AGM and Lithium Fe PO4 batteries. N 0706A 15W N 0704A 10W SAVE 24% SAVE 24% $ 45 $ 60 GREAT FOR: • Motorbikes • Caravans • Boats • Jet Skis Powertran® Solar Battery Charger/Maintainers These compact solar panels are designed for keeping your vehicle batteries topped up when parked. Easy croc clip or car accessory plug connection. Can even be permanently installed outdoors. SAVE $40 Size: 360W x 185H x 138Dmm. Connect-It-All DC Power Hub T 5099 179 $ A complete central power hub for your 4WD, caravan or camping remote power set up. 6 contura style illuminated rocker switches, 3 car accessory sockets (12V), USB charger, volt meter, protected by a 6 way weatherproof fuse block. Sale Ends April 30th 2025 Shop in-store at one of our 11 locations around Australia: WA » PERTH » JOONDALUP » CANNINGTON » MIDLAND » MYAREE » BALCATTA VIC » SPRINGVALE » AIRPORT WEST QLD » VIRGINIA NSW » AUBURN SA » PROSPECT Or find a local reseller at: altronics.com.au/storelocations/dealers/ Shop online 24/7 <at> altronics.com.au © Altronics 2025. E&OE. Prices stated herein are only valid until date shown or until stocks run out. Prices include GST and exclude freight and insurance. See latest catalogue for freight rates. B 0004 Subscribe to MARCH 2025 ISSN 1030-2662 03 The VERY BEST DIY Projects ! Artificial Limbs 9 771030 266001 $13 00* NZ $13 90 INC GST INC GST Modern prosthetics serve as replacements for lost limbs Power LCR Tester Measures inductance from 50μH to 1H at up to 30A to determine the saturat ion point, capacitance from 50nF to 1F and resistan ce from 1mΩ to 300Ω Waveform Generator Handy for audio equpme nt analysis, circuit development/demos and as a pulse source The Future of the Grid What will our energy grid be powered by in the future? What benefits & downsides do the current types of energy generation have? RPi Pico 2 Audio Analyse Including a built-in signal Australia’s top electronics magazine Silicon Chip is one of the best DIY electronics magazines in the world. Each month is filled with a variety of projects that you can build yourself, along with features on a wide range of topics from in-depth electronics articles to general tech overviews. r generator, oscilloscope and spectrum display in a handheld format ...and much more in this issue! Published in Silicon Chip If you have an active subscription you receive 10% OFF orders from our Online Shop (siliconchip.com.au/Shop/)* Rest of World New Zealand Australia * does not include the cost of postage Length Print Combined Online 6 months $70 $80 $52.50 1 year $130 $150 $100 2 years $245 $280 $190 6 months $82.50 $92.50 1 year $155 $175 2 years $290 $325 6 months $100 $110 1 year $195 $215 All prices are in Australian dollars (AUD). Combined subscriptions include both the printed magazine and online access. 2 years $380 $415 Prices are valid for month of issue. Try our Online Subscription – now with PDF downloads! The Future of the Grid; March 2025 High-Bandwidth Differential Probe; February 2025 Waveform Generator; March 2025 An online issue is perfect for those who don’t want too much clutter around the house and is the same price worldwide. Issues can be viewed online, or downloaded as a PDF. To start your subscription go to siliconchip.com.au/Shop/Subscribe Rotating Light for Models Here’s a simple circuit that has various applications, such as for a model lighthouse, or as a siren on top of a model emergency vehicle. It sequences eight LEDs, using PWM brightness control, to form a pretty convincing imitation of a rotating light. T his project originated from a family member’s desire for a white revolving light atop a miniature lighthouse. Kits to build this sort of thing are available, but we hadn’t published such a circuit, and I thought it might have other uses. I also thought it could be done simply, on a tiny PCB. We could use a single logic IC if all we wanted was essentially a circular LED chaser. However, I have seen that approach used (eg, on garbage trucks); while eye-catching, it doesn’t provide a convincing illusion of rotation. Moreover, most digital logic ICs can’t deliver much current, meaning the LEDs wouldn’t be that bright without extra transistors. With a microcontroller, we can fade the LEDs in and out, creating a much more impressive effect, even with just eight LEDs at 45° intervals. We can also make it adjustable; not just the rate of rotation, but also the direction and the brightness/beam angle. We can even have multiple ‘beams’ by lighting opposite LEDs, as shown in Fig.1. Adjusting the beam angle effectively controls how many LEDs can be lit at once. It can range from just one (with varying brightness) up to almost all of them being lit at once, with just a dim spot rotating. If you build it with white LEDs, it’s suitable for a model lighthouse, and with a compact, black PCB that’s just 20mm in diameter, it will fit in most models unless they are tiny. If you want to make a siren, you could use amber, blue, red, yellow or green LEDs, or even unusual colours like cyan or pink (yes, they’re available). Fig.1: by fading in one LED at the edge of each beam and fading out the opposite one, we create the illusion of a smoothlyrotating light with just eight fixed LEDs. The ‘beam’ brightness and width varies depending on how many LEDs are lit at any given time. 58 Silicon Chip Australia's electronics magazine siliconchip.com.au Features & specifications » Simulates a rotating light with one or two ‘beams’ » Adjustable rotation speed, from about 10 rotations per second to 30 seconds per rotation (0.03Hz to 10Hz) » Reversible rotation direction » Adjustable beam brightness » Adjustable beam angle from 45° to nearly 360° » Runs from 5-12V DC, typically drawing 10-20mA » Compact circular PCB: 20mm in diameter and less than 10mm tall (with SMD LEDs) » Can use SMD (M3216/1206/SMA) or through-hole (3mm/5mm) LEDs » Use any colour LEDs Project by Nicholas Vinen Suppose you want a really special effect. In that case, you could build it with differently coloured LEDs, so the colour shifts as it rotates! It can run from a small DC supply from 5V up to about 12V, including most small battery packs, such as standard 9V batteries or 6V batteries of four AA/AAA cells. The current draw depends on the brightness, but it’s typically around 10mA. So four AA cells would power it for quite a while; possibly as long as two weeks for really high-capacity cells. Four AAAs might last 5-7 days. You can see a video of our prototype in operation at siliconchip.com. au/Videos/Rotating+Light Circuit details The complete Rotating Light circuit is shown in Fig.2. A 14-pin, 8-bit PIC16F15224 was chosen as it has just enough pins, is inexpensive and draws very little current. It can drive the LEDs directly with fairly decent brightness (its maximum per-pin current is 25mA). It’s also easy to program with the free version of Microchip’s XC8 compiler and MPLAB X IDE. The eight LEDs are connected to eight of its digital outputs via 68W current-­ l imiting resistors. Their anodes connect directly to the 5V rail, and they light when the microcontroller pulls that output pin low, to 0V. This configuration was selected as the micro’s output transistors are better at sinking current than sourcing it, as is typical. So they will deliver a higher maximum current like this. Assuming white or blue LEDs with a forward voltage of around 3.3V and a 5V supply, there will be around 1.7V (5V – 3.3V) across the combination of the 68W resistors and the micro’s output transistors. With a 3V supply, the data says that those output transistors can sink 10mA with a 0.6V saturation voltage, implying an output impedance of 60W (0.6V ÷ 10mA). It might be lower with the higher 5V supply voltage used in this circuit, but let’s use 60W as the worstcase value. That means the 1.7V is across 128W (68W + 60W), so we can expect the LEDs to be driven with a peak current of about 13mA (1.7V ÷ 128W). LEDs with a lower forward voltage, like red or amber, would receive more current, likely around 20mA. So the peak current is limited to a safe level. The microcontroller can control the average current using pulse-width modulation (PWM). One nice feature about this microcontroller is that its two PWM peripherals can be dynamically mapped to any of its I/O pins. So as the light ‘rotates’, we can determine the two edge LEDs and assign them to the PWM peripherals to dim them. The other LEDs are either fully off or full on, as determined by the states of the other digital outputs. That means that all the LEDs are controlled by hardware, with the software just needing to periodically recalculate which LEDs should be lit. It can then update the PORT and PWM registers to advance the rotating light to the next position. Two trimpots, VR1 & VR2, connect across the 5V supply with their wipers Fig.2: the circuit is little more than eight LEDs driven by the microcontroller via current-limiting resistors, two potentiometers to set the parameters and a simple linear power supply. siliconchip.com.au Australia's electronics magazine April 2025  59 Parts List – Rotating Light 1 double-sided black PCB coded 09101251, 20 × 20mm 8 high-brightness LEDs, 3mm/5mm through-hole or SMD (SMA, M3216/1206 or M2012/0805 size), colour to suit application 1 PIC16F15224-I/SL 8-bit micro programmed with 0910125A.HEX, SOIC-14 (IC1) 1 MCP1703AT-5002E/CB 5V 250mA low-dropout linear regulator, SOT-23 (REG1) 1 RB491D 20V 1A schottky diode, SOT-23 (D1) 2 1μF 16V X7R ceramic chip capacitors, M3216/1206 size 2 10kW TC33X-2-103E SMD trimpots (VR1, VR2) 1 5.1kW SMD chip resistor, M2012/0805 size 8 68W SMD chip resistors, M2012/0805 size 1 length of light duty figure-8 wire, to supply power 1 5-12V DC 100mA power source At upper right, the top side of the PCB is shown at actual size. The underside views of the SMD and through-hole versions of the Rotating Light project are shown enlarged. LED selection SMD LED kit (SC7462; $20 + postage) | TH LED kit (SC7463; $20 + postage) Both kits includes all the parts listed above, except the power supply and wire going to pins 8 & 11 of IC1. These are both analog-capable pins, so we can use the micro’s internal analog-to-­ digital converter (ADC) to measure these voltages. VR1 controls the speed & direction of ‘rotation’, while VR2 controls the beam width & brightness. Usually, you would put capacitors on these pins to keep the AC impedance low, making the ADC results more precise, but there isn’t a lot of room on the PCB, so we’ve left them off. We don’t need to make super accurate measurements, and we can compensate for the lack of capacitors either by tweaking the software or by eye when making the adjustments. In practice, we found that the ADC measurements were close enough to what you would expect based on the trimpot positions without these extra capacitors. The 5.1kW pull-up resistor on the MCLR pin (pin 4 of IC1) prevents spurious resets, while the 1µF capacitor across its supply pins provides 60 Silicon Chip If you’re going to power it from a regulated 5V supply like USB, you could omit REG1 and solder a bridge between its input and output pads. You could also bypass D1, or replace it with a 0W resistor, if you are certain that the supply polarity can’t be reversed. The maximum recommended supply voltage is 12V due to REG1’s absolute maximum rating of 16V. With a 12V supply and 50mA average current draw, REG1 will dissipate 350mW ([12V – 5V] × 0.05A), giving an expected temperature rise of nearly 120°C, which would put it close to its shutdown temperature of 150°C at an ambient temperature of just 30°C. The PCB draws enough heat away from REG1 that it’s unlikely to shut down unless the current draw exceeds 50mA. Still, if you intend to run the Light with a bright, wide beam, you’d be better off with a supply voltage below 12V; 6-9V would be ideal. If you manage to overheat REG1, it won’t be damaged; the light will just shut off and then restart when it cools down. bypassing for stability. The 5.1kW value is not critical; it could be 4.7kW, 10kW or another similar value. All that remains is the simple power supply. 5V low-dropout regulator REG1 ensures IC1 receives a steady and safe voltage, even if the incoming supply at CON1 varies. Schottky diode D1 prevents any damage from occurring if the supply polarity is accidentally reversed at CON1, while also having a modest (~0.3V) voltage drop. REG1 requires an input bypass and output filter capacitor for stability, so we have provided 1μF in both cases. That is the minimum value for unconditional stability on the output, and is more than enough for input bypassing. The circuit can be run from a 5V supply (eg, from USB), although the LED brightness will be reduced somewhat as IC1 and the LEDs will only have a supply of about 4.6V, ie, 5V minus D1’s forward voltage (~0.3V) and REG1’s dropout voltage (<100mV). Australia's electronics magazine The LEDs are arranged around the outside and can be through-hole (3mm or 5mm) or SMD types (M3216/1206 or M2012/0805). While side-emitting SMD LEDs exist, we reckon it’s easier just to use regular SMD LEDs and mount them on their sides, with the emitters facing out. That’s how we built our SMD prototype, shown in the photos. Some reasons we don’t think it’s worth getting side-emitting SMD LEDs are: 1. They are many times pricier than the normal top-emitting type. 2. They aren’t that much easier to solder than a top-emitting type facing sideways. 3. Many of them have a central pad for extra support that could short out the anode and cathode pads. 4. There are much more limited choices of size and colour compared to regular SMD LEDs. 5. Only the largest component sellers stock them. Through-hole LEDs can be soldered on either side of the board, while SMD LEDs have to go on the top. You could perhaps get away with soldering smaller SMD LEDs across the pads on the bottom if you have a particular reason to do that. siliconchip.com.au PCB design The PCB is circular with a 20mm diameter (10mm radius). By making it black, we can hide it inside models, so you only see the light when it’s on. In the middle of the top side of the PCB is the microcontroller, the two small SMD adjustment trimpots, one of the 1μF capacitors and the 5.1kW resistor. All the other components, like the LED current-limiting resistors and the remainder of the power supply, are in the middle of the underside. The power connections are two solder pads to which wires can be soldered from either side of the PCB, to suit the installation. Software The software (from siliconchip.au/ Shop/6/1837) is just under 200 lines of C code. The PIC runs at 8MHz with its internal Timer0 used to control the rotation speed of the light and Timer2 to run the PWM peripherals used for LED dimming. At power-up, it sets the pins as analog inputs and digital outputs as required. It then initialises the two timers and the ADC. The main loop waits for Timer0 to roll over, which happens every 4ms or so. Each time, it adds the rotation speed/direction to a 16-bit accumulator. It uses the accumulator value to calculate the brightness for the eight LEDs, then updates the output and PWM states. The 8-bit PWM runs at around 2kHz. When Timer0 rolls over, it also measures the voltages at the two analog inputs and applies a low-pass filter to remove noise and glitches from those readings. The new readings are used for future light update calculations. The code compiles to 1276 instruction words, taking up 2552 bytes of the 8kiB of the available flash memory (31.2%). The pro version of the XC8 compiler is not required. The critical part that generates the ‘rotating’ light is actually quite simple. If you mentally unwrap the circular light pattern into a line, you end up with a bi-directional chaser that ‘wraps around’ from one end to the other. The mathematics to calculate that, even with the LED brightness smoothly changing, is relatively simple. In twin-beam mode, with VR2 closer to the clockwise end than anti-clockwise, the chaser shifts so that there are two lit areas exactly four LEDs apart. Many lighthouses and sirens that use siliconchip.com.au actual rotating lights will emit light from both ends, so this mode better simulates that appearance. Construction The Rotating Light is built on a double-­ sided PCB coded 09101251 that measures 20 × 20mm. The top and bottom component overlay diagrams are in Fig.3, with two versions shown to suit SMD or through-hole LEDs. Refer to those during construction to see which parts go where. If you find the small board slides around while working on it, use a little Blu-Tack to temporarily stick it to your work surface. None of the components are terribly hard to solder individually. We found the main challenge to assembly was to avoid accidental bridges between adjacent pads because they are quite close together due to the small size of the PCB, especially the two trimpots and the two SOT-23 devices that mount side-by-side. So it’s best to feed in solder carefully and use the minimum necessary to form good fillets. The microcontroller IC has fairly widely spaced pins, on a 1.27mm pitch and other parts have larger or more widely spaced pins. So the actual soldering of individual components is not too difficult. As there are parts on both sides, once you have fitted the parts on one side, the PCB won’t easily sit flat and will tend to rock as you work on it. To deal with this, you can either use Blu-Tack as mentioned, or you could do what we did and place the PCB on top of a roll of solder-wicking braid. This has a hole in the middle for the components to fit in, so it can rest on its edges and sit flat. Of course, that depends on you having a similarly sized roll of braid to us, but it worked surprisingly well for us. Fig.3: the top side of the PCB has the microcontroller, both trimpots, one capacitor, one resistor and either SMD or through-hole LEDs, although TH LEDs can also be inserted from the bottom side. All the currentlimiting resistors are on the underside, along with most of the power supply. Australia's electronics magazine April 2025  61 There is no provision to program the microcontroller on the board, so you’ll need to either purchase a pre-­ programmed micro (from our Online Shop, either individually or in a kit), or program it yourself before soldering it. Our article on the PIC Programming Adaptor in the September 2023 issue (siliconchip.au/Article/15943) explains how you can do it. Once programmed, make sure you have identified pin 1 on the chip and lined it up with the marking on the PCB (very important!). Also check it against Fig.3, then tack-solder one pin. Adding a little flux paste will help the solder flow. Check the alignment of all the other pins with their pads (now is also a good time to double-check that pin 1 is in the right place!). If the positioning is not perfect, remelt the solder joint and gently nudge the chip into position. Once it’s located correctly, solder the diagonally opposite pin, then spread a little flux paste down both sides of the chip, over all the pins, and solder the remaining pins. You can drag solder them, or do them one at a time. If solder bridges have developed between any pins, clean them up by adding a dab of flux paste and then using a clean piece of solder wick to remove the excess solder. Clean off the flux residue with a suitable solvent, then inspect the pins under magnification to ensure all the solder joints are good (solder has flowed onto both the pin and pad) and no bridges remain. Solder the two trimpots similarly, being careful to avoid bridges to adjacent pads due to their proximity to IC pins and LED pads. We found the trimpots were the trickiest parts of all to solder because the pads didn’t stick out very far from underneath them. We’ve extended them in the final version of the PCB, but there was limited space available to do so. Add flux paste on both the PCB and component leads before soldering need to be careful to check that the solder has flowed down on the PCB pads before moving on to other components. With the trimpots soldered correctly, add the sole top-side capacitor and one resistor. Finally, if you are fitting SMD LEDs, you can do that now. Soldering the LEDs We recommend soldering standard SMD LEDs on their side, like in our photos. First, figure out which end of 62 Silicon Chip the LED is the cathode. You can do this with a DMM on diode test mode. When the LED lights up, the black probe is on the cathode. It must go to one of the pads marked “K” in Fig.3. Start soldering each SMD LED by adding solder to one of its pads. Due to the through-holes, you’ll need to add more than you might expect until you get sufficient solder on the top surface. You want a visible bulge so enough solder will reach the pad on the side of the LED, rather than the bottom as usual. The hardest part of soldering the SMD LEDs on their side was picking them up with the tweezers. We found the easiest way was to pick them up from the bench with one hand, rotate them on their side, then grab them with tweezers using the other hand. Make sure the tweezer tips don’t extend past the bottom of the LED or you won’t be able to get it to sit flat on the PCB. Once we had picked them up correctly, we found that soldering them wasn’t too hard. Position the LED with tweezers while keeping the solder on that initial pad molten with your soldering iron. Remove the iron for a few seconds to let it solidify, then check if the position is good. If it is, add a fillet to the other pad. The LEDs don’t need to be perfectly aligned but it helps if they are close. If you aren’t happy with the LED position, you can grab it again with the tweezers, reheat the initial joint and nudge it into place. Once both sides are soldered, you may need to add a dab of flux paste to the first pad and heat it to reflow the solder and form a nice, shiny fillet. With all the top-side components fitted, flip the board over and add the remaining SMDs, as shown in Fig.3. Don’t get D1 & REG1 mixed up. None of the other components are polarised. If using through-hole LEDs, bend their leads consistently and solder them in place now. You can insert them from either side of the PCB but make sure when you bend the leads that the shorter (cathode) lead will always go into a pad marked “K” in Fig.3. Now solder the power leads to their pads. They are marked with + and – symbols on one side of the PCB. You can solder them from either side. Testing If you have a current-limited supply, set it to 6V and 25-50mA. Otherwise, Australia's electronics magazine you could include a series resistor (eg, 100W 5W from a 12V supply) to limit the current in the event of a fault. Apply power and check the current flow. Depending on the trimpot positions, it should be around 10-20mA and should definitely not exceed 50mA. Verify that the LEDs light up and start to sequence. If the current draw is too high, switch off and inspect the board for faults, such as solder bridges between pads or pins, or incorrectly placed or orientated components. Perform similar checks if there is no current draw or nothing happens. Also check that all solder joints have been made correctly. If it operates but some LEDs don’t light, likely they are faulty, their solder joints are bad, or they are shorted to an adjacent pad. If it appears to be working, try adjusting VR1 & VR2 to verify that you can change the rotation speed, direction and beam brightness/ width as expected. We found that many of our Phillips head screwdrivers of various sizes failed to actually rotate the trimpot. We had to search around until we found a slotted screwdriver of the perfect size before we could get sufficient purchase. After that, we could make easy and precise adjustments. Usage With VR1 centred, rotation is very slow; if it is perfectly centred, the light will not rotate, or just barely. It ‘accelerates’ in either direction as you move towards the clockwise or anti-­ clockwise extremes. This gives a reasonable range of speed options without making the adjustment super fiddly. With VR2 centred, you will have a narrow (45°), dim beam. As you move it anti-clockwise, the beam will first start to brighten, then widen. At about halfway between anti-­clockwise and the centre, you will have a bright 45° beam. As you move closer to anti-clockwise, the beam will get wider and wider until it occupies almost the whole circumference. Rotating it from the centre clockwise is similar except that you will have two opposing 45° beams that get brighter, then wider. If you want to power this board from a USB supply, we have an upcoming article on USB Power Adaptors. You would need to join the two boards with a short length of light-duty twin lead or similar. SC siliconchip.com.au SOnline ilicon Chip Shop Kits, parts and much more www.siliconchip.com.au/Shop/ Compact OLED Clock & Timer September 2024 Short-Form Kit SC6979: $45 siliconchip.au/Article/16570 This kit includes everything needed to build the OLED clock, except the UB5 Jiffy box and Li-ion cell. Dual Mini LED Dice August 2024 Micromite-Explore 40 October 2024 Complete Kit SC6991: $35 SMD LED Complete Kit SC6961: $17.50 TH LED Complete Kit SC6849: $17.50 siliconchip.au/Article/16418 siliconchip.au/Article/16677 Includes either 3mm through-hole or 1206sized SMD LEDs. Choice of either white or black PCB. CR2032 coin cell not included. Includes the PCB and all onboard parts. Audio Breakout board and Pico BackPack are sold separately. Compact HiFi Headphone Amplifier Complete Kit SC6885: $70 Capacitor Discharger December 2024 December 2024 & January 2025 siliconchip.au/Series/432 This kit includes everything required to build the Compact HiFi Headphone Amplifier. The case is included, but you will need your own power supply. Programmable Frequency Divider Complete Kit SC6959: $60 Feb25: siliconchip.au/Article/17733 Includes all onboard components, except for a power supply and the optional programming header. Short-Form Kit SC7404: $30 siliconchip.au/Article/17310 Includes the PCB, resistors, semis, mounting hardware and banana sockets. Case, heatsink, thermal switch and wiring are not supplied. → Subscribers receive a 10% discount on all purchases, except for subscriptions (postage is not discounted). → Prices listed do not include postage. Postage rates within Australia start at $12, rates are calculated at the checkout. By Andrew Levido Precision Electronics Part 6: Digital-to-Analog Conversion The first five parts of this series have concentrated on sources of error & imprecision in analog circuitry, such as gain errors, offset voltages, bias currents & noise. These days, most analog circuitry eventually interfaces with the digital realm, such as with a microcontroller, so we also need to understand the precision approach to ADCs & DACs. I n this series so far, we have covered precision electronics from a purely analog perspective. We designed a simple multi-range current-sensing circuit that could be used in a bench-top power supply. It had an analog precision of around ±0.06% at 25°C after trimming out fixed errors. The idea was that the analog measurement voltage would be digitised via an analog-to-digital converter (ADC), with the trimming and calibration carried out digitally. We will therefore ultimately need to design an ADC subsystem that will have similar or better performance to the analog circuitry we have already developed. However, the topic of interfacing digital and analog electronics is a big one and quite complex. So I am going to cover it over two articles, starting here with some general principles and then taking a look a digital-to-analog converters (DACs) with a real-world example, before moving on to ADCs next month. We will also need to take a look at voltage references in a later article, since any discussion of ADCs/DACs would be otherwise incomplete. Quantisation errors A digital system like a microcontroller has to represent an analog quantity using a binary-coded number. This can be a simple unsigned integer (zero or a positive whole number) if we are representing a quantity that is always positive, or an offset binary or signed (two’s complement) integer if the quantity can be both positive and negative. Table 1 shows how typical ADCs and DACs represent such quantities, using 8-bit numbers as an example. Notice that the resolution of the two’s-complement and offset integers are lower because the input span is twice the full scale value. It might seem obvious, but you need to take this into account when choosing a converter. It’s all too easy to fall into the trap of thinking about Fig.1: the relationship between analog input or output and digital code for an ideal 3-bit converter. (b) is more representative of real-life devices where the quantisation error is usually ±½LSB. (a) 64 Silicon Chip (b) Australia's electronics magazine full-scale instead of span in these circumstances. In an ideal converter, the relationship between the analog input or output and the digital code will be as shown in Fig.1. The horizontal axis is the digital code (in this case, three bits for simplicity), while the vertical scale is the analog value as a fraction of full scale. The green line represents an ideal transfer function, but since the digital codes are discrete, there must be transition points, shown by the black dots. Taking the chart on the left as an example, for an ADC, the code would be zero for an input of zero volts and it would remain so for input voltages up to ⅛ of full scale, at which point there will be a transition to the 001 code. For a DAC, the output voltage will be somewhere below ⅛ of full scale if the code is 000. It’s obvious that there is a degree of error inherent in any such converter, since the transition points The vertical axes in Figs.2a-2d are the analog voltages. (a) siliconchip.com.au Table 1 – ADC and DAC numerial coding schemes Voltage Unsigned Integer Two's Complement Offset Binary +Full Scale 11111111 01111111 11111111 +Full Scale − 1 ... 11111110 ... 01111110 ... 11111110 ... +1 00000001 00000001 10000001 0 00000000 00000000 10000000 −1 ... – ... 11111111 ... 01111111 ... −Full Scale + 1 – 10000001 00000001 −Full Scale – 10000000 00000000 are separated by one least significant bit (LSB) of the input code. This is called quantisation error and is an inescapable consequence of the discrete nature of digital systems. In this example, the quantisation error at any point on the transfer function will be somewhere between zero and one LSB. The chart on the right side of Fig.1 is a more realistic example of how a converter is configured; the ideal transfer function is the same, but the transitions are shifted ½ of one LSB so that they occur between the nominal analog values. In this case, the quantisation error is the same magnitude, but is ±½ LSB either side of the nominal value. Quantisation error can add noise to an AC signal. If we were to apply a linear ramp signal to an ADC, or try to generate a linear ramp with a DAC, the error would look like that shown at the bottom of Fig.1b. The quantisation error would appear as a sawtooth wave with an amplitude of ±½LSB. We can calculate the signal-to-noise ratio (SNR) of a digitised sawtooth or triangular waveform. If we have a converter with n bits, the maximum amplitude of an AC waveform can be ½ × 2n LSB. The noise amplitude is ½LSB, so the SNR is 20log10(2n) or approximately 6.02n decibels. But this only applies to sawtooth and triangle waveforms where the signal has a uniform error distribution. For sinewaves, we have to use the approximation SNR = 1.76 + 6.02n decibels to allow for the uneven error distribution. For an 8-bit converter, the SNR due only to quantising will be around 50dB, and for a 12-bit converter, it will be 74dB. This is pretty significant; hence, high-fidelity audio ADCs and DACs use many bits and careful filtering to maximise the SNR. Quantisation error is therefore defined by the resolution of the converter. You obviously need to select a converter with sufficient bits to give you the resolution that your application requires. Typically, you need even more bits to account for some of the other errors that can occur. Further ADC & DAC errors You will see ADC and DAC errors expressed in a range of terms, so it can be a bit confusing at first. We have already seen quantisation error expressed in least significant bits (LSB), but you will also see errors expressed as relative errors (percentages or parts per million) and in absolute terms like millivolts. You can convert LSB to a percentage error by dividing it by 2n (where n is the number of bits, ie, the resolution) and applying the appropriate scaling. For example, the ½LSB quantisation error on an 8-bit converter will be about 0.2% (100% × ½ ÷ 28), while on a 12-bit converter, it would be 122ppm (106 × ½ ÷ 212). We have seen in past instalments that it can be useful to have errors in both absolute and relative terms, depending on whether we are adding or multiplying uncertain quantities. Fig.2 shows the four most common types of error that are relevant for ADCs and DACs. Offset error (Fig.2a) is a fixed shift in transition points away from their ideal locations. If measured in LSB, it is defined by the difference between the value of the first code transition and its ideal value. It is most often specified as an absolute voltage, just like an op amp’s offset voltage. As you might expect, there can also be a gain error (Fig.2b) if the slope of the transfer function deviates from the ideal. In LSB, it is defined by the Fig.2: ADCs and DACs have four main types of error shown in these graphs: offset error, gain error, integral nonlinearity (INL) and differential nonlinearity (DNL). If the DNL exceeds ±1LSB, the converter can exhibit non-monotonicity, as shown in (d). (b) siliconchip.com.au (c) Australia's electronics magazine (d) April 2025  65 difference between the last code transition point and its ideal counterpart, but it is more often specified as a relative error. There is also the possibility that the transfer function will deviate from the ideal by not being completely linear (Fig.2c). There are two common measures of linearity error: differential nonlinearity (DNL) and integral nonlinearity (INL). INL is the maximum deviation of the transfer function from the ideal over the whole conversion range, while DNL is the maximum difference between the width of a code and its ideal width (1 LSB). A DNL of more than ±1LSB implies a loss of monotonicity, as shown in Fig.2d. A monotonic curve is one that always increases (or decreases). Fig.3: a resistor string DAC has 2n matched resistors forming a voltage divider. Analog switches select one ‘tap’ off this divider for each input code. A 16bit DAC of this type will have 65,536 matched resistor and a similar number of analog switches! Fig.4: an R-2R DAC uses just 2n matched resistors and switches to achieve 2n output voltage steps. This architecture is also useful for lowresolution ‘roll-your-own’ DACs using microcontroller GPIOs in the place of the analog switches. 66 Silicon Chip The INL is the specification to care about if you are looking for the best overall accuracy – for example, to generate or measure a voltage with minimal error. However, if you are using the DAC or ADC in a control loop, you may want to focus on DNL. The control loop’s servo action will look after the INL if it is relatively ‘smooth’, but ‘patches’ of inconsistent gain (or worse, non-monotonicity) can cause control glitches like dead spots or even points of instability. Total unadjusted error The ‘total unadjusted error’ (TUE) is a figure that describes the total maximum error for a converter. This is very handy for calculating the error budget. Sometimes manufacturers specify the TUE in the data sheet – either in LSB or as a relative error – but you can calculate it yourself if necessary. To do so, you convert the offset, gain and INL errors to the same format, and add them using the root-sumof-squares method (since the error sources are uncorrelated). We will do this for our DAC example later in this article. Resistor string DACs Enough theory – let’s take a look at a few practical DACs. One common (and fairly obvious) way to construct a DAC is with a resistor string, as shown in Fig.3. A string of 2n equal-value resistors are used together with a set of binary-weighted analog switches to switch one ‘tap’ of the string to a buffer and out to an external pin. The output voltage is Vref(N ÷ 2n), where n is the DAC resolution in bits and N is the input code, which ranges from zero to 2n – 1. This type of DAC is guaranteed monotonic by design, and can have quite good linearity since it is possible to match on-chip resistors well. They can also have good temperature stability for the same reason. It is possible to get DACs with up to 16 bits of resolution that use this architecture. The AD5689 we will use in our test circuit is a good example. This means the chip contains a string of 65,536 matched resistors and a similar number of double-throw analog switches, or equivalent, for each channel. Amazing! R-2R ladder DACs The R-2R ladder is a variant on the Australia's electronics magazine resistor string DAC that uses a lot fewer resistors. Instead of requiring 2n resistors, we can get away with just 2n, using the circuit shown in Fig.4. Only one double-throw analog switch is required for each bit. The simplified circuit means you can get R-2R DACs with up to 20 bits or more of resolution. They also can have quite good linearity and temperature coefficients. The output voltage is Vref(N ÷ 2n), just like the resistor string DAC. Another useful property of the R-2R ladder DAC is that you can easily improvise one with an op amp, a handful of resistors and a few digital outputs. The analog switches in Fig.4 effectively switch between Vref and 0V, so they could be replaced with totem-pole digital outputs (say, microcontroller GPIOs), creating a basic 3-bit DAC. The performance will be average, since the reference voltage will be the digital supply voltage and you will probably use 1% resistors, but if you only need a few levels, this can be a handy technique to create a ‘free’ DAC. Current output & multiplying DACs A variation of the R-2R ladder DAC that provides an output current rather than a voltage allows us to build ‘multiplying DACs’. Strictly speaking, all DACs effectively multiply the reference voltage by the digital code, but many have internal references or external ones that only accept voltages of one polarity. A multiplying DAC can operate in two or four quadrants, as shown in Fig.5. In both cases, the DACs, shown inside the dashed box, are identical. Since the Iout pin is sitting at 0V courtesy of op amp IC1, the current coming out of the pin is N ÷ 2n × Vin ÷ R. You can see from the circuit that this equation will hold even if Vin is negative (in which case, the current will flow into the pin). This current is converted to a voltage by IC1, which is configured as an inverting amplifier. The output voltage of the two-quadrant multiplying DAC will be -N ÷ 2n × Vin, where Vin can be positive or negative. You will notice that the feedback resistor is provided within the DAC IC and is matched to the resistors in the siliconchip.com.au Fig.5: multiplying DACs use a current source architecture to achieve twoquadrant or four-quadrant operation. In both circuits, the reference can be of either polarity or an AC voltage. R-2R ladder. This is important because the semiconductor manufacturing process can produce on-chip resistors that are very well matched in value or ratio, but their absolute value is more difficult to control. Using an external resistor would almost certainly introduce large gain errors and poor temperature stability. Fig.5 also shows a four-quadrant version of the same circuit, which is identical in operation to the two-quadrant one but has an added (inverting) summing amplifier stage (IC2) that scales up the DAC output and offsets it by Vin. If we consider the code to be a signed value using the offset binary representation, we can effectively multiply a bipolar input voltage by a positive or negative integer. Multiplying DACs can be very useful in signal processing, for example, as a very fine-grained programmable gain stage. They are also useful in making precise ratiometric measurements. For example, if you excite some sensor (such as strain gauge) using a voltage produced by a multiplying DAC and digitise the resulting signal with an ADC that uses the same voltage reference, any error or drift in the reference is cancelled. The resulting readings are the true ratio of input to siliconchip.com.au output independent of the excitation voltage. Delta-sigma DACs Another class of DACs worth mentioning are the delta-sigma converters that you frequently encounter in audio applications. Delta-sigma DACs typically have very high resolution (20+ bits) to minimise quantisation noise and have spectacular linearity to ensure low harmonic distortion. We don’t normally use delta-sigma DACs in precision applications because their DC performance is generally not great, probably because this does not matter in audio applications. Oddly, there are plenty of very high-precision delta-sigma ADCs that work very well at DC, as we shall see next time. Pulse-width modulation We should not neglect pulse-width modulation (PWM) as a potential type of DAC, especially in microcontroller circuits where dedicated PWM peripherals are commonplace. Fig.6 shows the simplest possible configuration with a PWM output and an RC low-pass filter. The time constant of the low-pass filter has to be much longer than the PWM period (Tpwm) to produce an average of the PWM waveform Australia's electronics magazine Fig.6: a PWM DAC can be as simple as a microcontroller’s PWM output and an RC filter. The lower circuit uses a complimentary PWM signal to improve ripple and settling time eightfold. proportional to its duty cycle. A longer time constant with respect to Tpwm means better averaging and a lower output ripple. The worst-case ripple occurs at 50% duty cycle and is given by Vripple = Vfs(Tpwm ÷ 4RC). The downside of having a long time constant (and lower ripple) is a slow response of the output voltage to changes in the PWM duty cycle. Stephen Woodward published a really neat technique to address this problem in an EDN article published in 2017. Woodward showed that the time for the output voltage to settle to within the ripple voltage for a given change in duty cycle is Tsettle = RC × loge(Vfs ÷ Vripple). If we wanted to make a PWM DAC equivalent to a conventional DAC with 8-bit resolution, we would require that the peak-to-peak ripple be 1LSB (or 1/256 of the full-scale voltage), corresponding to the quantising noise. The first equation above tells us this requires an RC time constant 64 times the PWM period. The settling time (from the second equation) will therefore be 355 × Tpwm. Depending on what you are doing, that could be a long time! For 10kHz PWM, this is a settling time of 35ms. If you wanted a 10-bit resolution, the settling time would be even worse at 177ms. April 2025  67 Woodward’s ingenious solution is shown at the bottom of Fig.6. Here, an inverted version of the PWM signal is injected into the output via a series RC network to cancel the ripple. I have shown this coming from a complimentary PWM output on the microcontroller, but a logic gate inverter would work equally well. If the new resistor and capacitor are the same value as the original ones, the ripple equation becomes Vripple = ½Vfs × (Tpwm ÷ 4RC)2 and the settling time becomes Tsettle = ½Tpwm × √(Vfs ÷ Vripple) × loge(Vfs ÷ Vripple). Using the 8-bit example above, the RC time constant required to achieve the ripple target reduces from 64 to just four PWM periods, and the settling time reduces from 355Tpwm to 44Tpwm (or 4.4ms at 10kHz). This is an eightfold improvement in settling time! In the 10-bit case, the settling time reduces to 11ms from 177ms for the original circuit. It turns out you don’t even need precision components to achieve these improvements. It is sufficient to use 1% tolerance resistors and 10% tolerance capacitors. This is a circuit well worth knowing. Other DAC types If we agree that a duty-cycle-to-­ analog converter like the PWM example above is a form of DAC, we should also consider a frequency-to-voltage converter to be one too. Fig.7 shows a typical example using an LM331 IC. This circuit works as follows. A square wave of frequency Fin is differentiated by the RC network connected to pin 6, creating negative-­ going spikes on each falling edge. When these fall below the threshold set by the resistor divider connected to pin 7, the upper comparator sets the RS flip-flop. When the flip-flop is set, the discharge transistor on pin 5 is switched off, and capacitor Ct begins to charge via Rt. When this voltage reaches 2/3Vcc, the lower comparator resets the flip-flop, discharging Ct. This means the flip-flop is set for a fixed duration each time a falling edge on the input occurs. The analog switch connected to the flip-flop’s output directs a fixed current out of pin 1 during this period. The current is converted to a voltage by Rl and averaged by the RlCl lowpass filter. Fig.7: a frequency-to-voltage converter is also a useful type of DAC. This circuit shows how an LM331 can be used to create a simple but quite respectable DAC. 68 Silicon Chip Australia's electronics magazine The level of this current is precisely controlled by the circuit connected to pin 2. An on-board bandgap reference and op amp ensure a precise 1.9V is always present on pin 2, meaning a current of 1.9V ÷ Rs flows out of this pin and therefore out of the current mirror to the analog switch. A practical example Fig.8 shows an extract of a circuit I designed some time ago for a precision instrument. This part of the circuit includes a voltage reference, a two-channel DAC and a couple of op amps. The output is intended to be a ±2.0V precision voltage programmable by the microcontroller. I will step through the operation and error analysis for this circuit, so it will be helpful to follow both the schematic and the error budget (Table 2). The operating temperature range for this device is 15-35°C, reflecting its intended use in a laboratory setting. The MAX6225 provides a very stable, very accurate 2.5V reference (±200ppm initial accuracy, ±2ppm/°C). An inverted copy of the reference is provided by the inverting amplifier (IC1), an LTC2057 zero-drift op amp. The inverting op amp uses relatively high value resistors (10kW) to minimise the load on the precision reference. These resistors are high-precision (0.01%) types with very low thermal drift (±5ppm/°C), since we don’t want to compromise the performance of an expensive reference. The LTC2057 has very low input offset voltage (±4µV) and even lower offset drift (±15nV/°C), as we would expect from a “zero drift” op amp. Because we have used relatively high value resistors, I have included the op amp’s input offset current error in the table on line 3. You can see from line 4 of the table that the total error at the input of the reference inverter is dominated by the reference error and is 0.02%. You can see why I selected 0.01% resistors for this circuit – to keep the gain error down to a similar order of magnitude as the reference error. This leads to a total error for the negative reference of 0.047% over the temperature range, compared to a total error of 0.022% for the positive reference. I was not 100% happy with the negative reference error, but figured that as I was only making a handful of these siliconchip.com.au Fig.8: this is an excerpt from a circuit which uses a voltage reference, a DAC and a couple of op amps to create a programmable ±2V with a resolution of less than 200µV. The untrimmed error is better than ±0.03%. devices, my results would likely be much better. The odds of two identical resistors being at the opposite extremes of tolerance are low enough that if I did find an outlier, I could manually select resistors to fix it. The measured results show both references to be within 0.01% of each other on the prototype. The positive 2.5V reference is applied to the DAC. I used a very nice dual-channel, 16-bit resistor-string DAC. It can be configured for a gain of one or two. I used a gain of two to make the subsequent circuit design simpler. This means the full-scale output is 5.0V. In this configuration, the DAC has an offset error of ±1.5mV, a gain error of 0.1% and an INL of ±1LSB. Thanks to the manufacturer for specifying the three key figures in three different ways! That said, this is pretty good performance for a DAC, especially the INL. I converted all these figures to relative errors in the table and added them using the root-sum-of-squares method to arrive at a TUE of 0.104%, dominated by the gain error. The data sheet actually provides a TUE figure of 0.1%, so I did this exercise to just demonstrate how TUE is calculated. It is nice when theory and reality agree! Table 2: positive/negative voltage references, DAC & offset amp errors Error Since the voltage at the output of the DAC is the reference multiplied through the DAC coefficient, the total error is calculated as the sum of relative errors on line 12. We get a total error here of 0.127% over the full temperature range (0.124% + 0.003%), dominated by the DAC gain error of 0.1%. I’m not usually a proponent of ‘typical’ specifications, but if our DAC was within the typical range, the total error would be closer to 0.05%. Finally, the 0-5V DAC output is summed with the negative reference (and inverted) by IC2. This produces an output voltage that ranges from +2.5V when the DAC code is zero At Nominal 25°C Additional error over 15-35°C (Nominal ±10°C) Nominal Value Abs. Error Rel. Error Abs. Error Rel. Error 1 MAX6225ACASA+ (±200ppm, ±2ppm/˚C) 2.5V 500μV 0.02% 0.002% 2 Op Amp: LTC2057 (Vos ±4µV, 15nV/˚C) 0V 4μV 100nV 3 Op Amp Ios × 10kW || 10kW: LM7301 (Ios ±400pA, ±1pA/˚C) 0V 2μV 50nV 4 Voltage at Op Amp Input (Line 1 + Line 2 + Line 3) 2.5V 506μV 5 Op Amp Gain: R/R Stackpole RNCF0603TKY10K00 (0.01%, 5ppm/˚C) 1 6 +Vref error (Line 1) 2.5V 500μV 0.02% 50μV 0.002% 7 −Vref error (Line 4 × Line 5) 8 DAC Offset error: AD5689 (±1.5mV, ±1µV/˚C) -2.5V 1mV 0.04% 175.2μV 0.007% 5V 1.5mV 0.03% 10μV 0.000% 0.02% 50μV 50.2μV 0.02% 0.002% 0.005% 9 DAC Gain error: AD5689 (0.1%, ±1ppm/˚C) 2 0.1% 0.001% 10 DAC Linearity: AD5689 (INL ±1LSB, DNL ±1LSB) 0 0.002% 0.000% 11 DAC total unadjusted error (root sum of squares Lines 8-10) 5V 5.2mV 0.104% 51μV 0.001% 12 DAC total error (Line 6 × Line 11) 5V 6.2mV 0.124% 620μV 0.003% 13 Op Amp: LTC2057 (Vos ±4µv, 15nV/˚C) 0V 4μV 150nV 14 Op Amp Ios × 1kW || 1kW || 1kW: LM7301 (Ios ±400pA, ±1pA/˚C) 0V 133.2nV 3.3nV 15 Op Amp Gain: R/R Stackpole ACASA1002U1002P1AT (0.05%, 5ppm/˚C) 1 0.05% 0.005% 16 Voltage at Op Amp Input (Line 7 + Line 12 + Line 13 + Line 14) 2.5V 7.2mV 0.29% 250.6μV 0.01% 17 Voltage at Op Amp Output (Line 15 × Line 16) 2.5V 8.5mV 0.34% 375.6μV 0.015% Silicon Chip kcaBBack Issues $10.00 + post $11.50 + post $12.50 + post $13.00 + post January 1997 to October 2021 November 2021 to September 2023 October 2023 to September 2024 October 2024 onwards All back issues after February 2015 are in stock, while most from January 1997 to December 2014 are available. For a full list of all available issues, visit: siliconchip.com.au/ Shop/2 PDF versions are available for all issues at siliconchip.com.au/Shop/12 (0x0000), to almost -2.5V when the DAC code is full scale (0xFFFF). The error budget for this circuit is similar to the previous examples. This time, I used low-cost resistor arrays with ±0.05% tolerances on the matching. If I had used individual 0.01% resistors for this, there would be a 0.03% error due to there being three resistors involved. The improvement of 0.02% does not justify the extra cost given the other sources of error in the circuit. There are a couple of things worth noting before we discuss the results this circuit produced. First, you will see that I am only using a range of ±2V, not the full ±2.5V the circuit is capable of. This is because near zero and full scale, the DAC output is prone to errors, since these are right at its power supply rails. We already know that no output can truly swing all the way to the rails. Avoiding the ends of the span costs us some precision, since the whole range of codes is not used. In this case, the valid codes are 6553 to 58,981 for a 4V span, giving us a resolution of about 191µV, which is plenty for my application. You should always avoid the extremes of DAC & ADC ranges in precision applications. You can go much closer than I have here, but there will probably be errors right at the edges. I have also taken a lot of care with the power supplies. It is not worth spending good money on precision components and skimping on the power supply. The digital and analog supplies come from separate linear regulators. I included a ferrite bead on the analog supply to the DAC, more to protect the rest of the circuit from glitches caused by the DAC switching than vice versa. Results I first measured the positive and negative references and found they were +2.50015V and -2.49989V, both well under 0.01% away from the nominal value and within 0.01% of each other. I measured the output voltages at code intervals of 400 hexadecimal (1024 decimal) over the full range. With a code of zero, I measured +2.49930V (0.028% error). At 0x8000, I measured 481µV (0.019% error), while at full scale (0xFFFF), I measured -2.45720 (-1.7% error). As mentioned above, we expect the extremes to be poor. If we look at the range of interest, the error is never worse than +0.024%; in fact, it is also never less than +0.018%, suggesting we have an offset error, albeit a small one. Sure enough, the absolute error ranges from 460µV to 610µV and averages 550µV. Can we trim this error somehow? My circuit also includes an ADC, and switching that allows me to measure the voltage we are concerned with. If we were to measure the voltage with the code 0x8000 (corresponding with 0V out), we would be able to measure the 481µV offset and correct for it in software. We could similarly measure the voltage at either end of the span (±2V nominal) and correct for that. This is easier said than done, and we will look into it in more detail in a later article. I also performed a full noise analysis of this circuit, which I have included in Table 3. I won’t go into the gory details since I used the same techniques I described in the last article. Overall, the RMS noise voltage should be around 1.4µV over a 10Hz bandwidth. That does not include quantisation noise, since this is a DC application. The biggest contributor is the DAC noise, with the reference coming in next. Given the 191µV resolution of the DAC, this level of noise is not going to impact the precision of our circuit. In summary, we can almost certainly get to an overall precision close to that of the reference at 0.02%, and it is pretty hard to do any better than SC that! Table 3 – noise analysis with 10Hz nominal bandwidth Noise Source Notes Noise Voltage (RMS) 1 Positive reference noise MAX6225 (15nV/√Hz, fc100Hz) fb straddles fc so use f = fb + fcloge(10) = 240Hz 232.4nV 2 Ref inverter amp voltage noise: LTC2057 (11nV/√Hz) Data shows noise flat from 0.1Hz to 10Hz 34.8nV 3 Ref inverter amp voltage noise: LTC2057 (170fA/√Hz) 10kW || 10kW resistors in inverting input 2.68nV 4 Ref inverter 10kW input resistor & feedback resistor √4kTRfb 40.7nV 5 Negative reference noise Line 1 + Noise Gain (2) × Line 2 # 237.5nV 6 DAC noise AD5689 (300nV/√Hz) Use fb = 10Hz 948.7nV 7 DAC output noise Line 6 + Line 1 1.2µV 8 Summing amp voltage noise: LTC2057 (11nV/√Hz) Data shows noise flat from 0.1Hz to 10Hz 34.8nV 9 Summing amp voltage noise: LTC2057 (170fA/√Hz) 10kW || 10kW || 10kW resistors in inverting input 1.79nV 10 10kW input resistors & feedback resistor √4kTRfb 40.7nV 11 Scaled output noise Line 7 + Line 5 # 1.4µV # other errors are at least an order of magnitude smaller, so they can be ignored 70 Silicon Chip Australia's electronics magazine siliconchip.com.au SILICON CHIP .com.au/shop ONLINESHOP HOW TO ORDER INTERNET (24/7) PAYPAL (24/7) eMAIL (24/7) MAIL (24/7) PHONE – (9-5:00 AET, Mon-Fri) siliconchip.com.au/Shop silicon<at>siliconchip.com.au silicon<at>siliconchip.com.au PO Box 194, MATRAVILLE, NSW 2036 (02) 9939 3295, +612 for international You can also pay by cheque/money order (Orders by mail only) or bank transfer. Make cheques payable to Silicon Chip. 04/25 YES! You can also order or renew your Silicon Chip subscription via any of these methods as well! The best benefit, apart from the magazine? Subscribers get a 10% discount on all orders for parts. PRE-PROGRAMMED MICROS For a complete list, go to siliconchip.com.au/Shop/9 $10 MICROS $15 MICROS ATmega328P ATtiny45-20PU PIC10LF322-I/OT PIC12F617-I/P 110dB RF Attenuator (Jul22), Basic RF Signal Generator (Jun23) ATSAML10E16A-AUT High-Current Battery Balancer (Mar21) 2m VHF CW/FM Test Generator (Oct23) PIC16F1847-I/P Digital Capacitance Meter (Jan25) Range Extender IR-to-UHF (Jan22) PIC16F18877-I/P USB Cable Tester (Nov21) Active Mains Soft Starter (Feb23), Model Railway Uncoupler (Jul23) PIC16F18877-I/PT Dual-Channel Breadboard PSU Display Adaptor (Dec22) Battery-Powered Model Railway Transmitter (Jan25) Wideband Fuel Mixture Display (WFMD; Apr23) PIC12F617-I/SN Model Railway Carriage Lights (Nov21) PIC16F88-I/P Battery Charge Controller (Jun22), Railway Semaphore (Apr22) PIC12F675-I/P Train Chuff Sound Generator (Oct22) PIC24FJ256GA702-I/SS Ohmmeter (Aug22), Advanced SMD Test Tweezers (Feb23) ESR Test Tweezers (Jun24) PIC16F1455-I/P Auto Train Controller (Oct22), GPS Disciplined Oscillator (May23) Railway Points Controller Transmitter / Receiver (2 versions; Feb24) PIC32MX170F256D-501P/T 44-pin Micromite Mk2 (Aug14), 4DoF Simulation Seat (Sep19) Battery-Powered Model Railway TH Receiver (Jan25) PIC32MX170F256B-50I/SP Micromite LCD BackPack V1-V3 (Feb16 / May17 / Aug19) PIC16F1455-I/SL Battery Multi Logger (Feb21), USB-C Serial Adaptor (Jun24) Advanced GPS Computer (Jun21), Touchscreen Digital Preamp (Sep21) Battery-Powered Model Railway SMD Receiver (Jan25) PIC32MX170F256B-I/SO Battery Multi Logger (Feb21), Battery Manager BackPack (Aug21) USB Programmable Frequency Divider (Feb25) PIC32MX270F256B-50I/SP ASCII Video Terminal (Jul14), USB M&K Adaptor (Feb19) PIC16LF1455-I/P New GPS-Synchronised Analog Clock (Sep22) $20 MICROS PIC16F1459-I/P K-Type Thermostat (Nov23), Secure Remote Switch (RX, Dec23) ATmega32U4 Wii Nunchuk RGB Light Driver (Mar24) Mains Power-Up Sequencer (Feb24 | repurposed firmware Jul24) ATmega644PA-AU AM-FM DDS Signal Generator (May22) 8-Channel Learning IR Remote (Oct24) PIC32MK0128MCA048 Power LCR Meter (Mar25) PIC16F1459-I/SO Multimeter Calibrator (Jul22), Buck/Boost Charger Adaptor (Oct22) PIC16F15214-I/SN Digital Volume Control Pot (SMD; Mar23), Silicon Chirp Cricket (Apr23) $25 MICROS PIC16F15214-I/P Digital Volume Control Pot (TH; Mar23), Filament Dryer (Oct24) PIC32MX170F256B-50I/SO + PIC16F1455-I/SL Micromite Explore-40 (SC5157, Oct24) PIC16F15224-I/SL Multi-Channel Volume Control (OLED Module; Dec23) PIC32MX470F512H-120/PT Micromite Explore 64 (Aug 16), Micromite Plus (Nov16) NFC IR Keyfob Transmitter (Feb25), Rotating Light (Apr25) PIC32MX470F512L-120/PT Micromite Explore 100 (Sep16) PIC16F18146-I/SO Volume Control (Control Module, Dec23), Coin Cell Emulator (Dec23) $30 MICROS Compact OLED Clock & Timer (Sep24), Flexidice (Nov24) PIC16LF15323-I/SL Remote Mains Switch (TX, Jul22), Secure Remote Switch (TX, Dec23) PIC32MX695F512H-80I/PT Touchscreen Audio Recorder (Jun14) PIC32MZ2048EFH064-I/PT DSP Crossover/Equaliser (May19), Low-Distortion DDS (Feb20) STM32G030K6T6 Variable Speed Drive Mk2 (Nov24) DIY Reflow Oven Controller (Apr20), Dual Hybrid Supply (Feb22) W27C020 Noughts & Crosses Computer (Jan23) KITS, SPECIALISED COMPONENTS ETC siliconchip.com.au/Shop/ PICO/2/COMPUTER (SC7468) (APR 25) MICROMITE EXPLORE-40 KIT (SC6991) (OCT 24) ROTATING LIGHT FOR MODELS KIT (APR 25) DUAL-RAIL LOAD PROTECTOR (SC7366) (OCT 24) PicoMSA PARTS (SC7323) (SEP 24) COMPACT OLED CLOCK & TIMER KIT (SC6979) (SEP 24) DISCRETE IDEAL BRIDGE RECTIFIER (SEP 24) DUAL MINI LED DICE (AUG 24) AUTOMATIC LQ METER KIT (SC6939) (JUL 24) ESR TEST TWEEZERS COMPLETE KIT (SC6952) (JUN 24) DC SUPPLY PROTECTOR (JUN 24) Includes an assembled PCB, separate Raspberry Pi Pico 2 and front/rear panels $120.00 Includes all required parts (see p83, Oct24) Complete kit which includes the PCB and all onboard components (see p60, Apr25): - SMD LEDs (SC7462) $20.00 - Through-hole LEDs (SC7463) $20.00 433MHz TRANSMITTER KIT (SC7430) (APR 25) PICO 2 AUDIO ANALYSER SHORT-FORM KIT (SC6772) (MAR 25) USB PROGRAMMABLE FREQUENCY DIVIDER (SC6959) (FEB 25) NFC PROGRAMMABLE IR KEYFOB (SC7421) (FEB 25) COMPACT HIFI HEADPHONE AMP (SC6885) (DEC 24) CAPACITOR DISCHARGER KIT (SC7404) (DEC 24) Includes the PCB and all onboard parts (see p75, Apr25) The Pico Audio Analyser kit from Nov23, but with an unprogrammed Pico 2 Complete kit: includes all components (see p85, Feb25) Complete kit: includes all required items, except the cell (see p67, Feb25) Complete kit: includes everything except the power supply (see p47, Dec24) Includes the PCB and all components that mount on it, the mounting hardware (without heatsink) and banana sockets (see p36, Dec24) PICO COMPUTER $20.00 $50.00 $60.00 $25.00 $70.00 $30.00 (DEC 24) For full functionality both the Pico Computer Board and Digital Video Terminal kits are required, see page 71 in the December 2024 issue for more details. - Pico Computer Board kit (SC7374) $40.00 - Pico Digital Video Terminal kit (SC6917) $65.00 Separate/Optional Components: - PWM Audio Module kit (SC7376) $10.00 - ESP-PSRAM64H 64Mb SPI PSRAM chip (SC7377) $5.00 - DS3231 real-time clock SOIC-16 IC (SC5103) $7.50 - DS3231MZ real-time clock SOIC-8 IC (SC5779) $10.00 FLEXIDICE COMPLETE KIT (SC7361) Includes all required parts except the coin cell (see p71, Nov24) (NOV 24) $30.00 Hard-to-get parts: includes the PCB and all semiconductors except the optional/variable diodes (see p73, Oct24) Hard-to-get parts: includes the PCB, Raspberry Pi Pico (unprogrammed), plus all semiconductors, capacitors and resistors (see p63, Sep24) Includes everything except the case & Li-ion cell (see p34, Sep24) $35.00 $35.00 $50.00 $45.00 Both kits include the PCB and everything that mounts to it (see page 83, Sep24) - All through-hole (TH) kit (SC6987) $30.00 - SMD kit (SC6988) $27.50 Complete kit: choice of white or black PCB solder mask (see page 50, August 2024) - Through-hole LEDs kit (SC6849) $17.50 - SMD LEDs kit (SC6961) $17.50 Includes everything except the case & debugging interface (see p33, July24) - Rotary encoder with integral pushbutton (available separately, SC5601) Includes all parts and OLED, except the coin cell and optional header - 0.96in white OLED with SSD1306 controller (also sold separately, SC6936) All kits come with the PCB and all onboard components (see page 81, June24) - Adjustable SMD kit (SC6948) - Adjustable TH kit (SC6949) - Fixed TH kit – ZD3 & R1-R7 vary so are not included (SC6950) USB-C SERIAL ADAPTOR COMPLETE KIT (SC6652) Includes the PCB, programmed micro and all other required parts *Prices valid for month of magazine issue only. All prices in Australian dollars and include GST where applicable. # Overseas? Place an order on our website for a quote. (JUN 24) $100.00 $3.00 $50.00 $10.00 $17.50 $22.50 $20.00 $20.00 PROJECT BY TIM BLYTHMAN 433MHz Transmitter Module The 433MHz LIPD (low interference potential device) band is used by many devices for radio control. There is no need to apply for a class licence, and the availability of low-cost modules means it is easy to create a custom wireless link. As it is no longer so easy to get these modules locally, we have created our own version, which is a direct replacement for commercial equivalents. W e’ve published many projects that operate on the 433MHz LIPD band. Many include a transmitter and receiver module pair, or one or the other to interface with existing equipment. The existence of drop-in radio modules has made this quite easy. ACMA, the Australian Communications and Media Authority, is responsible for the regulation of this radio band. According to the Radiocommunications (Low Interference Potential Devices) Class Licence document, this band covers 433.05MHz to 434.79MHz, actually excluding 433MHz and being fairly well centred on 434MHz. Many devices (including this Transmitter) operate at a nominal 433.92MHz. The general provisions of the licence are that a person may operate a radio transmitter in one of the LIPD bands so long as the prescribed frequencies and equivalent isotropically radiated power (EIRP) restrictions are observed. In other words, a paid licence is not needed if the licence conditions are obeyed. For the spectrum between 433.05MHz and 434.79MHz, the EIRP is not to exceed 25mW. The full licence details can be found at siliconchip.au/ link/ac3z There is the proviso that a device “is generally not expected to suffer interference, but may under specific circumstances”. We’ve heard stories of garage or car remote controls that 72 Silicon Chip stopped working because of interference from another source, so it can happen. There are some other restrictions relating to areas near specific astronomical observatories and the like. The upshot is that as long as you can comply with some fairly simple restrictions, anyone can transmit on this band. Past projects Nearly all of our projects using 433/434MHz transmitters and receivers have used prebuilt modules. Typical transmitter modules include Jaycar’s ZW3100 and Altronics’ Z6900; the corresponding receiver modules are the Jaycar ZW3102 and Altronics Z6905A. We’ve written about these as far back as Part 6 of Stan Swan’s PICAXE series (July 2003; siliconchip.au/ Article/3908). That was about when these modules started to become available. We have noticed some local suppliers warning they will be discontinued soon, though. We recently started incorporating discrete transmitter circuits into some of our designs. The Remote Control Range Extender from January 2022 (siliconchip.au/Article/15182) and the Secure Remote Switch from December 2023 and January 2024 (siliconchip. au/Series/408) both did so. They incorporate custom circuitry built around an RF chip, the MICRF113 ASK Transmitter IC. ASK is short for ‘amplitude-shift keying’, where digital data is encoded as different amplitude levels of the carrier wave. To ensure all these projects could still be built, we decided to design this drop-in replacement for the tiny transmitter modules. Our Transmitter design has the same size, shape and pinout, so is interchangeable. Later we’ll look at some minor differences from the modules; they are mostly Fig.1: a typical application for these sorts of modules is to create a wireless link. The DATA signal is usually a bit stream with a specific type of serial encoding, rather than a plain HIGH/LOW digital level. The antenna is often a quarterwave monopole in the form of a piece of wire about 170mm long. Australia's electronics magazine siliconchip.com.au Fig.2: the Transmitter circuit mostly consists of the recommended RF capacitors and inductors for the MICRF113, along with a 13.56MHz crystal to set the carrier frequency. We added a voltage regulator and DATA input protection circuitry to allow operation with 3.3V or 5V systems. improvements made possible by the MICRF113 chip. Module overview We also thought that this was a good time to take a deeper look at these modules, their general operating parameters and what their limitations are. There are of course many cheap versions available from online sellers, and the available information can be limited. The transmitter module is our focus in this article, although we will look at how the receiver extracts the signal of interest from the busy radio spectrum. This information is presented at the end of the article, in case you are only interested in building the Transmitter. Fig.1 shows a typical arrangement used to provide a wireless link. It is as simple as supplying power to the transmitter module and providing a data signal. What goes in the DATA pin on the transmitter generally comes out at the DATA pin on the receiver, subject to the limitations of the modules, which we also discuss later. Transmission over distances of up to 100m are possible in open air. Usually, the data is an encoded packet reporting a status or a command. The packets are brief (perhaps 0.1s in duration) and are only transmitted occasionally, which helps to avoid interference with other equipment. Even if one packet interferes with another device, it’s unlikely the intervals between packets will match siliconchip.com.au exactly, so they will only interfere occasionally. The software at each end is often designed to encode an identity (possibly using DIP switches or the like), as well as the command or data. This is so that a similar nearby transmitter does not trigger a receiver unless both share the same identity value. Some codes also include a checksum to validate the integrity of the data. Circuit details Much of the circuitry needed, shown in Fig.2, is outlined in the data sheet for the MICRF113 chip. It should come as no surprise that it is also similar to the Discrete Transmitter for the Secure Remote Switch. As well as IC1, the MICRF113, there is 13.56MHz crystal X1, along with its requisite load capacitors, operating as a frequency reference. IC1 uses an internal ×32 PLL (phase-locked loop) to multiply that up to 433.92MHz. The capacitors and inductors on the PAOUT pin (which delivers the RF signal) form a signal matching network to limit the transmitted harmonics. The LINK resistor should normally be a 0W jumper, but a resistor can be used instead to reduce the output RF power and thus the necessary supply current. There are more details on this in the MICRF113 data sheet. A solder blob across its pads would also work! The MICRF113 has an operating supply voltage range of 1.8-3.6V; we Australia's electronics magazine have chosen a nominal 3.3V and added a voltage regulator to provide this. The current demand is about 2mA in the quiescent state and 12mA when transmitting, so a small regulator is all that is needed. We used the MCP1700, which can handle up to 6V at its input, as we expect in most cases it will be receiving 5V. Its dropout voltage at the expected load levels is less than 0.1V, so it won’t cause a problem if a lower supply voltage like 3.3V is used instead. The two 1μF capacitors are required by the regulator for stability, while the 100nF capacitor provides local supply bypassing for IC1. Finally, the incoming data signal (from a microcontroller or the like) comes through a series resistor with a shunt diode to limit the voltage at the ASK pin. Thus, our Transmitter is compatible with 5V and 3.3V supply and logic levels. Assembly The Transmitter has been designed to fit in the same compact footprint as the Jaycar ZW3100; the Altronics part is slightly smaller, and all have the same pinout. So the assembly will involve working with small SMD parts. The MICRF113 chip comes in the SOT-23-6 package, while most of the passives are M2012 (0805 imperial) at 2 × 1.2mm, with one slightly smaller (M1608/0603 or 1.6 × 0.8mm) part. You’ll need the standard SMD gear such as flux paste, tweezers and so forth. A magnifier and good light will be a great help with such small parts. Fume extraction is always recommended when working with flux paste, and you should have a solvent to clean up the flux residue, too. We’ve managed to fit all the parts on one side, so you can use some BluTack or similar to keep small the PCB in position on your workbench. Solder wicking braid will be handy to remove any solder bridges that might occur. We’ve managed to fit practically all the component designators on the PCB silkscreen, but it’s best to refer to the overlay diagram (Fig.3) and accompanying photo to confirm the location of the components. Start by spreading a thin layer of flux paste on all the PCB pads. IC1 has the smallest pin pitch, so place it first. Its pin 1 designator is very small, so you will probably need to examine April 2025  73 You can also check your assembly against the photo here (shown enlarged and at actual size on the left). We used a right-angled header, but you might like to use straight headers to allow the module to be mounted parallel to a PCB, like we did with the Secure Remote Switch. We’ve designed our Transmitter to be a drop-in replacement for the Jaycar ZW3100 shown in the centre and on the right (as well as Altronics’ equivalent). It is even the same size with the same pinout. Fig.3: the PCB is quite small, but we managed to fit most of the component designators on the PCB silkscreen. You can check the components on the overlay diagram here as you go. Fig.4: with the components fitted as shown here, you can probe for continuity at the marked pads, which will indicate whether the inductors have been soldered correctly. 74 Silicon Chip it under magnification to confirm the correct orientation. It must be aligned with the matching mark on the PCB silkscreen, which is near L1. If you have the PCB upright, with X1 at the top and the external connections at the bottom, the text on IC1 should be upright, too. Clean the tip of your iron and add some fresh solder. Tack one lead and confirm that the others are flat and within their pads. If needed, adjust the part position by using the iron to remelt the solder. Then solder the remaining leads. Use solder-wicking braid to remove any solder bridges as you go. It will be trickier as more components are added. Add a little more flux paste, press the braid against the bridge with the iron and then slowly draw both away when the excess solder is taken up. Next, fit the two SOT-23-3 parts. D1 is near IC1, while REG1 is near the external connector. Fortunately, they should both only fit one way. You can use the same idea; tack one lead, adjust and then solder the remaining leads. These parts have leads that are quite spread out, so they should not bridge easily. Next, fit the two inductors. These will be a bit fiddly, since their leads are only on the undersides. You will need to apply the iron to the PCB pad and add solder, allowing it to melt and flow onto the leads. The smaller 68nH part is L2, which sits between the 5pF Australia's electronics magazine and 12pF capacitors, while 470nH inductor L1 is between IC1 and REG1. While L2 is the smaller M1608/0603sized part, we have used M2012/0805 pads to make soldering it easier. All the remaining two-lead parts are in M2012/0805 packages. If you are unsure that the parts have been soldered correctly, you can check this with a multimeter set to continuity or resistance modes. The inductors have low DC resistance, so both should read well under 10W. With none of the surrounding parts fitted, they are safe to probe. For L2, probe the adjacent pads on the 12pF and 5pF capacitors, as shown in Fig.4. Using a nearby pad eliminates the chance of a false positive in the event that the component is connected to the solder in the joint but not the pad below. For L1, try the other end of the 12pF capacitor and the end of the 0W link next to L1. If you get a low resistance reading across each inductor then all is well. If not, try adding some flux paste to each joint and reflow the solder with your iron before checking again. The passives Fit the crystal, X1, next; it is unpolarised, as are the other remaining parts. Its leads are quite small, so you might need to use a similar soldering technique to the inductors. The larger PCB pads should make this easy, although you won’t be able to test for continuity in the same fashion. siliconchip.com.au Be careful not to mix up the capacitors. The two 18pF capacitors are adjacent to the crystal and then, on the left of the PCB, are the 12pF capacitor above L2 and the 5pF crystal below it. The two 1μF capacitors are near REG1, while the 100nF part is next to IC1. Move on to install the 4.7kW resistor in the bottom-right corner and the 0W link nearby. Finally, fit the header of your choice; we used a right-angle header to match the prebuilt transmitters. Use a solvent to thoroughly clean the flux from the PCB and allow it to dry. Inspect your soldering with a magnifier and confirm that all the components are soldered correctly with no bridges. Testing If you wish to proceed cautiously, you should power up the Transmitter from a current-limited supply. The Transmitter should draw around 3mA while idle or 15mA when transmitting. Something simple, like a 330W resistor in series with a 5V supply, would also be suitable. Add a jumper wire or similar between the GND and DATA pins to ensure that the Transmitter is initially in the idle state. Then apply power and measure voltage across the resistor; it should be no more than around 1V. The next step is to apply a waveform to the DATA pin and see that it is received correctly. You may have a project planned for the Transmitter, in which case you should simply connect it and try it out. Another simple test we tried can be done with a piezo transducer and a 433MHz receiver module. Wire up the receiver module so that the piezo is connected between its DATA output and GND, then connect 5V power. You might not need an antenna to test over short distances, such as on a workbench. The piezo should emit a sound like white noise or static; this indicates that it is picking up normal background RF noise. If you then power on the Transmitter and drive its DATA input high, the noise should cease as that signal overwhelms the background noise and saturates the receiver’s automatic gain control (AGC). Driving the DATA pin low should similarly cause the background noise to resume. Applying a 1kHz square wave to the DATA pin should cause siliconchip.com.au a high-pitched noise to be emitted from the piezo. Note that you should always make sure the DATA pin is driven, since it is a high-impedance input and could otherwise float to an unknown level. Conclusion You can read more about some of our comparative tests in the text below, but we have found our Transmitter to be just as good, if not better than, other similar transmitter modules it is a drop-in replacement for. It is capable of transmitting at 10mW, which could fall foul of the EIRP restrictions if used with a highly directional antenna. So we recommend sticking to simple antenna designs, such as a quarter-wave dipole, (~170mm for 433MHz) to ensure that you do not exceed the licence limits. If you must use a directional antenna, replace the 0W link with a resistor to reduce the output power (refer to the MICRF113 data sheet). A detailed analysis of 433MHz modules We performed some testing on these modules to ensure our replacement performed at least as well. We purchased a ZW3100 (transmitter) and ZW3102 (receiver) recently from Jaycar; these will be our test subjects, alongside our new Transmitter. The information on the Jaycar website indicates a maximum supply current of 10mA and a maximum output power of 3dBm (or 2mW) for the ZW3100. The suggested data rate is 300bps to 10kbps. While the 2mW might sound comfortably within the 25mW limit, the EIRP (equivalent isotropically radiated power) could be higher. It is calculated as though the maximum signal strength (which might only occur in one direction) was radiated in all directions. In fact, the EIRP can be no less than the actual power. For a theoretical isotropic (outputting the same power in all directions) antenna, the figures will be the same. A highly directional antenna will have higher EIRP since the radiation is concentrated. Fortunately, we nearly always use non-directional antennas with these modules. The typical gain of quarter-­ wave monopole antennas is no more than 3dBi, which is about a factor of 2, keeping the EIRP to around 4mW, well under the 25mW limit. The MICRF113 in the Transmitter specifies an output power up to 10dBm or 10mW. So it too is unlikely to fall foul of the restrictions with a quarter-­ wave monopole antenna. The current draw on the MICRF113 peaks at around 13mA during transmission, so it does draw slightly more current for a substantially higher output power. Receiver operation If you have ever monitored the output of the receiver modules while Parts List – 433MHz Transmitter Module 1 double-sided PCB coded 15103251 measuring 19 × 15mm 1 4-way pin header, straight or right-angled to suit application (CON1) 1 13.56MHz 5.0 × 3.2mm surface-mounting crystal (X1) [Abracon ABM3-13.560MHZ-B2-T] 1 470nH inductor, M2012/0805 size, >434MHz SRF (L1) [Coilcraft 0805HT-R47TJLB or Murata LQW21HNR47J00L] 1 68nH inductor, M1608/0603 size, >434MHz SRF (L2) [Bourns CW16080868NJ, Coilcraft 0603CS-68NXJLU or Murata LQW18AS68NJ00D] 1 4.7kW M2012/0805 size SMD resistor, ⅛W 1 0W M2012/0805 size SMD resistor or value to suit (see text) Semiconductors 1 MICRF113YM6 SOT-23 ASK transmitter IC, SOT-23-6 (IC1) 1 BAT54C/BAT54S/BAT54 200mA 25V schottky diode, SOT-23 (D1) 1 MCP1700-3302 3.3V LDO voltage regulator, SOT-23 (REG1) Capacitors (all SMD M2012/0805 size, 50V ceramic unless noted) 2 1μF 50V X7R 1 100nF 50V X7R 2 18pF C0G/NP0 (or to suit crystal) SC7430 Kit ($20 + postage): 1 12pF C0G/NP0 includes all the parts listed here 1 5pF C0G/NP0 Australia's electronics magazine April 2025  75 Fig.5: examples of various encodings that are used with RF (and IR) systems. The pulses are different to encode a 0 or 1 while maintaining a duty cycle near 50%. Manchester encoding is often decoded by looking for the rising or falling edge in the middle of the bit time, rather than the pulse length or spacing. Scope 1: in the absence of a strong, nearby transmitter, the receiver modules will deliver noise. When connected to a piezo transducer or similar, it sounds like white noise. Scope 2: the current draw (and thus output power) of the Jaycar ZW3100 shows an analog response to a triangle wave, suggesting it is capable of amplitude modulation (AM) to some extent. 76 Silicon Chip Australia's electronics magazine nothing is transmitting, you might have noticed the signal is just noise; there is no always low or high idle state. Scope 1 shows a scope grab of the receiver output when no transmission is occurring. This is due to the way that the receivers resolve signals with different strengths. During transmission, they must be able to deliver a valid signal, whether the transmitter is nearby or far away. In other words, they must be able to cope with receiving weak or strong signals and produce the same output. The operation of many infrared (IR) receivers is much the same too, although most IR receivers have extra circuitry to suppress the output noise during idle periods. Though the modules are described as using ASK modulation, it is typically the most simple form called OOK (on-off keying). With OOK, one of the levels is ‘carrier on’ and the other is ‘carrier off’. Here, the carrier refers to the underlying 434MHz signal. The raw data is also further encoded with the likes of pulse-length, pulsewidth or pulse-distance modulation (as also used in IR remote controls). Manchester encoding is another system that is also used in these scenarios. Fig.5 shows representative waveforms of some of these encodings. The simple on-off nature of the data means that the receiver only needs to recognise two signal levels, and this is done with the assistance of automatic gain control (AGC). This is much the same principle that ensures that nearby and far AM radio stations are received at the same volume. The AGC takes an average of the incoming signal strength, compares it to the instantaneous strength and adjusts the receiver gain to compensate. The AGC response time will also dictate a minimum data rate; if the receiver sees a carrier on state for too long, it will saturate and start producing noise. This is why the various encoding schemes have a duty cycle close to 50%. It means that the carrier on and carrier off levels are a similar distance from the average that the AGC sees, and both output levels are decoded correctly. Analog behaviour There are some reports of these modules being capable of transmitting and siliconchip.com.au receiving analog data, such as voice or audio, using AM (amplitude modulation). But it’s doesn’t appear to be possible with either the Jaycar receiver or our Transmitter. The Jaycar transmitter might be capable of AM transmission, so could be used for this purpose with an appropriate receiver. Scope grabs of our tests are shown in Scope 2 (Jaycar Transmitter) and Scope 3 (our Transmitter). In these, the blue trace is a triangle waveform from a signal generator, which was fed into the DATA input of the transmitter module. The red trace is the output of a nearby receiver, and the green trace is the voltage across a resistor in the transmitter’s positive supply; a crude current measuring shunt. Being in the positive supply, a lower voltage means more current being consumed by the transmitter. Assuming that the current reflects the strength of the RF transmission, we can get an idea of whether the modulation is AM or simple OOK. For Scope 2 (the ZW3100), we can see that the current does indeed follow the incoming signal level over a range, while Scope 3 (our Transmitter) shows a very digital response, with a hysteresis between 1.3V (falling) and 2.0V (rising). So we don’t think our module will be suitable for AM transmission. In both cases, you can see that the receiver has a very ‘digital’ response, so we don’t think it could be used for AM reception. There is what appears to be some hysteresis in the receiver output, but part of that may be delays in the receiver and its AGC response. Comparative tests To keep the tests between our module and the one from Jaycar fair, we set everything up on a breadboard so that we could swap between the two transmitters without changing anything else. We did not attach any external antenna to the modules. The breadboard strips provide a very short antenna that was sufficient for transmission over short ranges. We looked at aspects like signal delay between the transmitter and receiver and the response to transmitting at different data rates. For all these, we used our Arduino Uno to generate a pulse-width waveform of the type used in 433MHz applications. siliconchip.com.au Scope 3: our Transmitter has a more digital response, even showing hysteresis. This should provide a cleaner signal when used in digital applications, as is usually the case for these modules. Scope 4: the blue trace shows a transmitter input, while the red trace is a receiver output. The short delay between the two is around 20-40μs and differs slightly between the modules. Each cycle is nominally 720μs, giving a 1.4kbit/s data rate. We used an oscilloscope to compare the data coming from the Uno to the data received by the receiver module. Scope 4 shows the delay between the input to the transmitter and the output from the receiver. For both transmitters, the rising edge was propagated more quickly than the falling edge (by about 10μs). Also, our Transmitter showed quicker response times in general; around 20μs compared to 35μs for the ZW3100. This could be partially due to the higher output power of our Transmitter, but there may also be some difference in the way that the incoming signal is modulated. Our findings in Australia's electronics magazine the Analog behaviour section above are consistent with that. The next test involved speeding up the waveform until we started to see missed and distorted pulses. Our Transmitter did not miss a pulse until we reached cycle times under 95μs or around 10.5kbit/s, while the ZW3100 started missing pulses at around 105μs or 9.5kbit/s. Scope 5 shows the conditions we looked for. The transmitter input is delayed to align it better to the receiver’s output; you can see a few locations where the receiver waveform has stayed high when it should have gone low. SDR analysis We also examined the output of the April 2025  77 Scope 5: the green trace is the transmitter input delayed slightly to roughly align it with the red trace of the receiver output. You can see that at higher pulse rates than in Scope 4, pulses are missed and the asymmetry between rising and falling edges is more pronounced. transmitters using a software defined radio (SDR) receiver. The necessary hardware can be found in cheap USB TV receiver dongles. The free AirSpy SDR# program provides a way of receiving and viewing the spectrum of the RF signal. We used much the same hardware as mentioned in Jim Rowe’s Software Defined Radio article from May 2013 (siliconchip.au/Article/3778). Since we were performing simple comparative tests, we used the basic whip antenna included with the dongle. Screen 1 (our Transmitter) and Screen 2 (ZW3100) show the received spectra, with a displayed bandwidth covering the approximate range of the 433MHz LIPD band. The SDR is not a precision device, so the readings are simply relative to its full scale (0dB at the top of the spectra). The peak of both transmitters sits very close to 433.92MHz, as expected. Our Transmitter has a few spurs and it peaks at around -4dB, while the ZW3100 has a wider spread and more spurs. It peaks at around -8dB. That Screen 1: the spectrum of our Transmitter is neatly confined to the 433MHz LIPD band. 78 Silicon Chip is not surprising, given the expected output power given by the respective data sheets. Summary Our Transmitter works as a drop-in replacement for the ZW3100; it should work in all applications that require a digital transmitter on the 433MHz LIPD band. The ZW3100 does seem capable of amplitude modulation, so you might prefer if you want to experiment with audio or other analog transSC missions. Screen 2: the ZW3100 transmitter module has a wider spread, more spurs and lower peak power than our Transmitter. Australia's electronics magazine siliconchip.com.au PRINTED CIRCUIT BOARDS & CASE PIECES PRINTED CIRCUIT BOARD TO SUIT PROJECT BUCK/BOOST CHARGER ADAPTOR AUTO TRAIN CONTROLLER ↳ TRAIN CHUFF SOUND GENERATOR PIC16F18xxx BREAKOUT BOARD (DIP-VERSION) ↳ SOIC-VERSION AVR64DD32 BREAKOUT BOARD LC METER MK3 ↳ ADAPTOR BOARD DC TRANSIENT SUPPLY FILTER TINY LED ICICLE (WHITE) DUAL-CHANNEL BREADBOARD PSU ↳ DISPLAY BOARD DIGITAL BOOST REGULATOR ACTIVE MONITOR SPEAKERS POWER SUPPLY PICO W BACKPACK Q METER MAIN PCB ↳ FRONT PANEL (BLACK) NOUGHTS & CROSSES COMPUTER GAME BOARD ↳ COMPUTE BOARD ACTIVE MAINS SOFT STARTER ADVANCED SMD TEST TWEEZERS SET DIGITAL VOLUME CONTROL POT (SMD VERSION) ↳ THROUGH-HOLE VERSION MODEL RAILWAY TURNTABLE CONTROL PCB ↳ CONTACT PCB (GOLD-PLATED) WIDEBAND FUEL MIXTURE DISPLAY (BLUE) TEST BENCH SWISS ARMY KNIFE (BLUE) SILICON CHIRP CRICKET GPS DISCIPLINED OSCILLATOR SONGBIRD (RED, GREEN, PURPLE or YELLOW) DUAL RF AMPLIFIER (GREEN or BLUE) LOUDSPEAKER TESTING JIG BASIC RF SIGNAL GENERATOR (AD9834) ↳ FRONT PANEL V6295 VIBRATOR REPLACEMENT PCB SET DYNAMIC RFID / NFC TAG (SMALL, PURPLE) ↳ NFC TAG (LARGE, BLACK) RECIPROCAL FREQUENCY COUNTER MAIN PCB ↳ FRONT PANEL (BLACK) PI PICO-BASED THERMAL CAMERA MODEL RAILWAY UNCOUPLER MOSFET VIBRATOR REPLACEMENT ARDUINO ESR METER (STANDALONE VERSION) ↳ COMBINED VERSION WITH LC METER WATERING SYSTEM CONTROLLER CALIBRATED MEASUREMENT MICROPHONE (SMD) ↳ THROUGH-HOLE VERSION SALAD BOWL SPEAKER CROSSOVER PIC PROGRAMMING ADAPTOR REVISED 30V 2A BENCH SUPPLY MAIN PCB ↳ FRONT PANEL CONTROL PCB ↳ VOLTAGE INVERTER / DOUBLER 2M VHF CW/FM TEST GENERATOR TQFP-32 PROGRAMMING ADAPTOR ↳ TQFP-44 ↳ TQFP-48 ↳ TQFP-64 K-TYPE THERMOMETER / THERMOSTAT (SET; RED) MODEM / ROUTER WATCHDOG (BLUE) DISCRETE MICROAMP LED FLASHER MAGNETIC LEVITATION DEMONSTRATION MULTI-CHANNEL VOLUME CONTROL: VOLUME PCB ↳ CONTROL PCB ↳ OLED PCB SECURE REMOTE SWITCH RECEIVER ↳ TRANSMITTER (MODULE VERSION) ↳ TRANSMITTER (DISCRETE VERSION COIN CELL EMULATOR (BLACK) IDEAL BRIDGE RECTIFIER, 28mm SQUARE SPADE ↳ 21mm SQUARE PIN ↳ 5mm PITCH SIL ↳ MINI SOT-23 ↳ STANDALONE D2PAK SMD DATE OCT22 OCT22 OCT22 OCT22 OCT22 OCT22 NOV22 NOV22 NOV22 NOV22 DEC22 DEC22 DEC22 DEC22 JAN23 JAN23 JAN23 JAN23 JAN23 FEB23 FEB23 MAR23 MAR23 MAR23 MAR23 APR23 APR23 APR23 MAY23 MAY23 MAY23 JUN23 JUN23 JUN23 JUN23 JUL23 JUL23 JUL23 JUL23 JUL23 JUL23 JUL23 AUG23 AUG23 AUG23 AUG23 AUG23 SEP23 SEP23 SEP23 OCT22 SEP23 OCT23 OCT23 OCT23 OCT23 OCT23 NOV23 NOV23 NOV23 NOV23 DEC23 DEC23 DEC23 DEC23 DEC23 DEC23 DEC23 DEC23 DEC23 DEC23 DEC23 DEC23 PCB CODE 14108221 09109221 09109222 24110222 24110225 24110223 CSE220503C CSE200603 08108221 16111192 04112221 04112222 24110224 01112221 07101221 CSE220701 CSE220704 08111221 08111222 10110221 SC6658 01101231 01101232 09103231 09103232 05104231 04110221 08101231 04103231 08103231 CSE220602A 04106231 CSE221001 CSE220902B 18105231/2 06101231 06101232 CSE230101C CSE230102 04105231 09105231 18106231 04106181 04106182 15110231 01108231 01108232 01109231 24105231 04105223 04105222 04107222 06107231 24108231 24108232 24108233 24108234 04108231/2 10111231 SC6868 SC6866 01111221 01111222 01111223 10109231 10109232 10109233 18101231 18101241 18101242 18101243 18101244 18101245 Price $5.00 $2.50 $2.50 $2.50 $2.50 $2.50 $7.50 $2.50 $5.00 $2.50 $5.00 $5.00 $5.00 $10.00 $5.00 $5.00 $5.00 $12.50 $12.50 $10.00 $10.00 $2.50 $5.00 $5.00 $10.00 $10.00 $10.00 $5.00 $5.00 $4.00 $2.50 $12.50 $5.00 $5.00 $5.00 $1.50 $4.00 $5.00 $5.00 $5.00 $2.50 $2.50 $5.00 $7.50 $12.50 $2.50 $2.50 $10.00 $5.00 $10.00 $2.50 $2.50 $5.00 $5.00 $5.00 $5.00 $5.00 $10.00 $2.50 $2.50 $5.00 $5.00 $5.00 $3.00 $5.00 $2.50 $2.50 $5.00 $2.00 $2.00 $2.00 $1.00 $3.00 For a complete list, go to siliconchip.com.au/Shop/8 PRINTED CIRCUIT BOARD TO SUIT PROJECT ↳ STANDALONE TO-220 (70μm COPPER) RASPBERRY PI CLOCK RADIO MAIN PCB ↳ DISPLAY PCB KEYBOARD ADAPTOR (VGA PICOMITE) ↳ PS2X2PICO VERSION MICROPHONE PREAMPLIFIER ↳ EMBEDDED VERSION RAILWAY POINTS CONTROLLER TRANSMITTER ↳ RECEIVER LASER COMMUNICATOR TRANSMITTER ↳ RECEIVER PICO DIGITAL VIDEO TERMINAL ↳ FRONT PANEL FOR ALTRONICS H0190 (BLACK) ↳ FRONT PANEL FOR ALTRONICS H0191 (BLACK) WII NUNCHUK RGB LIGHT DRIVER (BLACK) ARDUINO FOR ARDUINIANS (PACK OF SIX PCBS) ↳ PROJECT 27 PCB SKILL TESTER 9000 PICO GAMER ESP32-CAM BACKPACK WIFI DDS FUNCTION GENERATOR 10MHz to 1MHz / 1Hz FREQUENCY DIVIDER (BLUE) FAN SPEED CONTROLLER MK2 ESR TEST TWEEZERS (SET OF FOUR, WHITE) DC SUPPLY PROTECTOR (ADJUSTABLE SMD) ↳ ADJUSTABLE THROUGH-HOLE ↳ FIXED THROUGH-HOLE USB-C SERIAL ADAPTOR (BLACK) AUTOMATIC LQ METER MAIN AUTOMATIC LQ METER FRONT PANEL (BLACK) 180-230V DC MOTOR SPEED CONTROLLER STYLOCLONE (CASE VERSION) ↳ STANDALONE VERSION DUAL MINI LED DICE (THROUGH-HOLE LEDs) ↳ SMD LEDs GUITAR PICKGUARD (FENDER JAZZ BASS) ↳ J&D T-STYLE BASS ↳ MUSIC MAN STINGRAY BASS ↳ FENDER TELECASTER COMPACT OLED CLOCK & TIMER USB MIXED-SIGNAL LOGIC ANALYSER (PicoMSA) DISCRETE IDEAL BRIDGE RECTIFIER (TH) ↳ SMD VERSION MICROMITE EXPLORE-40 (BLUE) PICO BACKPACK AUDIO BREAKOUT (with conns.) 8-CHANNEL LEARNING IR REMOTE (BLUE) 3D PRINTER FILAMENT DRYER DUAL-RAIL LOAD PROTECTOR VARIABLE SPEED DRIVE Mk2 (BLACK) FLEXIDICE (RED, PAIR OF PCBs) SURF SOUND SIMULATOR (BLUE) COMPACT HIFI HEADPHONE AMP (BLUE) CAPACITOR DISCHARGER PICO COMPUTER ↳ FRONT PANEL (BLACK) ↳ PWM AUDIO MODULE DIGITAL CAPACITANCE METER BATTERY MODEL RAILWAY TRANSMITTER ↳ THROUGH-HOLE (TH) RECEIVER ↳ SMD RECEIVER ↳ CHARGER 5MHZ 40A CURRENT PROBE (BLACK) USB PROGRAMMABLE FREQUENCY DIVIDER HIGH-BANDWIDTH DIFFERENTIAL PROBE NFC IR KEYFOB TRANSMITTER POWER LCR METER WAVEFORM GENERATOR PICO 2 AUDIO ANALYSER (BLACK) DATE DEC23 JAN24 JAN24 JAN24 JAN24 FEB24 FEB24 FEB24 FEB24 MAR24 MAR24 MAR24 MAR24 MAR24 MAR24 MAR24 MAR24 APR24 APR24 APR24 MAY24 MAY24 MAY24 JUN24 JUN24 JUN24 JUN24 JUN24 JUL24 JUL24 JUL24 AUG24 AUG24 AUG24 AUG24 SEP24 SEP24 SEP24 SEP24 SEP24 SEP24 SEP24 SEP24 OCT24 OCT24 OCT24 OCT24 OCT24 NOV24 NOV24 NOV24 DEC24 DEC24 DEC24 DEC24 DEC24 JAN25 JAN25 JAN25 JAN25 JAN25 JAN25 FEB25 FEB25 FEB25 MAR25 MAR25 MAR25 PCB CODE 18101246 19101241 19101242 07111231 07111232 01110231 01110232 09101241 09101242 16102241 16102242 07112231 07112232 07112233 16103241 SC6903 SC6904 08101241 08104241 07102241 04104241 04112231 10104241 SC6963 08106241 08106242 08106243 24106241 CSE240203A CSE240204A 11104241 23106241 23106242 08103241 08103242 23109241 23109242 23109243 23109244 19101231 04109241 18108241 18108242 07106241 07101222 15108241 28110241 18109241 11111241 08107241/2 01111241 01103241 9047-01 07112234 07112235 07112238 04111241 09110241 09110242 09110243 09110244 9049-01 04108241 9015-D 15109231 04103251 04104251 04107231 Price $5.00 $12.50 $7.50 $2.50 $2.50 $7.50 $7.50 $5.00 $2.50 $5.00 $2.50 $5.00 $2.50 $2.50 $20.00 $20.00 $7.50 $15.00 $10.00 $5.00 $10.00 $2.50 $5.00 $10.00 $2.50 $2.50 $2.50 $2.50 $5.00 $5.00 $15.00 $10.00 $12.50 $2.50 $2.50 $10.00 $10.00 $10.00 $5.00 $5.00 $7.50 $5.00 $2.50 $2.50 $2.50 $7.50 $7.50 $5.00 $15.00 $5.00 $10.00 $7.50 $5.00 $5.00 $2.50 $2.50 $5.00 $2.50 $2.50 $2.50 $2.50 $5.00 $5.00 $5.00 $2.50 $10.00 $5.00 $5.00 PICO/2/COMPUTER ↳ FRONT & REAR PANELS (BLACK) ROTATING LIGHT (BLACK) 433MHZ TRANSMITTER APR25 APR25 APR25 APR25 07104251 07104252/3 09101251 15103251 $5.00 $10.00 $2.50 $2.50 NEW PCBs We also sell the Silicon Chip PDFs on USB, RTV&H USB, Vintage Radio USB and more at siliconchip.com.au/Shop/3 CIRCUIT NOTEBOOK Interesting circuit ideas which we have checked but not built and tested. Contributions will be paid for at standard rates. All submissions should include full name, address & phone number. WiFi 8×8 RGB LED matrix display This circuit allows you to control an 8×8 RGB LED matrix display using WiFi and your smart device with various web pages and/or a gesture control module. It can be housed in a 3D-printed box. Its features include: • real-time clock display (via NTP) • text scrolling and animations • adjustable pixel colours, brightness, scroll speed, fade and random colours • 15 customisable web page buttons • five fixed web page buttons for Time, Design, Coding, Screen and Settings. • a low-power audio amplifier provides sounds for some functions. You can see a video of it in action at siliconchip.au/link/ac43 The heart of the circuit is an ESP32 WiFi microcomputer programmed with the Arduino IDE. I used a 30-pin ESP32 Dev Kit board, but almost any ESP32 module will work provided the correct board is selected in the Arduino IDE and the pin allocations are modified in the sketch to accommodate the different ESP32 board. The micro connects to your local WiFi network using the credentials you initially enter into the sketch. It then runs web and WebSocket servers to allow you to call up the web pages to control the display using your smart device. The web pages allow you to choose the display you want, such as Clock, Text, Letters, Designs(Icons) Animations and Games. Also included are preset buttons to access programming web pages such as Design, Coding, Screen and Settings. The web server serves the data using HTML, CSS and JavaScript code. The WebSocket server allows rapid bidirectional data transfer between the server and the client. This data is a text string made up of data segments with a ~ (tilde) used as a field separator. The data generally includes a command/data pair to indicate the request. For example, “brightness=50” sets the display brightness to 50%. The software includes arrays for both uppercase and lowercase alphanumeric characters. Users can also program up to 100 custom displays with any pixel set to any colour or brightness. Animations can be created using consecutive designs to give the illusion of motion. A simple coding language allocates a label and action to be specified for buttons on the main web page to display preferred designs or text. Designs, Names and Coding definitions are saved to the micro’s SPIFFS file system, while settings like WiFi credentials and screen colours are saved to EEPROM. The display comprises an 8×8 array of WS2812B RGB LED chips. As well as containing the LEDs, each chip includes digital control and data shaping circuits. Each LED has a data input and data output pin, so the LEDs can be series-connected and the data signal passed through each from the first input pin to the last output pin. The LEDs operate from 5V DC. The micro sends a 24-bit data stream 64 times from its D13 digital output to the display module to switch the pixels on or off and set their colours. By turning specific pixels on, alphanumeric letters can be shown, as well as colourful icons and scrolling text. The optional PAJ7620 gesture sensing module provides another input method, allowing the device to be controlled with your hands. It connects to the micro via an I2C interface using pin D22 for SCL and D21 for SDA. The PAM8302 low-power audio amplifier and speaker provide the ability to create a limited range of sounds. The micro drives the D12 pin with PWM signals to produce sounds. The unit is powered by a 5V 1A USB plugpack (CON1). Each LED can draw up to 50mA at full brightness. The software limits the maximum brightness so that the maximum current drawn with all LEDs on with white colour is around 0.5A. Diode D1 appears to be over-rated, but this keeps temperature rise modest at higher currents. If the 5V plugpack supply is not connected and the micro is powered from your PC via the micro’s USB port, D1 prevents 5V from being backfed to the display panel, avoiding the potential for computer USB port overloading. The user manual and Arduino source code can be downloaded from siliconchip.au/Shop/6/1824 Phillip Webb, Hope Valley, SA. ($120) Making dual triode valves interchangeable One of the most common valves still in use is the 9-pin dual triode in a “Noval B9a” bottle. There are very few guitar amplifier builders indeed who haven’t come across the 12AX7. Similar valves include the 12AU7, 12AT7, 12AY7 and ECC88, also used in preamps and power amplifiers. The pinouts of the 12A*7 family and the ECC88 family are identical except for the heaters. The 12A*7 family has two heaters: one between pins 4 and 9, the other between pins 5 and 9 (on the left in Fig.1 above). If these two heaters are connected in series, the heater voltage is 12.6V <at> 150mA current. If connected in parallel, the heater voltage is 6.3V <at> 300mA. The ECC88 family has only one heater, between pins 4 and 5; the voltage is always 6.3V, and the current is about 350mA. Pin 9 is instead connected to the shield that separates the two triodes, which is usually grounded, as shown on the right in Fig.1. The problem is that when you try to replace a valve of one family (for example, 12AT7) with one from the other family (say, E88CC), you have to rewire the heaters. That is a fiddly undertaking at the best of times; for some circuits, it is nigh impossible. The heater wires are often the first that get connected to the sockets. All other components are soldered above them. These components can be bulky and challenging to push aside to gain access to the heating wires. The simple ‘universal heater’ circuits shown in Fig.2 at lower right allow you to change between the different type of valves by simply swapping the envelopes. The universal heater circuit requires a DC heater supply of 6.7-7.0V. If you have a standard 6.3V AC heater supply, the conversion should not be too much of a challenge, especially if you use schottky diodes in the rectifier bridge. A 6.3V AC heater supply will give a peak voltage of about 8.9V before the rectifier bridge reduces it by 0.6-1.4V, depending on the type of diodes you use. Keep in mind that whilst they have a lower forward voltage drop than standard silicon rectifiers, schottky diodes are leaky and their leakiness increases with temperature. That’s why it’s a good idea to use schottky diodes rated siliconchip.com.au for 3A or more for a 1A heater supply. You will have to rig up a small auxiliary DC supply, which must be a few volts higher than the heater voltage. In the circuit shown on the left in Fig.2, the auxiliary voltage is 14V, with a draw of about 50mA for each valve socket. In the second example on the right, the auxiliary voltage can be just a volt or two above the heater and minimal current is drawn. The heater pins of the valve socket have to be connected as follows: • Pin 4: to the 0V of the heater supply • Pin 5: to a switch that can flip it to either 0V or to the positive side of the heater supply • Pin 9: to the positive side of the heater supply (via a diode, either standard or schottky) The universal heater works by sensing whether pin 9 is a part of the heater circuit or not. If the inserted valve is a 12A*7 type, there are two heater windings of about 38W each, one between pins 9 and 4, and the other one between pins 9 and 5. It senses that the voltage on pin 9 is around 6.3V as a result, and pulls pin 5 down to 0V. This causes the heater current to run from pin 9 to pins 4 and 5, in parallel. On the other hand, if the inserted valve is a member of the ECC88 family, it has just one heater, between pins 4 and 5. Pin 9 is not connected to them. D1 is reverse-biased and the voltage on pin 9 goes up to the value of the auxiliary DC supply. The circuit senses this voltage increase and pulls pin 5 up, so the voltage on it becomes 6.3V. This causes the heater current to run from pin 5 to pin 4. Australia's electronics magazine In most circuits that have valves from the ECC88 family, pin 9 is wired directly to ground or chassis, depending on the requirements. With this circuit, the grounding wire must be removed from pin 9 and an appropriately-­ sized capacitor to ground installed in its place. That means the pin is still grounded from an AC perspective. The circuits shown in Fig.2 do the same job, with an electromechanical approach on the left, and a transistorised approach on the right. For the one on the left, I used an NG8N1C0.64 12V signal relay, which draws about 50mA. The resulting PCB can easily fit onto the valve socket lugs, although the relay is a bit tall if you’re looking for a really unobtrusive addition. For the sake of simplicity, the relay flyback diode, the grounding capacitor on pin 9 and the two customary 47W grounding resistors between the heater positive and ground have all been omitted, but they should ideally be added to a practical implementation. In the circuit on the right, the pair of Mosfets act like an SPDT switch. The value of Re will depend on your auxiliary voltage. If D1 is a schottky, you’ll also have to take into account its leakage current and make sure that it will not cause the PNP transistor to conduct. A compact SMD version of the circuit can be made with a BC856 transistor, NX2301P and PMF63UN Mosfets. It will fit on a very small PCB that can easily fit between the valve socket lugs. Vedran Simunovic, Chatswood, NSW. ($100) April 2025  81 Part 2 by Phil Prosser POWER LCR METER We introduced this new device last month. It isn’t just another LC meter; it can deliver a range of currents up to 30A to determine how an inductor behaves as its core starts to saturate. This tester can also measure very high capacitances and very low resistances. This article covers its assembly, testing, calibration and use. T he Power LCR Meter has two basic modes: it either applies a fixed current or a fixed voltage to the device under test (DUT) and samples the voltage across it and current through it many times over a short period. It then examines those samples to determine either its resistance, capacitance or inductance. Because it can control the current used for the test, for power inductors, it can step through a range of currents and calculate the inductances, allowing you to see how it changes. For a typical inductor with a ferrite, iron or mu-metal core, the inductance will remain relatively steady until a certain current level is reached, then it will fall off as the core saturates. Having this information is invaluable as it allows you to determine whether the inductor will be suitable for applications that demand a certain inductance up to a certain current level, like a loudspeaker crossover or switch-mode power supply. Construction The Power LCR Meter is built on a double-sided 156 × 118mm PCB coded 82 Silicon Chip 04103251. It mostly uses through-hole parts, but there are a few SMDs, which should be fitted first. During assembly, refer to the component overlay diagrams, Figs.10 & 11, to see which parts go where. You can see in the photos that we didn’t have a 5W 0.39W resistor, so we used two smaller resistors in series. We only installed one 47,000μF capacitor on this prototype, which was enough for the test inductors used. Fit both if you want to test large, low-­resistance inductors. You will also see that we have used 1μF & 10μF SMD tantalum capacitors, while the final parts list suggests ceramic capacitors instead. You can use either, but the specific ceramic capacitors should be cheaper, more reliable and perform better. If you use tantalums, make sure you orientate them with the positive stripes as shown on the PCB and in the photos. We always like to fit all the power supply parts before the remaining active semiconductors to make testing easier. So start by mounting all the parts in the power supply section, which is everything to the left of the Australia's electronics magazine white vertical line on the silkscreen (the black line in Fig.10, including the parts in the lower-left corner). It’s easiest to start with low-profile components like resistors and then work your way up to the taller ones, ending with the bulky and heavy inductors. Watch the orientations of the diodes, electrolytic capacitors, regulators and transistor. For the regulators and transistor, pay attention to which side the metal tab goes (REG3 & REG5) or flat face (the others) so that they match Fig.10. There is space for a heatsink for the LM2576 (REG5), but it is not required. The average dissipation is low enough that it will be fine without it. With all the power supply components installed, you can connect a 12-20V DC power supply to CON4 (with the positive lead nearest the fuse) and check the following: 1. Check the 10V filtered rail is 9-11V; our four prototypes all measured about 9.8V. You can measure this on the DUT+ terminals. There is a GND test point just next to the power switch; we found it convenient to siliconchip.com.au 100nF Spare S4 NO S1 siliconchip.com.au (S4 SPARE) 4.7kW 100W 1kW 33nF 4.7kW 100W Q5 TIP121 Q1 0 BC558 100nF 470W 1W 1W CON11 TRIGGER 470W 100nF IC7 TLC072 100nF 47kW 47kW 4.7kW 470W S5 Power DUT− 4.7kW 100nF BAT85 S3 NO DUT+ 4.7kW + 4.7kW (S1 ENTER) 470W BAT85 100nF TP3 NC S2 NO 4.7kW 4.7kW BAT85 Fig.10: we recommend you fit the power supply components first (the whole leftmost section) so you can verify that is all working before adding the rest of the parts. Be very careful to orientate IC1 correctly, with its pin 1 dot at upper left, before soldering it. Also watch the orientations of the other ICs, diodes, electros, and transistors (including the Mosfets). Q9 BC548 Fig.11: there aren’t many parts on the back of the PCB; just the four or five switches. The main measurement terminals pass through the two large holes near the middle. Down NC NO BAT85 D9 Up Enter NC 4.7kW IPP013N04NF2S 4.7kW 4.7kW 4.7kW 4.7kW 4013B IC3 D7 IC8 INA281B1 1m F V1.2 SILICON CHIP Power LCR Tester NC (S2 UP) T P5 TP4 BAT85 + 47,000mF IC4 LM393 100nF D6 Q2 4.7kW 4.7kW BC548 IC6 INA281B1 CON1 + KELVIN SUP70101EL 12V ZD12 CON5 10W 10W + CON6 100nF − Q4 1m F 47,000mF IC2 MCP4822 Q7 Q6 4.7kW 1mF Q8 BC548 330W 1W D5 JP8 10kW 18pF ZD11 12V 56 0 W 33 0 W 1 0m F 1kW 220pF D8 REG3 100nF LM337 1 00 m F 10mF 4.7kW (S3 DOWN) 4.7kW 10kW 4.7kW 100nF 100nF 100nF 10mF 100nF 4.7kW Q3 BC548 10mF 18pF 100nF 8MHz D10 TP6 100nF IMON 2.7kW CON7 X1 470W 0.005W 0.39W 5W + TP8 +3.3VA CON3 + L2 330 m H 1 00 m F REG2 LM2950-33 + −3.3V 100W 1W 1 4148 1kW 100W 100nF + +3.3V REG1 LM2950-33 100nF GND 100mF 100nF RAIL IC1 SENSE 100nF 1 TP7 TP2 100nF VR1 20kW D3 4148 PIC3MK0128MCA048 10mF + + JP10 4.7kW Q1 BC558 33kW 100mF 10mF 1 00 m F 100nF IC5 25AA256 100nF JP9 4148 D2 10mF 16 1000mF D1 4148 10 m F +10V GND CON2 100nF D4 L1 330mH 4.7kW 4.7kW POWER SUPPLY 10 0 W 58 22 100nF REG5 LM2576 1000mF RS E CON4 POWER 1000mF (S5 POWER) F1 1A + + + DUT− DUT+ Australia's electronics magazine April 2025  83 solder a piece of tinned copper wire into this to clip onto. 2. Check the +3.3VD, +3.VA and -3.3V voltages. Test points for these are just above the circular cutouts for the DUT connectors. We expect the two positive rails to be within 100mV; note that in normal operation running from 12V, these regulators get quite warm. If any of these are off significantly, or something gets hot, check the orientation of all capacitors and diodes. We tried to keep all capacitors orientated the same way, but because switch-mode power supplies have exacting layout requirements, the diode placement in that area is not so consistent. The 330W resistor just above the 47,000μF capacitors is there to put a sufficient load on the switch-mode power supply that it runs continuously. We need this to generate the -3.3V supply. If your -3.3V supply does not come up properly, but everything else looks OK, check it. The following surface mount parts can go on next. With the power supplies behaving, it is time to get the fiddly bits on while there is still room. That includes: ● The PIC32MK0128MCA048 (IC1). ● The two 10μF surface-mount capacitors. ● The eight 100nF surface-mount bypass capacitors, which are mostly around IC1. ● The two 18pF SMD capacitors near the crystal oscillator. ● The three 1μF SMD capacitors, which are next to the INA281s and across the DUT terminals. ● The 25AA256-I/SN serial EEPROM. ● 470W series resistor for the crystal oscillator. ● The 10kW and 1kW resistors next to the reset header. ● The two 10W resistors for the Kelvin connection option. ● The two INA281B1 devices (IC6 & IC8). The INA281 devices are in SOT23-5 packages, which are a little on the small side. However, if you approach 13 15 – A them with some care, they are not too difficult to solder. The PIC microcontroller is in a 48-pin thin quad flat pack (TQFP), which has a 0.5mm lead spacing. This was the most easily soldered IC in the series we could find, alternative devices being in leadless packages, which are daunting to solder. We have provided soldering guides for TQFP and SOT-23 packages in the past. Our key tip is to use plenty of flux paste and to use a magnifying loupe to check for bridges between pins when you’ve finished. Use solder wick to remove any bridges you find. If the joint on a pin looks a little dry, resolder it before it causes you trouble later. When you’ve finished construction and apply power, if the LCD does not fire up immediately, come back and double check those pins for shorts. We have had to fix plenty of solder bridges ourselves in the past; the PIC microcontrollers are very tolerant of shorted pins and we have not managed to blow one up yet from a solder bridge (but it’s still better to clear them before applying power). Pro tip: after soldering all the SMDs, you will probably have flux residue that gets in the way of a proper inspection. Clean it off using a flux solvent (or isopropyl alcohol or methylated spirits if that’s all you have) and it will be much easier to spot any problems. Your board will also look a lot nicer and be less sticky! Mounting the LCD We want to connect the 16×2 LCD to the main PCB with a 16-way ribbon cable. To fit neatly in the case, we directly soldered the ribbon cable to the 14 through-holes on the LCD. This was a nuisance, but there was not room in the case for the IDC header we wanted to use. We say 14 and not 16 because the backlight connections are at the other end of the LCD. We show how we connected this in the photo below. Ensure that the red wire on your ribbon cable goes to pin 1 at both ends. Also make sure that once crimped, the IDC cable comes out in the right direction. The total length of ribbon cable we used was 300mm, with about 200mm between the IDC header and LCD board, leaving that extra length to connect to the backlight on the LCD board. Pins 1-14 of the ribbon cable are connected to the same pin number on the LCD. Note that the pins alternate between the two columns on the LCD. For the two remaining wires on pin 15 and 16 from the main board, strip the end of these and solder them to the anode and cathode backlight pads. Importantly, for the Altronics screen, you must place jumpers horizontally on JP9 and JP10 on the main board as shown in Fig.10. This applies 3.3V to Vdd (pin 2) on the LCD and grounds pin 1. If you are using a different display, check its data sheet, as these pins are sometimes swapped between manufacturers. If this is the case, you can install JP9 and JP10 vertically, which will swap the rails. Getting the microcontroller working At this point, we can install the remaining parts in the microcontroller section. That is the section at upperright bordered by a solid vertical line on the left and a broken horizontal line below. The four pushbutton switches mount to the rear of the PCB (S4 is not needed). For these, it is important that you rotate them so the normally open (NO) pins are at the bottom. Double-check this using a continuity meter; if on startup the system always goes into calibration mode, you almost certainly have the switches in the wrong way around. Also watch the orientations of the BAT85 protection diodes as they are not consistent. We also note that you can save quite a bit purchasing these from the larger online suppliers. We have used a lot of 4.7kW resistors to make it easier to purchase and manage the parts for this project. However, there are some 470W resistors as well, which will have similar colour 14 16 – K 1 2 This shows how to solder the ribbon cable to the Altronics 16×2 LCD. We tried to use an Altronics P5162A 14-way IDC-toPCB adaptor, but it wouldn’t fit in the space available. If you are installing it in a larger case, you may be able to use it. 84 Silicon Chip Australia's electronics magazine siliconchip.com.au codes, so take care not to mix them up. Mount all the 4.7kW resistors at once and you can be confident you won’t confuse them. Plug the LCD onto the main PCB, making sure that you get the pin 1 ends right at both the PCB and display end. We can now test this part of the board. Apply power and check the power rail voltages again. The voltages should be about the same; if any are very low, look for things getting hot or capacitors in the wrong way around. You should see the LCD backlight come on. If not, check the connections on the LCD from the header to the backlight LED and check that the headers are plugged in the right way around. You now need to adjust trimpot VR1, which controls the LCD contrast. Start at one end and turn it until you get good contrast on the display. There should be legible text if everything is fine, but if the LCD has not fully booted, you will still see lines of boxes or characters. If you get no display at all, double-check your LCD data sheet to make sure JP9 and JP10 are in the right locations. If the LCD is not displaying anything at all, check the soldering on the microcontroller and your cabling. If this all looks good, you probably want to check for activity on the LCD RS and E lines with an oscilloscope (if you have one). We put test points on the PCB for these – although we didn’t have to use them, as the 16×2 LCDs seem to mostly just work. If you still think nothing is happening and the display is blank, check the crystal oscillator drive on its associated 470W resistor. There must be an 8MHz sinewave here; if it is missing, double-check the microcontroller solder joints. You should now have a screen with text on it. Installing the measurement section You can solder all the remaining parts in place now. The only heatsink that you need to attach is on Q5, as shown in Fig.10. The other devices don’t dissipate enough power to warrant heatsinks, even though we have space for them on the board. With all parts mounted, you should be able to fire the meter up and get a screen saying “Resistance < 300R, Enter to Meas” and similar for siliconchip.com.au You can see how we wired the sockets to the PCB, all via polarised plugs or screw terminals. Capacitance, Inductance and Inductance Saturation. If you press the Enter/OK button, the meter will display “Measuring Resistance”, “Measuring Capacitance”, “Measuring Inductance” or “Measuring Inductance Sat’n” respectively and go off and measure the value. Note on our case we labelled “Enter” as “OK”. The standard measurements take about a second, while the inductor saturation tests need to perform quite a lot of measurements and take longer. Because we are dealing with inductors carrying a lot of current, we also need to provide a decent charge and decay time. So the inductance saturation test can take a few seconds, depending on the value of the inductor under test. Australia's electronics magazine The results are displayed on the screen and, once presented, you can press Enter/OK to repeat the measurement. If you want to change between resistance, capacitance and inductance measurements, press the up/down keys to cycle through the options, then press Enter/OK to measure. After a saturation current measurement is complete, you can cycle through the 10 inductance values across the range the meter can provide. The maximum current the meter will test to is 30A, plus readings from 5% to 90% of the maximum current. We have selected this range to ensure that noise at the start of the measurement does not grossly affect results (although it may still affect it if April 2025  85 the inductor rings badly). By pressing up and down, you can review: • The current at which the measurement is made. • The percentage of the inductance value of the second inductance measurement, which is considered 100%. We chose the second measurement, as this was always ‘clean’ in our tests. • The value of inductance at the displayed current. Calibration If you don’t calibrate the meter, it will load defaults, which will work but definitely compromise accuracy. To calibrate the meter, apply power and hold both the up and down buttons continuously. The meter will present the question “Calibrate meter?”, “Y/N, Up/Dn”. Press the up button, and a series of help screens will walk you through the process. As you will see in operation, inductance values are ‘all over the shop’ with current, so we have kept calibration focused on the few key parameters. We can calibrate critical parameters, but we do not seek to create a ‘lab standard’; this is more of a working measurement system for power devices where a few percent precision is sufficient. There are five steps to calibration: Fig.12: drill the holes in the lid as shown here. It’s best to start with pilot holes and then enlarge them to size. For the rectangular cutout, you could use something like a jigsaw, but you can also drill lots of small holes within the outline, knock the centre out, then file it to shape. It doesn’t have to be perfect as the bezel will cover minor imperfections. 86 Silicon Chip siliconchip.com.au #1: 10mA constant current test The current measurements in steps 1-3 are important for resistance and capacitance tests. Connect a milliammeter across the DUT terminals. The Meter will drive a 10mA current. Measure this and use the up/down buttons to enter your measured value. Get this to within 0.1mA of your meter reading. prototypes, the minimum measurable capacitance was around 20nF, and we achieved reasonable performance for values of 100nF and above. This is a power device tester, and does not seek to measure low-value capacitors. Once this is all done, it stores the new calibration factors in EEPROM, and you are ready to start testing! #2: 100mA constant current test This is the same as step 1 but at 100mA. We housed our tester in an Altronics H0310 ABS box. The board mounts on the lid, with onboard buttons and switches passing through holes in that lid. The specified switches all have the same height, so provided you make holes in the lid that all align with the switches, this provides a very neat mounting arrangement. We have always struggled with mounting 16×2 LCDs as they don’t generally come with a bezel. Therefore, we designed a bezel that can be 3D-printed to match the Altronics Z7018 LCD. You can download the STL file from siliconchip.au/ Shop/6/605 If you use a different LCD screen, you might want to design a similar bezel to match yours, as it makes assembly easier and neater. Fig.12 shows the front panel/lid cutouts and drilling details, while Fig.13 (overleaf) shows the drilling required for the side of the case. The Kelvin probe connectors mount on the side; we used banana sockets, allowing us #3: 1A constant current test The meter pulses the current on for two seconds, then off for about eight. This reduces heating in the constant current sink. Make sure your meter is not on a low-current range when you connect it. Adjust the value displayed until it is within 1mA of your meter’s reading. #4: Measure 3.3VA This voltage defines the full-scale value for the ADC and affects all measurements. Measure the voltage between ground and the 3.3VA rail at TP8. Enter this into the meter using the up and down buttons. #5: Null capacitance Leave the DUT terminals open circuit for this stage. This measures the internal minimum capacitance and uses it to correct low readings. In our Putting it in the case to use Kelvin probes when we want to measure really low resistances. You don’t need to use them for normal inductor and capacitor tests. We also installed BNC connectors so that we could use an oscilloscope to monitor the current waveform – see the photo below. These are optional. You do need to mount a power socket. This meter needs a minimum of 12V. We selected a socket that matched our power supply; there are many options. We chose a convenient spot on the side of the enclosure for this, as shown. The arrangement of holes and connectors on the side is what we recommend, but you can customise this to your needs. Ensure that all holes are centred in the lower half of the case so the connectors will not interfere with the PCB. Fit the LCD bezel to the LCD now. Test-fit it before gluing anything in place, as we have seen 16×2 LCDs in so many configurations. Make sure that yours will fit before committing to glue. If you use the Altronics screen and our 3D-printed bezel, it should be fine. The bezel is a tight fit, so expect to jiggle the display to get it on. If necessary, you can use a knife to scratch/trim the printed bezel. Those who have used a 3D printer will be used to this fettling process. Glue the bezel in place with a few drops of superglue on the inside of This shows how we arranged the connectors on the side of the case. You can also see our snazzy Dymo labels. At least we’ll be remember what everything does when we come back to it in six months! On this side, everything but the power socket is optional. Still, if you want to measure low resistances, the Kelvin connectors are required. siliconchip.com.au April 2025  87 the enclosure. Then install the LCD in the bezel and glue that in place after double-­ checking that you have the LCD the right way up. The DUT screw terminals affix to the front panel and project through two matching holes in the PCB. Mount them and do them up tight; we will wire them up later. Mount the four 10mm standoffs to the PCB using machine screw and shakeproof washers, then jiggle the PCB to get the pushbuttons through the holes in the front panel. Make sure the back of the LCD is clear of your PCB. The LCD ribbon cable comes out to the side of the PCB and will reach the header. The PCB mounts to the front panel as shown in the adjacent photo. Now you can install 80mm of 7.5A or 10A rated wire between the DUT+ and DUT– terminals on the PCB and the red and black screw terminals. Fig.13: this is how we arranged the connectors on the side of the case. You might decide to leave some of these out so verify which connectors you actually need before drilling the holes. The front and side panels are shown opposite. The front panel is shown at 40% actual size, while the side panel is at full size. You can download both of them from siliconchip.com.au/ Shop/11/1832 We soldered ours directly to the PCB to minimise resistance, but the board accepts 6.3mm spade terminals and you could crimp 6.3mm spade lugs to these wires. If doing that, make sure the connections are nice and tight. We need to make provision for Kelvin connections required for measuring low resistances accurately. These connect to the PCB via CON1. We simply ran two 150mm wires to banana sockets on the side of the case. For monitoring the operation via an oscilloscope, we recommend mounting two BNC sockets. One connects to CON11 and provides a trigger signal, while the other goes to CON7 for current monitoring. We used 150mm ribbon cable offcuts to wires these up. We put these on the side of our case next to the Kelvin connectors as we don’t use them much and that was where there is room. These oscilloscope connections are optional but present some interesting data. You need a digital ‘scope set to single-shot mode to capture the data. Set the trigger level to 1V. The vertical scale of the current curve is 100mV per amp. Most pulses are pretty short; for low-value inductors, they are in the 10s of microseconds. Large inductors can be tested over a few milliseconds. If you look at the waveforms presented last month, you will see that inductor current curves are almost never straight. Where there is a reasonably high DC resistance but no saturation, they curve downwards, while if the inductor saturates, they curve upwards. Usage hints Never use this meter to test components in circuit. The currents it drives may destroy something. Never connect this meter to powered circuitry. We have protection for inductor back-EMF, but if the input is driven above the 10V rail, you will damage the Meter. Always discharge capacitors before connecting them – if they hold a charge above 10V, you might damage the Meter. You don’t need to use the Kelvin connections for anything but low resistances. If you want to measure resistances below a few ohms, you really should use them. With these, you can measure right down into the milliohm region. SC 88 Silicon Chip Australia's electronics magazine siliconchip.com.au ENTER Inductance | Capacitance | Resistance DUT+ DUT− UP DOWN POWER Our assembled board; the one below uses two large capacitors, as recommended, but it can be used with one. The heatsink shown here on REG5 is not necessary. ENTER Inductance | Capacitance | Resistance OWER LCR METER This shows the PCB mounted to the inside of the case lid, with the LCD ribbon cable in place. siliconchip.com.au Australia's electronics magazine + − POWER KELVIN DUT− DOWN 12V DC DUT+ UP TRIG MONITOR POWER LCR METER The lid artwork & connector labels – see the Fig.13 caption for details. April 2025  89 SERVICEMAN’S LOG The camera eye Dave Thompson I’ve always loved cameras; the idea of capturing a precise moment in time has always been alluring. Back in the 1970s, as a wee tacker, I had one of those ‘kids’ camera that they sold by the millions. I got it for my birthday one year. I don’t recall which, but I know I was very young and had visions of becoming a photographer. It may well have been one of those “Daisy” branded cameras but I can’t remember exactly now. I know I used to have to buy flash cubes for it, and often I didn’t have them nor the money to buy them, so I made do with bright light when I could. It soon became obvious that I would not be the next darling of the photography circuit with this camera! Of course, one had to take the film (with only 12 exposures from memory – not even a full roll) to a local pharmacy or photo place and pay to get it developed. The problem was that most of the shots were blurred, or out of focus, or just rubbish because I was not a good photographer (the latter was the most likely reason). Still, in my defence, the gear was not the best (I know, an engineer blaming his tools rather than lack of talent!). The lenses in those mass-produced cameras were cheap and nasty, and that didn’t help at all! A few years later, I got a ‘proper’ camera: a single-lens-­ reflex (SLR) type that took 35mm film, which made a huge difference to my photos. For a start, I could actually see what I wanted to look at, and what came out on the film was almost what I was seeing through the viewfinder! No 90 Silicon Chip more parallax errors, foggy views or wasted film; unless you considered my resulting photographs wasted film! I’m sure the clever among you have realised my career goal of being a photographer didn’t develop (yes, I know, a bad pun). But I still have a love of cameras, and as early as the 1980s, I was asking myself why they didn’t take images on some kind of EPROM and digitally store them. Obviously, that was a lame and naïve attempt at thinking about the technology side of it; there was no way a technologically challenged individual like me could figure out how that would work out in the real world. But as we all now know, it is very similar to what they ended up doing. I knew I should have registered a patent! Anyway, despite not ending up as a professional photographer, I have still had a keen interest in cameras all my life. That old Yashica served me well for many years. I bought different lenses for it on my many trips overseas so I could try to take better photos. Digital man Then, all of a sudden, digital cameras were all the rage. And my old SLR with all those lenses was pretty much delegated to the scrap heap. I gave it all to a friend’s daughter who was going to take a photography course that involved Australia's electronics magazine Items Covered This Month • The camera eye • Zoom G2.1u guitar effects unit repair • Fixing the fan controller in a Honda Jazz Dave Thompson runs PC Anytime in Christchurch, NZ. Website: www.pcanytime.co.nz Email: dave<at>pcanytime.co.nz Cartoonist – Louis Decrevel Website: loueee.com developing and dark rooms and all that jazz. She seemed to appreciate it, so I guess it went to a good home. I went overseas in 1997, so at that time I bought a new-­ fangled (for the time) digital camera from a local big-box store before I went. This thing wasn’t cheap. From memory, it was about a grand, and it was a Casio-branded camera with a resolution of 800 × 600 pixels. It took very average photos, especially with me behind the lens. I still have the photos I took with it, and being able to just snap away and delete any dud ones as I went was a revelation for me. No more taking rolls of film to the local pharmacy and being disappointed with the out of focus or terribly composed shots – of course, I was charged for developing all of them. Now I only got to keep the photos I liked, no matter if they were low-res and not very well taken. It made that European trip that much easier. The problem is that it ate batteries like a kid eating cake at a party. And back then, batteries were not cheap. It took four AA cells and chewed through them like thermite through a paper plate. I was lucky to get 30 photos before the dreaded low battery warning. There was no built-in flash and there had to be so much light to get a decent photo it just about required one of those NightSun spotlights; the ones police choppers use to illuminate the scene. Still, it was a revelation, and I could see the writing on the wall that this was the future of photography. From that time on I was a digital camera fan, upgrading that old Casio (which I still have somewhere) to various new models, each more advanced than the last. Each one did me well and are still sitting in a drawer somewhere. Usually, it was an overseas trip that triggered my new camera searches and, while the (high) prices remained pretty much the same, the lenses and photo quality (due to better sensors with progressively higher resolutions) improved greatly. These were still point-and-shoot digital cameras, which suited my portability and lack of ability requirements. While digital SLRs were starting to appear on the shelves, they and their accessories were far out of my price range. Those cameras served me well and I still have most of the photos I took with them, stored on external hard drives or CDs and DVDs. Of course, none of those media will likely give up their data after all this time, with siliconchip.com.au burned CDs and DVDs and hard drives notoriously breaking down and failing. For a long time, I just copied them to new discs every year or so, keeping them refreshed, but my backup routine has fallen behind lately. I must get an external SSD and copy everything I can to it from those media. Moving pictures Then another revelation came along: digital camcorders. They were typically far smaller than their VHS, Super8 or even MiniDV cassette-based cameras. These were easily portable, had reasonable image quality and were not ridiculously expensive. One downside was that screen grabs from the videos were of course low-resolution, until I bought a camera that could take high-res snapshots, even while filming video. I just had to be adept enough to push the photo capture button while operating everything else! I wrote once about repairing one of my camcorders, which started faulting well out of warranty (April 2022 issue; siliconchip.au/Article/15283). The side screen would often not work properly; it was covered in lines and missing bits. Google suggested it was the interconnecting strap, one of those really thin, printed Mylar strap (flexible) PCBs with a push-in connector at either end. The replacement was available surprisingly cheaply from China, so I thought I’d give it a go. It was successful, but those things are really complicated and built so tightly, making them a challenge to work on with my fat fingers. So I hope I don’t have to do such a repair again. I’ve also repaired several SLRs over the years, though mainly mechanical faults from being dropped. Again, they are so compactly constructed, with parts just jammed into them everywhere. All these cameras really are a wonder of engineering and design. My fascination with cameras hasn’t stopped there. A while back, we lost a cat, and a friend offered us a supposedly working trail camera, one of those night-vision, motion triggered ones in full camouflage livery. These things too require many batteries for the days or weeks they might spend tied to a tree. I opened it up to put batteries in it but found that it was rotten inside. Someone had left cells in it and, of course, they leaked and corroded everything in the acid’s (or alkaline’s) Australia's electronics magazine April 2025  91 path. Some tracks on the PCB were almost eaten right through in places, and many of the surface-mount components were just fuzzy globs of corrosion. This thing obviously would not fly at all, and its days of snapping wildlife were over. I recovered the IR LED array from it and biffed the rest. There was nothing else for it but to buy another one. Chinese websites are awash with these cameras, but I bought an American-branded one from Amazon. Of course, it was made in China anyway, but it was of good quality and took very good, high-resolution photos, even in the monochrome night-vision mode. We set it up in areas we had credible reports of our cat being, but all it captured were birds and hedgehogs. We never found that cat. Available light So, the jungle camera sat on the shelf until we suspected someone was coming up our long drive – a brave act considering it is shingle and at night, every step must sound like someone opening one of those plastic cake containers you get from the supermarket. Whoever, or whatever, it was triggered our security lights, so there was definitely something, or someone, there. There had also been a spate of vehicle break-ins in the neighbourhood, so while I felt a bit paranoid about it, I set up the trail camera on one of our fence posts. It was relatively hidden from view, unless someone was scouring for one. At night, it would be almost invisible. For the following few days, I checked the camera, which was relatively easy as it has a small built-in colour screen. With a 64GB microSD card, it can fit a fair few images and videos before needing emptying. It is set with motion detection enabled by default, so no real setup was required, except for setting the resolution of the photos and videos, which would obviously impact the capacity of the card. Scrolling through the images, I could see the usual array of cats passing through, and the odd shuffling hedgehog, but neither of these types of critters usually trigger the security lights. So, there was nothing untoward the first night, albeit some good wildlife shots but not much else. The next night, however, the camera picked up some miscreant carefully coming up our drive to the gate. The Servicing Stories Wanted Do you have any good servicing stories that you would like to share in The Serviceman column in SILICON CHIP? If so, why not send those stories in to us? It doesn’t matter what the story is about as long as it’s in some way related to the electronics or electrical industries, to computers or even to cars and similar. We pay for all contributions published but please note that your material must be original. Send your contribution by email to: editor<at>siliconchip.com.au Please be sure to include your full name and address details. gate is just a rusty old wrought-iron thing I’ve been meaning to change for six years. I even have the replacement, a proper heavy duty motorised unit. But, after buying it, I found I’d need to lay a concrete track for it and because it crosses waste pipes, council consent. I filed it in the too-hard basket and simply put an alarm on the gate itself. This is a simple siren with a magnetically controlled trigger (the magnet goes on the other half of the split gate). If it isn’t present and the alarm is set, it sounds quite a piercing piezo siren. It also has a small keyfob style remote control, so I can arm and disarm it from the house, 30 meters away. So if anyone tried to open the gate, I would know. Jumping over it would require some athleticism and would likely set off the alarm, anyway. There was also an automatic security light, so they’d hopefully be scared off and not come back. The photos, taken a few seconds apart, showed him looking at the gate and alarm, so he obviously wanted to get in. Such is urban life. I called the police and told them, but they really couldn’t care less. If the crims broke in and hurt someone, the cops might be interested, but in burglary itself, not so much it seems. The images were good, especially with the light on him, but the intruder was not identifiable from these pictures. He was wearing dark clothes, gloves and a hoodie, the usual burglars’ fashion du jour. I guess a patrolling rottweiler would be the order of the day. Still, it was a warning that we were not immune and couldn’t let our guard down. I left the camera up for another month, but it didn’t pick up anyone else. Hopefully, the guy thought it was too hard. I only took it down because it failed to switch on one day, and a fresh set of batteries made no difference. I did open it up to have a look, but without any information on the circuit or the components, most of which had the numbers obfuscated, it was just more junk for the bin. It had lasted a reasonable time, but this throwaway consumerism is really not on. One little victory So, I did what any gadget guy would do and bought a better camera. This is on another level again. It is a fraction of the size of the trail cam, takes higher resolution photos, and is solar powered from an array that charges the 92 Silicon Chip Australia's electronics magazine siliconchip.com.au onboard batteries. And it only cost $40! There was a twopack for $65, but this would do me fine. Plus, it is controlled by an app over Bluetooth or WiFi and, using the app, I can swivel the bottom half of the camera, which contains one of the two lenses. One is fixed; one is controllable. It is also motion sensitive, with zones that can be included or excluded from the image, and provides a live feed of what is happening, with an alert to our phones. On top of that, it has night vision, and a microSD card stores the images as well! You can imagine my chagrin when I installed it and it didn’t work! I mean, 40 bucks is 40 bucks, right? I tore it open, as one does, and looked as to what could be the cause. When I say tore it open, I’m not kidding; this thing is sealed shut with clips. There are no screws. It is an outdoor camera, so one would expect seals and weatherproofing, and to be fair, it is very well made and has excellent rubberised hatches and sockets. Our camera will be installed under eaves, so it isn’t that critical; it won’t be out in all weathers, but preventing any condensation or moisture ingress is important. The batteries, two 18650 types, were hard-wired in. I measured them as best I could and typically, for cheaper gear, one appeared to be dead. I cut the links between them and the board and took them both out. One was indeed measuring just 0.3V; the other, 3.5V, a little flat but it should be OK. I hit the low one with my power supply to see if I could kick it into life, but even multiple tries resulted in only 0.7V. I was just wasting time here. I sent out for two spares; I’d replace them both, and they cost the same as the camera! At least they had solder terminals, and it was easy enough to solder them back in. This time, pressing the button resulted in some chimes and a voice saying power on. I measured the output of the little solar panel that came with it and confirmed a nice voltage was being generated, even inside my workshop. I stuck it all back together with sealant and it works brilliantly and is patrolling as we speak. Cameras are such fun! Zoom G2.1u Guitar Effects unit repair On a rainy afternoon, I was ‘bogged down’ trying to write some Arduino code. There were several household jobs I could have been asked to do if my wife discovered I was idle, so instead, I decided to browse Facebook Marketplace. I occasionally have a look to see if there are any bargains on offer. Normally, any genuine bargains don’t last very long. After scrolling down a couple of pages, I found a Zoom G2.1u guitar effects unit listed as non-working for sale for $5. I was interested in this because I have attempted to teach myself the electric guitar over many years. I usually only stick at it for a few months before some project demands all my spare time. As the seller was only ten minutes away from me, I thought it might be worth risking five dollars. I messaged the seller, and he said it was still for sale as a previous potential purchaser had not turned up. So it was mine if I came straight away. He was a stereotypical young muso and could not find the effects unit buried amongst all his musical equipment (junk). He even offered to pay me five dollars for my wasted trip. I said I was willing to wait a little longer while he dug a little deeper. He eventually found it, almost in plain sight, in a desk drawer. siliconchip.com.au Australia's electronics magazine April 2025  93 From top-to-bottom: • The Zoom G2.1u guitar effects unit. • A close-up of the XC9502BO92A DC/DC controller IC. I wasn’t initially sure whether the fault was due to this controller IC or the transistor in the photo below. • I used a TO-126 package BD140 transistor to replace the faulty SMD 2SB1706. 94 Silicon Chip The Zoom G2.1u was released in 2009 and has many inbuilt sound effects, called patches, some of which replicate the sounds produced by various guitar amplifiers. It could be powered by either four AA cells, a 9V plug pack or via the USB port. When I got home, I opened up the battery compartment and noticed that a couple of the battery terminals were very rusty. I immediately got my hopes up, thinking maybe the only fault was the rusty terminals that needed cleaning. No such luck. After cleaning, the unit was still dead. I removed the separate battery compartment and found the positive wire had corroded away from the rusty battery terminal. I still had my hopes up for an easy fix. I cut a piece of brass shim to cover the rusty battery terminal, soldered the positive wire to it and refitted the batteries to the case. This time, the effects unit showed some sign of life; that is, a single LED lit. Not having much of an idea of how the unit worked, I pressed various buttons and twisted knobs with no change in the display. It was time to download the 21-page user manual to learn how to operate the unit. Using information from the manual, I performed a factory reset, which briefly brought alive more of the LED digital display before it reverted back to the single LED. It was time to do a Google search for some repair ideas. There was a vague mention that a single LED lit indicated a flat battery, although the manual indicated that “bt” would be displayed. I tried powering the unit from both the plugpack and the USB port with the same result. After a break, I ran another Google search and eventually found a service manual. Still sticking to the power supply investigation, I found that the various input power sources are regulated down to 3.3V by a simple surface-mount three-terminal regulator. This checked out to be working satisfactorily. Further studying of the almost unreadable service manual circuit diagrams showed there was another more complex dual power supply. This indicated that there should be 1.26V somewhere on the main board loaded with surface mount components. The 1.26V looked to be produced by a surface-mount XC9502BO92A two-channel step up/down DC/DC controller IC driving a 2SB1706 PNP surface-mount transistor. There was no 1.26V output. At last I had something to work with! Either the controller IC or the PNP transistor could be at fault. I checked for sources of these components. The controller IC is listed as obsolete but could be obtained secondhand from the USA for $4.30 plus $44 postage. There was no way I was going to pay that on the chance it was the regulator that was faulty. The transistor was available as a bundle of five from interstate. I did subsequently find some more affordable quotes for the controller IC from China. However, the controller IC seemed to be providing a drive voltage to the transistor. Before outlaying any money, I decided that the best option would be to locate and try a substitute transistor for the 2SB1706. It was not too demanding in its specifications, being listed as a low-frequency amplifier with a collector current of 2A. I had a TO-126 package BD140 PNP transistor in my spare parts. Although not an SMD, I thought it would be a close enough electrical replacement for testing purposes. Australia's electronics magazine siliconchip.com.au Silicon Chip PDFs on USB EACH BLOCK OF ISSUES COSTS $100 NOVEMBER 1987 – DECEMBER 1994 JANUARY 1995 – DECEMBER 1999 JANUARY 2000 – DECEMBER 2004 JANUARY 2005 – DECEMBER 2009 JANUARY 2010 – DECEMBER 2014 JANUARY 2015 – DECEMBER 2019 OUR NEWEST BLOCK COSTS $150 JANUARY 2020 – DECEMBER 2024 OR PAY $650 FOR THEM ALL (+ POST) WWW.SILICONCHIP.COM. AU/SHOP/DIGITAL_PDFS I unplugged all the jumper leads, removed the main circuit board from the case and located where the controller IC and transistor were positioned. I carefully gripped the 2SB1706 surface-mount transistor with my finest needle point pliers and de-soldered it. I knew the transistor was de-soldered when the pliers clicked together and the transistor disappeared into a corner of the room. Oh well, I was committed to changing the transistor, faulty or not. I used fine tinned wire to join the surface-­ mount pads to the relatively fat legs of the BD140 and used double-sided tape to hold the BD140 down. I just powered this board up by itself and was pleased to see 1.26V appear at the correct location. I reassembled and switched on the effects unit and saw the led display go through it self-check routine on startup. Now to test it properly. I dug out my guitar and plugged it in to the input, and the headphones into the dual headphone/amplifier output socket. On playing a note, I was greeted by the most horrible (to my ears) distorted sound coming from the headphones. I thought I would now have an audio problem to fix. This unit comes with a host of inbuilt sound effects that you can create yourself. A couple of pushbuttons select the various patches. I tried selecting a couple of patches and found different levels of distortion. I finally stumbled on some clean guitar sound patches. I had apparently started testing with patches that were deliberately distorted guitar effects. I grew up in my teenage years during the 1960s listening siliconchip.com.au to instrumental bands such as The Shadows. My guitar listening tastes have not matured much since those days. The later effects pedals that duplicate sounds from those days are reverb and delay, which mainly embellish the original guitar sound and don’t distort it so much. Anyway, I have a lot of learning to do to realise the potential of this effects unit bought at a bargain price and repaired at zero cost. If I decide later to sell the unit, I will replace the BD140 transistor with the correct surface-mount transistor to maintain its originality. B. B., Arana Hills, Qld. 2009 Honda Jazz fan controller repair My daughter rang and asked if I could have a look at her Honda Jazz, as the heater/air conditioner blower was only working at maximum speed. I have seen this sort of fault before. The fan speed is usually reduced by series resistors, mounted so the air from the fan flows over them to keep them cool. Newer cars use more efficient pulse-width modulation (PWM) to control the fan speed. I googled the problem before she arrived and found that there is a module that’s held into the fan shroud by two screws. It was deep inside the passenger foot well and difficult to get to. The first step was to empty the glove box and release the two arms that support the lid in the open position. That allowed the glove box lid folded down all the way, so a long-shaft Phillips screwdriver could be used to remove the top screw that holds the resistor module in place. The next step was to remove the under-dash shroud and lie on the floor to access the bottom screw using a very short Phillips head screwdriver. This was achieved after some uncomfortable contortions. The module came out easily and I unplugged the wiring connector. It has a perforated metal cage covering the resistors. I was able to bend back some tabs to remove the cover. The resistors are a series of wire coils of differing size and gauge. I could see the fault straight away. There is a springy metal arm that is soldered to a metal tab; if the module gets too hot, the solder melts and the springy metal arm loses contact, isolating the circuit. The solder had aged and cracked, letting the metal arm move away from the tab and open the circuit. The fix was to solder the arm back to the tab. I refitted the metal cover and installed the module back in the car. The fan now worked on all four speeds, so my daughter can now use the air conditioner without getting her hair blown all over the place. SC J. W., Hillarys, WA. Australia's electronics magazine This module (shown with the shroud removed) from a Honda Jazz is used to control the fan speed. The arrow shows the location of the failed solder joint. April 2025  95 Vintage Radio The Astor APK 4-Valve Superhet Radio Astor released the APK in 1958. It was available in ten two-tone cabinet colour combinations: ivory, cherry red & white, grey & white, coral & white, blossom pink & white, dark green & ivory, lime & white, tan & white, China red & white, and turquoise & white. T he APK is first mentioned in Mingay’s Price Guide for Autumn 1958. This set was purchased through the Historical Radio Society’s Victorian auction. It is similar to two other Electronic Industries sets at the time, the Astor Mickey HNQ and the Peter Pan FNQ. The Astor Technical Bulletin for the APK, dated 18/4/58, contains the circuit and alignment procedures. The valve line up is 6BE6, 6AD8, 6AQ5 & 6X4; it has permeability tuning with fixed capacitors and variable inductors, similar to most car radios before frequency synthesis. The tuning knob shaft has three brass bands around a metal cylinder. As the shaft rotates one (or two, in the other direction) of the bands push or pull a plastic sled above the chassis. The sled has ferrite cores attached to it and, as it moves, they move inside the antenna and oscillator coils to tune the radio. The technical note cautions against adjusting the cores, while providing information on how to set them if required. While there are only four valves, 96 Silicon Chip the circuit is more complex than usual since it has a reflexed intermediate frequency (IF) valve that also acts as the first audio amplification stage. This has the advantage that it provides higher output than a set with the same valve count but no reflexing, and the reduced valve count makes it less expensive and leads to less heat generated in the cabinet, which is important for plastic cabinet sets. However, there are disadvantages to reflexing: • Additional passive components • An increased tendency to overload on strong signals • A more complicated design (not as much of a concern for large production runs) • Increased distortion at high modulation levels • Play-through/minimum volume effect Play-through is the presence of an audio output with the volume control set to zero. It is caused by the rectification of the IF signal from the slight curvature of the anode characteristic (‘anode bend detection’) and amplification in the same valve. Australia's electronics magazine By Jim Greig When the volume control is set slightly above zero, the normal and the out-of-phase play-through signals roughly cancel. The audio is generally badly distorted at this point, as explained in the Radiotron Designers Handbook, pages 1140-1143. Bias to the reflexed stage is a careful balancing act between minimising play-through and preventing audio signals in excess of the bias voltage from drawing grid current. It is set to -1.8V in this set, a very linear part of the anode curve. To assist in maintaining the constant bias, automatic gain control (AGC) is applied to the converter only (see Fig.1). The converter stage employs an unusual oscillator configuration; the coil has no taps or secondary winding, and it and the capacitor are in series. The cathode is grounded and pins 6 (anode) and 1 (grid) form a triode with the capacitor from cathode to grid and the inductor from grid to anode. At resonance, the series impedance is at a minimum, and the signals across the capacitor and inductor are 180° out of phase. The triode anode has a 180° siliconchip.com.au Fig.1: Astor’s circuit for the set. It has a large number of components around the 6AD8 because it’s reflexed, handling both IF and audio amplification. Voltages on this circuit were measured with a 1000W/volt voltmeter. phase shift from the grid, so there is positive feedback, and the valve oscillates at that frequency. The IF amplifier is straightforward, but it has a very low anode voltage from the voltage drop across the 51kW anode load resistor (#28). Detected audio is filtered by capacitor 14 and applied to the volume control, while also supplying the AGC voltage to the converter. From the volume control, the audio is further filtered by capacitor 12 to remove all of the 455kHz IF signal and only pass audio, which is directed to the grid of the IF reflexed amplifier through the secondary of the first IF transformer (#46). The audio amplification stage provides reasonable audio gain, around 30 times, measured by applying a 1kHz sinewave to the volume control wiper and monitoring the grid and anode voltages. At the operating point, the anode current is 2.5mA, and the mutual conductance (gm) is around 1mA/1.25V or 800µmho, with the load resistance (Rl, 51kW) in parallel with the 470kW siliconchip.com.au grid resistance on the 6AQ5, giving an overall load of 46kW. The quick formula for a high internal resistance valve, gain = gm × Rl, gives a gain of 37 times, but it measured as 30 times. A more accurate formula for gain includes the valve internal impedance. The curve of anode voltage vs current for constant grid 1 (and 2) voltages for a pentode is very flat (see Fig.3). The (variational) anode resistance is “the incremental change in anode voltage divided by the incremental change in anode current which it produces, the other voltages remaining constant”, per the Radiotron Designers Handbook (page 14). To calculate Ra, the current was Table 1 – Anode current vs voltage Anode voltage (Va) Current (Ia) 60V 2.502mA 65V 2.510mA 70V 2.514mA 75V 2.518mA 80V 2.524mA Australia's electronics magazine measured at anode voltages around the nominal 71V, with the actual operating voltages on Grids 1 (-1.74V) and 2 (43V) in-circuit – see Table 1. I a increased by 8µA while V a increased from 65V to 75V. Ra is therefore 10V/8µA or 1.25MW. Gain = gm × Rl × Ra ÷ (Rl + Ra), so the calculated gain is 35, still higher than what I measured. Could the input impedance of the 6AQ5 be reduced by the negative feedback from the 100pF capacitor shown in Fig.4? It seems unlikely at 1kHz, but I checked that by adding a 68kW resistor in series with the 2.2nF (0.022μF) coupling capacitor and measuring the AC voltages around it. The 6AQ5 AC input impedance calculated from the voltages measured is 141kW. RLl is then 37kW (51kW || 141kW). The gain is now calculated to be 29.5, which is close enough to the measurement. Using the same test method and removing the 100pF capacitor increased the input impedance to 330kW, so the capacitor has a definite effect. April 2025  97 The audio gain from this stage allows the use of overall negative audio feedback (to the bottom of the volume control), reducing distortion and effectively increasing the audio bandwidth. The circuit shows a resistor (#22) and capacitor (#16) connected to the diode on pin 8 of the 6AD8. After some time looking for an electrical reason for the diode and finding none, it seems likely that the pin is used as a convenient tag for the connections, and saves adding a ground wire to it. The link has the effect of slightly increasing the bias on the 6AD8 for strong signals. It varied from -1.78V to -1.97V, possibly to allow for a greater voltage swing. When operating normally, there is around 8V (0.16mA through 51kW) deviation of the 6AD8 anode voltage from the nominal 71V; the valve is operating comfortably on the linear part of the transconductance curve (see Fig.2). Audio is coupled directly to the 6AQ5 output valve, which operates with -8V of fixed bias. The relatively low anode voltage (185V) reduces the heat dissipated (again, important in a plastic cabinet) and lowers the power transformer requirements. Values from the RCA Receiving Tube handbook (Frank’s electron tube data sheets, RCA 6AQ5A) show comparative anode dissipations of 11.2W (250V × 45mA) and 5.2W (180V × 29mA), with the audio power output reduced from 4.5W to 2W, which is still quite sufficient for the set’s intended use. Fig.2: 6AD8 valve mutual conductance plots from Frank’s Electron Tube Pages (black) and my measurement (red). Restoration The chassis is mounted diagonally The under-chassis view with major components labelled. C6 22nF C16 50pF R28 51kW 47kW 47k W screen resistors C15 100pF 98 Tuning shaft 3 brass bands Australia's electronics magazine siliconchip.com.au Fig.3: the 6AD8 pentode’s anode characteristics (measured) for varying control grid voltages. Fig.4: some voltage measurements I made to help determine the 6AD8’s gain was as expected, or low. in the cabinet, so the tuning shaft connects directly to the large centred dial, and the volume control is on the lower left. The cabinet was in good condition; a wash with soapy water and a little polishing had it looking in a reasonable state for its age. The chassis was clean with no rust and a small amount of accumulated dust. Note that the speaker is held onto the front panel with metal tags on plastic posts. It is hard to remove them and keep the posts intact. Careful work expanding the jaws of the clips before removing them cut the breakages to one in four. I regarded all paper and electrolytic capacitors in the set as potentially bad, so I replaced them. Work had been carried out on the radio at some point; the first filter capacitor (#18) was a newer 47µF type, not the 24µF specified; I replaced it with 22µF, which is closer to the original value. I replaced the other filter capacitor (#17) with a 16μF electrolytic that I placed inside the original can. I also replaced the 100pF mica capacitor (#15) on the output anode, as it is subject to high voltages, and there is a history of mica capacitors in this position breaking down. Any faults on powering on would not be from these components, and hopefully not from a wiring error while replacing them. The original power cable was a twincore cord knotted behind the plastic back, so I replaced it with a threecore cable, with the Earth connected to the chassis and the cable properly restrained. I carefully enlarged the small hole in the cabinet’s rear to fit the new the cord. The power transformer sits on the chassis, and the mains and HT lugs are exposed and uninsulated; a clear safety hazard. Now that the chassis was Earthed, there would be 230V AC from the mains Active to the chassis, around 440V AC from the out-ofphase HT secondary and 380V across the secondary. The back of the volume control also has exposed mains wiring. Beware if you are working on one of these radios; cover the exposed terminals before powering it on! Having powered the radio on, there was no smoke but its performance was poor. A check of the DC voltages showed some anomalies. I measured 146V on the 6AD8 anode, not 71V. Its screen was at 28V. The converter screen supply was also low. So I powered it off and checked the resistors. The ½-watt resistors were within tolerance, but the 1W types This top view of the chassis shows the permeability tuning system, which is attached to a plastic sled. 2nd IF Transformer 1st IF Transformer 6AD8 6AQ5 6BE6 6X4 siliconchip.com.au Australia's electronics magazine April 2025  99 Scope 1: the converter oscillator grid voltage (red) and its anode voltage (yellow). Scope 2: the 6AD8 reflexed audio amplifier’s grid (red) and anode (yellow). You can see some of the IF signal superimposed on the red trace. Scope 3: the small variation with signal in the 6AD8 anode voltage. that were connected to the B+ were all high in value: the 6AD8 anode resistor (#28) was 60kW instead of 51kW, the 6AD8 screen resistor (#29) was over 100kW rather than 47kW, and the converter screen resistor (#30) was 60kW instead of 47kW. I replaced them all and then the 6AD8 screen measured 44V but the anode was still over 100V. A new 6AD8 bought it back to 70V. The radio could now tune a Melbourne station, and a tweak of the IFs bought it in reasonably well. Many of the Vintage Radio articles include information on receiver sensitivity. I have no experience making these tests and no screened room. However, I built a dummy antenna based on Graeme Dennes’ in Radio Waves, October 2020 and set up my signal generator, oscilloscope and voltmeter. I measured the audio power across the speaker; ideally, a resistor would be used instead. To obtain the standard 50mW of output, I needed 1.5mV of modulated RF. Assuming the dummy antenna to be part of the receiver per the 1995 British Standard, the aerial voltage to achieve the standard output was 1500μV, which is way too high. I replaced the 6AQ5, which made no difference, but a new 6BE6 converter dropped the required signal level to 500μV. No doubt this is still too high, but the AGC level increased from -0.51V to -1.77V as the signal strength was increased from zero; this change would have reduced the sensitivity. The result shows a lack of knowledge of the testing process rather than the absolute performance of the receiver, but it did help to diagnose a weak converter valve. In Bendigo, this set receives 774 Melbourne with some noise. A Panasonic R-399 12-transistor set with an RF stage performed better, but there is still noise; clearly, my location is not ideal for receiving that station. Overall, this set is typical of the era, in an attractive shape and available in numerous colours. It is well-made and achieves quite good performance with SC a reasonable price tag. An advertisement from The Biz (Fairfield, NSW), 24th of September 1958, page 18. References ● Frank’s Electron Tube Pages (https://tubedata.wernull.com/ index.html) ● The Valve Museum (www.rtype.org) ● Radiotron Designers Handbook, F. Langford-Smith, Fourth Edition 1963, Wireless Press ● Vintage Radio March 2019 (The Astor HNQ Mickey; siliconchip.au/ Article/11451) ● Astor Technical Bulletin Mantle Model-APK (www.kevinchant. com) ● HRSA Radio Waves, October 2020, Ferrite Rod or loop Antenna-­ equipped Receiver Testing ● Advertisements from the National Libraries Trove (https:// trove.nla.gov.au/newspaper/) I added insulation around the power transformer terminals after Earthing the chassis (and thus the transformer frame). 100 Silicon Chip Australia's electronics magazine siliconchip.com.au ASK SILICON CHIP Got a technical problem? Can’t understand a piece of jargon or some technical principle? Drop us a line and we’ll answer your question. Send your email to silicon<at>siliconchip.com.au No sound from Pico Gamer I built the Pico Gamer (April 2024; siliconchip.au/Article/16207), and was most impressed, as was my eldest granddaughter, who loves it. However, in the last few weeks, a problem has developed. There is no longer any sound. The games all play perfectly, but there are no sound effects or music. Even the beeps as the menu is scrolled no longer work. It was working fine, but this has suddenly happened. Any ideas? (G. G., Figtree, NSW) ● First, check that the volume control is turned up. If it is, you have a hardware fault, possibly a failed solder joint or component. Check all solder joints using a magnifier and repair any suspect ones, particularly at the Pico module and volume control pot. Then check all the components in the audio path, especially the loudspeaker. Determining the ratings of a transformer Is there a formula to work out the secondary output current of a transformer based on the diameter of the winding wire? I found a rather large transformer in my shed when I was sorting things out. I could use it for a future project if I know how many amps it can put out. I was not able to find any information online. (B. P., Dundathu, Qld) ● Typically, a transformer will use wire with a cross-sectional area of one square millimetre for every 5A. So if you can measure the wire diameter d in millimetres (ignoring the thickness of the insulation), you can calculate the theoretical current capacity as I = d2 × 5A. However, that is only a rule of thumb; cheap transformers will use thinner wire than that (ie, they will use a figure higher than 5A/mm2). This also ignores the core, which may saturate before you reach the current that the wire itself is capable of handling. The only way to find out for sure is to test the transformer under load. You will need to check that the temperature of the core and wiring remains within suitable limits at the calculated current. Also check that the voltage doesn’t sag excessively compared to the no-load voltage. It should usually drop by no more than 20% at the full rated current compared to no load. Building the Hummingbird amplifier I’m building the Hummingbird Amplifier from your December 2021 issue (siliconchip.au/Article/15126) and I have a few questions, starting with the power supply. I am planning to order the following from Altronics: one M5325C 25V + 25V 160VA toroidal transformer, one Z0091A 35A 1000V bridge rectifier and four R5208 2200µF 63V electrolytic capacitors. Would they make a suitable power supply for three Hummingbird amplifier modules to power two bookshelf speakers and a subwoofer? I’m just in the process of winding the 10µH inductor on the Hummingbird amplifier; I’ve never wound an inductor before. I read the instructions but I want to clarify them. I have a metal hole punch that is 10mm in diameter; I’m going to wind the coils on it with masking tape underneath them. It sounds like I do nine turns against each other horizontally as a flat round layer, then glue it. Then I wind another eight turns on top of that layer, glue it, then another nine turns in a third layer and glue it, giving 26 turns in total. Is that right? My final question is about CON3 and the speaker wiring. When I wire in the speaker, I connect the left-hand side positive terminal of CON3 to one side of the speaker terminal. Should the other speaker terminal connect to the other side of CON3, or is that wire omitted? I’m a little confused. (E. M., Hawthorn, Vic) ● The parts you have suggested for the power supply are mostly OK. It depends on how hard you’re going to drive the system, but 160VA should be enough unless you’re going to be WARNING! Silicon Chip magazine regularly describes projects which employ a mains power supply or produce high voltage. All such projects should be considered dangerous or even lethal if not used safely. Readers are warned that high voltage wiring should be carried out according to the instructions in the articles. When working on these projects use extreme care to ensure that you do not accidentally come into contact with mains AC voltages or high voltage DC. If you are not confident about working with projects employing mains voltages or other high voltages, you are advised not to attempt work on them. Silicon Chip Publications Pty Ltd disclaims any liability for damages should anyone be killed or injured while working on a project or circuit described in any issue of Silicon Chip magazine. Devices or circuits described in Silicon Chip may be covered by patents. Silicon Chip disclaims any liability for the infringement of such patents by the manufacturing or selling of any such equipment. Silicon Chip also disclaims any liability for projects which are used in such a way as to infringe relevant government regulations and by-laws. Advertisers are warned that they are responsible for the content of all advertisements and that they must conform to the Competition & Consumer Act 2010 or as subsequently amended and to any governmental regulations which are applicable. siliconchip.com.au Australia's electronics magazine April 2025  101 maxing out the subwoofer power. If you can afford the cost, weight and size of a 300VA transformer that’d be even better, but for regular program material at normal volume levels, 160VA is fine. You can get more capacitance for a similar price and in the same volume if you use 50V rated caps rather than 63V. We suggest using four of Altronics R4926 instead of R5208. They’re the same size and only $1 more each, with more than double the capacitance and a higher ripple current rating. The 35A bridge is more than sufficient. What you propose for winding the inductor will work fine. The inductor is there to ensure stability when a complex load is connected. That might be super long speaker wires or a crossover that presents a highly capacitive or inductive load. The exact inductance of that coil is not critical, but it is definitely needed. So, when winding the coil, aim for 26 turns of 10mm diameter. If the coil is a touch longer or has a turn or two extra, that won’t make much difference to its function. We do like to glue layers together as we go, as it stops the coil from unwinding itself if we take a break. However, it is possible to wind these inductors without it. We tend to keep winding if a turn slips onto the wrong layer, as it gets too messy undoing whole layers. Finally, regarding the wiring, your speaker positive (red) connects to the output of the amplifier on CON3. Your speaker negative (black) connects to the 0V terminal of CON4. You could terminate the speaker negative wire at the unused ground terminal of CON3, but that is not the best way. You will get more distortion, and in extreme cases, it might affect stability. The best way to route the speaker negative is to run it parallel with the speaker positive wire to the amplifier as if you was going to terminate it on the amplifier board. However, don’t cut the speaker ground (negative wire) off. Extend it, wrapped with the positive, negative and ground wire (the one that does connect to that 0V point) all the way back to your power supply main ground point. You then connect the speaker negative wire and the amplifier 0V power supply wire to your star Earth point. This way, you only have a small current in the wire from the 0V terminal of CON4 through to the main Earth point, so there is minimal noise and voltage drop on this wire. If you connect the speaker negative to CON4’s 0V terminal, then all the current to the speaker is carried by that wire and it also has to flow through the power supply ground wire. This current causes a voltage drop in that wire this is injected into the amplifier ‘ground’. So investing in that extra wire is really important for minimising distortion. It is helpful if you remember that anything connecting to ground/Earth should go back to that single point, within reason. Arduino Board Profile update broke our code I have recently successfully constructed and fully tested the Mains Power Up Sequencer with the current detection option (February & March 2024; siliconchip. au/Series/412). I plan to use the current detection feature to detect the TV being switched on, followed by powering on various items, with the audio amplifier last (and sequentially off in reverse order) to avoid a speaker thump. Unfortunately, the TV is unable to be completely powered off without switching it off at the wall, and the sequencer is triggered by the TV’s standby current. I tried winding fewer turns on T1 to reduce the sensitivity, down to one turn. While it then no longer responds to the standby current, it also won’t detect when the TV is actually powered on with the remote control. I note in the description that it says a 100W load resistor would provide more linearity, but all we are trying to do is detect current (and I assume not measure it precisely), so 10kW was chosen. I’m reluctant to measure anything or fiddle with the lid off, so I was wondering if you could suggest how I can modify it to work around this problem. Otherwise, it is an excellent, well thought out and designed project. (S. D., Wantirna South, Vic) ● The trouble with altering the turns on the transformer is that adjustment is too large a step in sensitivity. We think the threshold between the TV standby current and fully-on current could be set by using a variable resistance for T1’s load. Instead of the fixed 10kW resistor, use a 2kW trim pot wired between the wiper and one end terminal. You can then adjust the resistance (with the power off!) so that it ignores the TV’s standby current but responds to you switching it on. There should be a range of resistances over which it will work. Once you’ve found the ideal centre position, you could measure the resistance across the pot (again, with the power off) and replace it with a fixed resistor of that value. I have finally gotten round to constructing a Mini WiFi LCD Backpack (October 2020; siliconchip.au/Article/ 14599) but have encountered an error during the sketch verify process. For the most part, the output appears to be technical warnings. However, there seems to be a fatal error relating to the library file “gui.h”. I am running Arduino IDE version 2.3.4 on a MacBook Air M2 with ‘’Rosetta” installed to allow the IDE to run on Apple silicon. I have been able to load “Blink” into the Mini WiFi LCD Backpack and also succeeded in getting the Jaycar WiFi Weather Logger going, so I assume my development environment is OK. Would you please have a look at the log and comment? In the long run, I would like to be able to display the WiFi Weather Logger data on the D1 Backpack, initially via my local WiFi network, then hopefully via the internet. I also have a couple of other applications in mind for monitoring a couple of digital inputs and an analog input with a WiFi Mini and displaying them on the BackPack. Thank you for your assistance. (A. S., Parkinson, Qld) ● We immediately thought this would be a problem with your version of the ESP8266 board profile, and one line of the error message confirms this is the case: error: call to ‘HTTPClient::begin’ declared with attribute error: obsolete API, use ::begin(WiFiClient, url) The creators of the ESP8266 board profile have changed the way some things work, which has ‘broken’ our sketch. The version release notes indicate that board profile version 3.0.0 made several breaking changes Australia's electronics magazine siliconchip.com.au Mains Power-Up Sequencer trigger current threshold 102 Silicon Chip MARKET CENTRE Advertise your product or services here in Silicon Chip KIT ASSEMBLY & REPAIR FOR SALE PCB PRODUCTION DAVE THOMPSON (the Serviceman from Silicon Chip) is available to help you with kit assembly, project troubleshooting, general electronics and custom design work. No job too small. Based in Christchurch, New Zealand, but service available Australia/NZ wide. Email dave<at>davethompson.co.nz LEDsales PCB MANUFACTURE: single to multilayer. Bare board tested. One-offs to any quantity. 48 hour service. Artwork design. Excellent prices. Check out our specials: www.ldelectronics.com.au KEITH RIPPON KIT ASSEMBLY & REPAIR: * Australia & New Zealand; * Small production runs. Phone Keith: 0409 662 794 keith.rippon<at>gmail.com LEDS, BRAND NAME AND GENERIC LEDs, filament LEDs, LED drivers, heatsinks, power supplies, kits and modules, components, breadboards, hardware, magnets. Please visit www.ledsales.com.au Silicon Chip Binders REAL VALUE AT $21.50 PLU S P&P PMD WAY offers (almost) everything for the electronics enthusiast – with full warranty, technical support and free delivery worldwide. Visit pmdway.com to get started. Order online from www.siliconchip. com.au/Shop/4 or call (02) 9939 3295. ADVERTISING IN MARKET CENTRE Classified Ad Rates: $32.00 for up to 20 words (punctuation not charged) plus $1.20 for each additional word. Display ads in Market Centre (minimum 2cm deep, maximum 10cm deep): $82.50 per column centimetre per insertion. All prices include GST. Closing date: 5 weeks prior to month of sale. To book, email the text to silicon<at>siliconchip.com.au and include your name, address & credit card details, or phone (02) 9939 3295. (https://github.com/esp8266/Arduino/ releases/tag/3.0.0). The quickest fix will be to change the installed version of the board profile from the Boards Manager (there is a drop-down to change the version). We don’t have a record of the version used for the original project, but it was likely one of the later 2.x.x versions. We suggest you try 2.7.4. There were similar problems with the Clayton’s GPS Time Source project and version 2.7.4 worked in that case. Wideband Differential Probe output is too high I’ve finally got around to building the Wideband Active Differential Probe (from the September 2014 issue; siliconchip.au/Article/7995). I ordered the parts in 2015, lost them, then finally ordered them again a couple of months ago... siliconchip.com.au I realise that this is an old project. However, I wondered if you might have any troubleshooting information available based on any questions at the time. I’ve looked in my old magazines for the couple of months after in case there was anything there, but didn’t find anything. My problem is that when feeding in a 2V peak-to-peak square wave at around 200kHz, the voltages displayed on the oscilloscope are either 3.32V peak-to-peak at ×1 or an overshooting 3.6V peak-to-peak (3.2V if I ignore the overshoots) at ×10. The overshooting looks much like the shape of needing to calibrate a probe. The connection to the oscilloscope is an SMA-to-BNC coaxial cable and I’m feeding the probe’s inputs from the 50W output of a signal generator via a BNC to crocodile clip coaxial cable. I’ve measured the output of the signal generator using its 600W output direct Australia's electronics magazine to the scope with a BNC-BNC coaxial cable and this matches the 2V peak-topeak the generator claims to put out. Any troubleshooting thoughts, hints or pointers would be much appreciated. (J. B., Little River, New Zealand) ● The Wideband Active Differential Probe is designed to be used with a 50W cable to the oscilloscope and with 50W termination. Without this, the displayed levels will be incorrect. Many oscilloscopes have a 50W termination option for each input. Make sure you’ve enabled it for the one the Probe is connected to. If you are using this 50W termination, perhaps the gain of the differential amplifier is not correct. Check the values of the 1kW and 1.3kW resistors that set the gains of IC1 and IC2. Also check that the three 3MW resistors in series at the input and the 1MW shunt resistors at the gates of Q1 and Q2 are correct. April 2025  103 For 10:1 input compensation, you may require a nominal 5pF capacitor across the 9MW resistance. Bass power amplifier suggestion Do you have a suggestion for a sizeable bass power amplifier in the range of 150-300W (RMS)? I don’t need a preamp as I have a small high-quality head from which I use the line out to drive a large 15-inch (380mm) speaker cab. (J. C. H., Mount Barker, SA) ● You can search our projects on our website at siliconchip.au/Articles/ ContentsSearch A search there reveals a few good options for you: 1kW+ Class-D Amplifier, Pt1 by Allan Linton-Smith (October 2023; siliconchip.au/Series/405) Advertising Index Altronics.................................53-56 Beware! The Loop......................... 6 Dave Thompson........................ 103 DigiKey Electronics....................... 3 Electronex..................................... 7 Emona Instruments.................. IBC Hare & Forbes............................. 19 Icom............................................... 5 Jaycar............................. IFC, 41-44 Keith Rippon Kit Assembly....... 103 LD Electronics........................... 103 LEDsales................................... 103 Microchip Technology.............OBC Mouser Electronics....................... 4 OurPCB Australia........................ 23 PCBWay......................................... 9 PMD Way................................... 103 Silicon Chip Back Issues........... 70 Silicon Chip Binders................ 103 Silicon Chip PDFs on USB......... 95 Silicon Chip Shop.......... 63, 71, 79 Silicon Chip Subscriptions........ 57 The Loudspeaker Kit.com.......... 93 Wagner Electronics..................... 10 104 Silicon Chip 500W Class-D Mono Amplifier by Phil Prosser (April 2023; siliconchip. au/Article/15730) 500W Power Amplifier, Part 1 by John Clarke (April 2022; siliconchip. au/Series/380) New SC200 Audio Amplifier by Nicholas Vinen (January 2017; siliconchip.au/Series/308) Ultra-LD Mk.4 200W RMS Power Amplifier, Pt.1 by Nicholas Vinen (August 2015; siliconchip.au/ Series/289) Ultra-LD Mk.3 200W Amplifier Module by Nicholas Vinen (July 2011; siliconchip.au/Series/286) Studio 350 Power Amplifier Module by Leo Simpson & Peter Smith (January 2004; siliconchip.au/Series/97) Programmable Ignition System queries I have some questions about the Programmable Ignition System from the March-May 2007 issues (siliconchip. au/Series/56). 1. I am presently building the Independent Electronic Boost Control from the Performance Electronics for Cars book, including the Hand Controller. I noticed circuit differences between the Hand Controllers used for that project and the Programmable Ignition System. I would like to use the same Hand Controller for both systems. Can I do that, or do I need to plan on having two Controllers? 2. The car I intend using the Ignition System on has a GM HEI distributor with an onboard coil; all the electronics are under the cap. This unit has no external connections beyond a 12V source, a vacuum source and a connection to an electronic tachometer. None of the six options shown with the ignition system appear to deal with this situation. I don’t think there are any user-­ serviceable parts under the distributor cap, so there are no wires to tap into. Can I use the Programmable Ignition System with my distributor, or would I be better served remaining stock or doing something else? Thank you for your advice and help. (A. M., Fairfax, Virginia, USA) ● The Hand Controller used for the Boost Controller and the Programmable Ignition are essentially the same and can be used interchangeably. The added resistor array is just there to minimise the possibility of corrupted data. The resistors terminate the signal to prevent transmitting or receiving incorrect signal data. The version with the resistor array is preferred, but the original version could still be used. As far as the GE ignition system goes, unless you can access the connections to the trigger or ignition coil primary, it is impossible to make a connection to the Programmable Ignition input. The Programmable Ignition System effectively has to be wired between the trigger source and the ignition coil so that it can advance and retard the spark timing. You could remove the GE ignition and add in a Hall Effect trigger or one of the optical triggers instead. That would allow you to use the Programmable Ignition System. 10-Plus-10 Stereo Amp circuit from EA How good are your Electronics Australia archives? Many years ago, I built (what I am sure) was a version of EA’s “Playmaster” 10 + 10 solid state stereo amplifier. I think it was from the late 1960s, possibly 1968. The amplifier used AD161 & AD162 transistors in the output stages and had a regulated 30V power supply. I am after a copy of the circuit if you can provide it. (P. W., Pukekohe, New Zealand) ● After much searching, we finally found this article: the “10-Plus-10 Stereo Amplifier” design was published in the November 1968 issue of EA, starting on page 44. It was not listed in any of the EA project indexes we have! SC Errata and on-sale data for the next issue Universal Loudspeaker Protector, November 2015: the 6.8kΩ resistor shown below IC1 in Fig.2 should be 5.6kΩ as per the circuit diagram and parts list. It may also be marked on the PCB as 6.8kΩ. Next Issue: the May 2025 issue is due on sale in newsagents by Monday, April 28th. Expect postal delivery of subscription copies in Australia between April 25th and May 15th. Australia's electronics magazine siliconchip.com.au “Rigol Offer Australia’s Best Value Test Instruments” New 2024 Products Oscilloscopes New 12Bit Scopes RIGOL DS-1000Z/E - FREE OPTIONS RIGOL DHO Series RIGOL MSO-5000 Series 450MHz to 200MHz, 2/4 Ch 41GS/s Real Time Sampling 424Mpts Standard Memory Depth 470MHz to 800MHz, 2/4 Ch 412Bit Vertical Resolution 4Ultra Low Noise Floor 470MHz to 350MHz, 2 Ch & 4Ch 48GS/s Real Time Sampling 4Up to 200Mpts Memory Depth FROM $ 499 FROM $ ex GST 659 FROM $ ex GST 1,489 Multimeters Function/Arbitrary Function Generators New Product New Product RIGOL DG-800/900 Pro Series RIGOL DG-1000Z Series RIGOL DM-858/E 425MHz to 200MHz, 1/2 Ch 416Bit, Up to 1.25GS/s 47” Colour Touch Screen 425MHz, 30MHz & 60MHz 42 Output Channels 4160 In-Built Waveforms 45 1/2 Digits 47” Colour Touch Screen 4USB & LAN FROM $ 713 FROM $ ex GST Power Supplies ex GST 725 FROM $ ex GST Spectrum Analysers 689 ex GST Real-Time Analysers New Product RIGOL DP-932E RIGOL DSA Series RIGOL RSA Series 4Triple Output 2 x 32V/3A & 6V/3A 43 Electrically Isolated Channels 4Internal Series/Parallel Operation 4500MHz to 7.5GHz 4RBW settable down to 10 Hz 4Optional Tracking Generator 41.5GHz to 6.5GHz 4Modes: Real Time, Swept, VSA & EMI 4Optional Tracking Generator ONLY $ 849 FROM $ ex GST 1,321 FROM $ ex GST 3,210 ex GST Buy on-line at www.emona.com.au/rigol Sydney Tel 02 9519 3933 Fax 02 9550 1378 Melbourne Tel 03 9889 0427 Fax 03 9889 0715 email testinst<at>emona.com.au Brisbane Tel 07 3392 7170 Fax 07 3848 9046 Adelaide Tel 08 8363 5733 Fax 08 83635799 Perth Tel 08 9361 4200 Fax 08 9361 4300 web www.emona.com.au EMONA