Fullscreen HTML5Med Res (??MB)HTML4

Part 2 by Tim Blythman This Multi-Channel Volume Control can handle up to 20 independent channels, allowing you to build your own home theatre or surround system. You can use a touchscreen LCD panel, an IR remote control or an OLED Module with a rotary encoder to control it. This article has all the construction details. Multi-Channel Volume Control O ur Multi-Channel Volume Control can adjust the levels of up to 20 audio channels by touchscreen, IR remote or a rotary encoder. It’s modular, so you can build it with four or eight (or twelve or sixteen) channels if that is all you need. It’s intended to be incorporated as a part of a larger amplifier system, perhaps using several of our Hummingbird Amplifier modules (December 2021; siliconchip.au/Article/15126). But there is no reason it couldn’t be built as a dedicated unit in its own case. With the principles of operation covered last month, it’s time to commence construction. We assume you have already worked out what modules to build and have the parts at hand. We’ll describe the construction of each type of module in turn. You will need one Control and Power Supply Module and at least one Volume Module. If you want a rotary encoder volume control, you must also build the OLED Module. After that, we’ll go over testing the modules and connecting them together into your system. Since all three module types feature surface-mounting parts, check that you have the necessary tools for this sort of work. A fine-tipped soldering iron, flux paste (and a corresponding cleaning solvent), solder-wicking braid, tweezers and fume extraction are all highly recommended. Some sort of magnifier 74 Silicon Chip and a good light source are helpful for those with diminishing eyesight. We’re talking from experience here! Working outside is a good alternative to fume extraction and should also provide sufficient illumination. Enclosures If you have decided on your choice of enclosure, you can use the blank PCBs to mark the mounting hole positions. It will be easier to do this now before parts are fitted to the PCBs. Look at Figs.12 & 13 to get an idea of the cuts that need to be made for the Control and Power Supply Module and OLED Module, respectively. Control and Power Supply Module The through-hole parts on this Module are mainly in the power section, while the SMD parts are mostly related to the microcontroller. We’ll start with the SMDs. The three different SOT-23 package parts are the smallest and are all different types, so don’t mix them up. Check their orientation against the photos and PCB overlay, Fig.8. SOT-23 parts are small, but their leads are spread out, so you shouldn’t get bridges between pins. REG3 is the MCP1700-3.3 type. Apply flux paste to the three pads and hold it roughly in place with the tweezers. Tack one pin and adjust its position (melting the solder with the iron if needed) until it is square with the pads and flat against the PCB. Then apply Australia's electronics magazine solder to each of the other pins in turn. Refresh the first pin if necessary. Use the same technique to solder the two Mosfets, Q1 and Q2. Q1 is the P-channel part, while Q2 is the N-channel 2N7002 part. Follow with IC9, the 20-pin PIC16F18146 microcontroller. Check its orientation and ensure that pin 1 is located near the dot on the PCB near where the capacitor will be fitted later. Like the earlier parts, apply flux and rest the IC in place. It is larger, so it might not need to be held down with tweezers. Tack one lead and adjust the IC location until it is centred on its pads and flat against the PCB. Next, carefully solder each IC pin to its pad on the PCB. If you do form a solder bridge, leave it for now. Solder the remaining pins to secure the chip in the correct place. To fix a solder bridge, apply more flux and press the solder braid against the bridge with the iron. When it has drawn up the solder, carefully slide it away from the IC and repeat as necessary. After using the braid, surface tension should retain enough solder to form a solid joint, as long as the IC is flat against the PCB. If you’re not sure, have a close look using a magnifier and refresh the pins with the iron using some more solder and fresh flux. The remaining surface mounting parts are all M3216/1206 size (3.2×1.6mm) passives and can be fitted using similar techniques. There siliconchip.com.au Fig.8: there is a mix of SMD and through-hole parts, with components on both sides of this Module. Fortunately, none of the SMD parts are too small. Just take care not to mix up the components and watch the orientations of the IC, bridge rectifier, electrolytic capacitors and box header. are two 100nF capacitors and one 1μF capacitor, which won’t be marked, plus some resistors. Five SMD resistors are fitted to the same side of the PCB, plus three on the other side. To check the resistance codes printed on the parts, refer to Table 1 for the expected markings. Use a solvent to clean up the excess flux on the PCB. Isopropyl alcohol (isopropanol) is a suitable general-­purpose solvent for this. Wipe off as much excess as possible and then allow the remainder to evaporate. Inspect the PCB with a magnifier to ensure that your soldering looks correct. It will be much easier to make corrections now, before any other components are fitted. Through-hole parts The through-hole parts on the Control and Power Supply Module should generally be fitted from shortest to tallest, as that simplifies the process. Refer to the photos and overlay diagrams if needed. Start with the 5.6V zener diode, ZD1. Bend the leads by 90° and thread through the PCB, ensuring the cathode band matches the PCB silkscreen. Solder the leads and trim so they are neat. Follow with REG1, the sole TO-220 package regulator. Bend the leads backward by 90° at a point about 7mm from the body. Thread the leads through their PCB pads and affix the regulator with the machine screw, nut and washer. Once you are happy with the location, solder the leads and trim as needed. Fit bridge rectifier BR1 next, with the + polarity mark on the PCB matching the one on the rectifier. Push it down flat against the PCB before soldering it. Then adjust 500W trimpot siliconchip.com.au VR2 so its wiper is at its midpoint and solder it to the PCB. CON7 and CON8 are next. You don’t need to fit both, as only one is needed to supply power, but we used both on our prototype for testing. CON8 is required for a 24V AC centre-tapped supply, while either can be used with a single 12V AC tap. If you have a choice, the 24V AC centre-tapped transformer with CON8 is preferred. Install the three different regulators in TO-92 cases next, being careful not to mix them up. REG2 is the 78L12, REG4 the 79L12 and REG5 the LM317L. These are also marked on the PCB silkscreen. Now mount CON11 with the key in the box header facing away from the other components on this side. There might also be a marker on the box header indicating pin 1, which goes near the top of the PCB. The three different types of electrolytic capacitors are fitted next. There are four 100μF parts and two 220μF parts around CON11. Ensure that the polarities and values are correct before soldering, with the longer, positive leads towards the + markings on the board. The polarity of the two larger 1000μF capacitors near BR1 are reversed compared to the others. The last remaining component on this side of the PCB is the 5W resistor. Bend its leads and fit it to the PCB pads. Space the body of the resistor about 5mm clear of the PCB. You can tack one lead and adjust its position (if necessary) before soldering the other lead. Components on the other side To help align CON9 for the LCD touchscreen module, fit the four 12mm The majority of the components on this side are SMD parts related to driving the LCD module. Note the mounting for IRRx1, circled in red. Australia's electronics magazine January 2024  75 M3-tapped spacers to the Module using four M3 screws, with the latter on the same side as the through-hole components. Rest the 14-way female header (CON9) in place, then slot the LCD module into it, allowing the header to sit at right angles to the PCB. Solder CON9 to the board. You can test the arrangement for the IR receiver, IRRx1, next. We mounted it so it peeks out just above the top of the LCD (see the photos). There are many ways to mount IRRx1, but we think this method will work in most cases. Regardless of how you do it, just be sure that the correct pins of IRRx1 go to the correct PCB pads. You should also fit CON10 for in-­ circuit programming unless you have a pre-programmed microcontroller. We placed it on the top of the PCB, but it could also be fitted to the reverse if necessary. Programming the micro If you need to program the microcontroller in-circuit, use a 3.3V supply voltage. Also, detach the LCD module before programming to reduce the load on the programmer’s power source. These newer PICs can only be programmed with a PICkit 4 (or later) or a Snap; with the Snap, you will need to provide power separately. We discussed modifying a Snap to supply power on page 69 of the June 2021 issue (“PIC Programming Helper”; siliconchip.au/Article/14889). Use the IPE to upload the 0111122B. HEX file (0111122C.HEX is for the OLED Module) and confirm that you get the “Program/Verify complete” message. You won’t see anything that indicates that it is working right away. Testing Leave the LCD module off when checking the supply rails on the Control and Power Supply Module. It’s a good idea to do this with nothing attached, especially as we need to trim the 5.5V rail. You can connect a current limited DC supply (eg, a bench supply) to the CON8 screw terminals. Connect the negative supply to CON8’s centre GND connection with the positive supply to either of the remaining terminals. This will provide power to the positive regulators. Reversing the polarity will power the negative regulator, which we will do later. Set the current limit to around 100mA and slowly wind up the supply voltage. With 15V applied, we found that our prototype’s 12V, 5.5V, 5V and 3.3V rails were correct (within 0.1V), with the Module drawing about 60mA. You can access the 12V, 5.5V and 3.3V rails at pins 2, 4 and 9 of CON11, respectively. The 5V rail can be sensed at pin 1 of CON9 (where the LCD panel connects). CON8’s centre pin or REG1’s tab are good places to connect to ground for referencing these readings. Assuming there is 12V on the 12V rail, adjust VR2 to get a reading of 5.50V, or as close as possible, across ZD1. Don’t exceed 5.6V, or ZD1 will start conducting and could get warm. If you can’t trim the 5.5V rail, check the resistor values. Since the other regulators are fixed, there isn’t much else that can go wrong apart from the wrong regulator being fitted or the bridge rectifier not being installed correctly. Reverse the polarity applied to CON8 to check the -12V rail at pin 3 Table 1 – SMD resistor codes Value 3-digit code 104 1003 47kW 473 4702 22kW 223 2202 10kW 103 1002 2.2kW 222 2201 76 1kW 102 1001 910W 911 910R 680W 681 680R 560W 561 560R 110W 111 110R 100W 101 100R Silicon Chip LCD module backlight One of the problems we encountered during the design and testing of this project and the earlier Digital Preamp is that the LCD backlight has the heaviest current draw of any component. In the Digital Preamp, we applied the well-known technique of modulating that draw by applying a PWM signal to the backlight control. For this project, we wanted to tackle this in a better way, as it was apparent that the PWM signal was having a small but noticeable effect on the measured audio quality. So for this project, we have avoided using PWM control of the LCD backlight. You can see that the power section of the circuit now uses a 5W resistor instead of several 1W resistors, so it is better able to handle the full backlight current. We still found that the 5W resistor was getting warm, so we had a closer look at what we could do to reduce dissipation. While getting the Module to run cooler is always an advantage, we hoped the lower current draw would lead to less ripple on the main supply capacitors and thus better performance. The “LCD screen backlight modifications” panel explains how the backlight works on these LCD modules and discusses a minor modification that can be made to reduce its current draw. This modification is optional, so you can skip it if you like. Reattach and secure the LCD module using the four remaining machine screws. We can now test that the microcontroller is working correctly and can produce a display on the LCD screen. Screen 1: if you see this screen when you power up your Multi-Channel Volume Control, the Control Module is functional. The red circle at upper right is an IR (infrared) telltale that lights up whenever an IR remote control signal is received (whether it is recognised or not). 4-digit code 100kW of CON11. In this case, we found that the Module only drew about 30mA. Australia's electronics magazine siliconchip.com.au Using the connections you used to test the positive regulators (ie negative to GND, positive to either of the AC[~] connections), set the limit to around 300mA and wind up the voltage. You should see something on the LCD with the input at about 8V or higher. If you don’t see anything by 15V, there may be a problem. The actual current draw will depend on the type of LCD backlight and may be different if it has been modified, but it shouldn’t be any higher than 300mA. You should see a screen similar to Screen 1, and the UP/DOWN/MUTE buttons should respond to presses. That’s as much as we can test at this point. Fig.9: all parts for the Volume Module mount on the top side. Slightly smaller M2012/0805 size passives will fit the same pads. Watch the orientations of the ICs, the electrolytic (including tantalum) capacitors and the box header. Volume Module The Volume Module can be built without the last op amp stage if you want to save a bit of money and time, and that will also improve the volume control range if you don’t need the high maximum gain. All our performance specs are based on the fully populated version; performance will likely be the same or better without those extra op amps. We’ll describe the assembly for all components being fitted. If you wish to leave out the last op amp stage, omit IC3, IC7, their respective 100nF capacitors (one each) and the eight 1kW resistors in that area of the PCB. The two remaining 1kW resistors that pad VR1 are at the other end of this PCB. They are still used. If you omit IC3 and IC7, short out the four PCB jumpers pairs, JP3-JP6. Apply your iron to the pads of the jumper and feed in a generous amount of solder until a bridge forms. You can use solder wicking braid if you need to remove the bridges. Fig.9 and the PCB photos show the fully populated version that we will now assemble. The Volume Module is mostly populated with SMD components, with just a handful of through-hole parts. Start by fitting the eight dual diodes in SOT-23 packages. Apply flux to the pads and rest each diode in place, noting the orientation from the photos and overlay diagram. Tack one lead, adjust the positioning and then solder the remaining leads. Add some fresh flux and touch the iron to the first lead if you need to refresh that joint. Follow with the eight op amps, siliconchip.com.au IC1-IC8. They all face the same way, with their pin 1 facing towards the bottom of the PCB. Small parts like this may not have a dot printed on their bodies, but may have a bevel along the edge nearest pin 1. This bevel is most easily seen from the end of the chip. IC10, the 28-pin SOIC part, should be soldered next. Its pin 1 is orientated in the opposite direction from IC1-IC8. If you have any solder bridges on these parts, rectify them using more flux and solder-wicking braid. The top half of the PCB is marked with horizontal lines and values down the middle, indicating that four identical parts are fitted across. Each part corresponds to one of the four channels, hence the symmetry (Fig.9 shows the values individually for clarity). The remaining SMD parts on this PCB are two-lead passives. Fit ferrite beads FB1-FB4 next. They are identical and marked as FB on the PCB silkscreen and overlay diagram. The ferrite beads will probably be dark grey, matching the ferrite material they are made from; ceramic capacitors are usually a lighter beige/brown colour. The four tantalum capacitors are in a row near IC1 and IC5. As these are polarised, observe the polarity markings. It’s important to note that, unlike electrolytic can capacitors, rectangular moulded (as well as tag tantalum) electros have a stripe on the positive end, similar to a diode’s cathode marking! Australia's electronics magazine Also, our prototype used ceramic capacitors, which look different to the tantalum parts we will supply in kits. You could use high-value SMD ceramics if absolutely necessary, but they are generally inferior for audio signal coupling compared to electrolytic caps. After that, install the 11 100nF capacitors, the four 470pF capacitors and the 100pF capacitors. There are many SMD resistors of different values; naturally, they should not be mixed up. Fortunately, their values will be marked, so you can check them as you go (you might need a magnifier) – see Table 1. If you’re unsure of reading the codes, carefully use a multimeter to measure their resistances. You could even measure them using our Advanced SMD Test Tweezers from the February and March 2023 issues (siliconchip.au/Series/396). Cleaning and checking Now use your preferred flux cleaning solvent to remove any excess flux from the PCB and allow it to dry. It’s a good time to inspect the assembly and check that all the components look to be soldered correctly in the right spots before fitting the remaining components. For JP2, you might like to use a simple wire link if you know what your configuration will be. If so, populate the first board with a link across CS1, the second with a link across CS2 etc. January 2024  77 The Volume Control Module shown fully populated. The op amps just behind the RCA sockets can be left off if a lower maximum output signal is required. The four 1μF ceramic capacitors have been replaced with 2.2μF tantalums in the final version for improved performance. If you’re not sure, install the double-­ row pin header and place the links as described for testing. Adjust the 500W VR1 trimpot to near its midpoint, then solder it in place (or centre it after soldering). Next, fit box header CON5. Its key should be to the left, with pin 1 towards the middle of the PCB, as indicated by the arrow on the silkscreen. You could use a double-row pin header at a pinch, although that won’t guarantee the correct plug orientation. Next, mount the nine electrolytic capacitors. Watch out for the polarities (the longer lead is positive, while the stripe indicates negative) and install them as shown. The last parts to be soldered are the RCA sockets. Their pins and alignment pegs take a bit of wrangling, so ensure their bases are flush against the PCB before soldering them in place. We also suggest adding a tapped spacer to each of the bottom corners of the PCB. Secure them from above with machine screws. These are used to mount these boards to your choice of enclosure but will also keep the PCBs off your bench during testing. OLED Module The optional OLED Module is the smallest of the three. It is little more than a microcontroller, a rotary 78 Silicon Chip encoder and an OLED screen. All the components are fitted to one side of the PCB; the other side forms its front panel. You can see this in the Fig.10 overlay diagram and the photos. Fit the PIC16F15224 microcontroller (IC11) first. Add flux to the PCB pads, rest the micro in place, tack one lead and check its alignment before soldering the remaining leads. There are four 100nF capacitors and four 10kW resistors. None of these are polarised, and can be soldered next. At this stage, you should also add a solder bridge to the CS5 (bottom-most) position of JP7. Now clean off the excess flux and allow the board to dry. Inspect the solder joints of the smaller components and rectify any concerns. This will be easier before the larger parts are fitted. If IC11 is not programmed, you will need to fit CON13, the ICSP header. As you can see from the photo, we used a right-angled header fitted as a surface-­ mounted part. To program IC11, set your programmer to provide 3.3V, connect it to the ICSP header and upload the 0111122C. HEX file. Next, solder CON12, the surface-­ mounting box header. Note the pin 1 marking indicating the orientation. The key for the tab on the cable should face towards the top edge of the PCB. You could use a standard surface-­ mounting dual-row pin header if you don’t have a box header, but it will lack the keying that ensures the plug is always inserted correctly. Like any other part, apply some flux, rest the header in place and tack one lead. Adjust the position if necessary, then solder the remaining leads. Since these larger pins are at 0.1in (2.54mm) spacing, you can be pretty generous with the solder. You should be able to look at the gap between the PCB and the box to see that there are no bridges. Fit the rotary encoder (RE1) next. Mount the encoder using the supplied nut and then add short lengths of component leads to make the connections to the pads below. When fitting the encoder, ensure that the pins match the PCB silkscreen markings (two pins on one side and three on the other). Once the encoder is aligned, you can mechanically secure it using the pads on each side of the body. We used lead offcuts around 1cm long, bent about 3mm from one end. Tin the PCB pads and the ends of the leads and then solder the short end of the leads to the PCB. We used tweezers to hold the other end of the leads while soldering them, then gently bent the other ends of the leads against the pins of the rotary encoder and soldered them together. You can see this in the photo below. Similarly, the OLED uses short lead offcuts for its four electrical connections. Don’t fit the headers to the This shows how the rotary encoder and OLED are attached to the PCB. They both use short lengths of wire, such as component lead offcuts, to connect to the PCB. Note how we’ve soldered a header to CON13 to program the microcontroller in circuit. siliconchip.com.au Fig.10: to allow the PCB of the OLED Module to be used as the front panel, all the components are surfacemounted, including the usually through-hole parts. You can also see this in our photos. OLED, as we aren’t using them. If one is already fitted, desolder it and clear the pad holes of solder. Tin each of the four pads on the PCB and then solder a lead offcut vertically. Remove the protective film from the OLED and ease the OLED module down over the leads until it is flush against the PCB. Gently adjust the position of the OLED so that it is square within the markings on the PCB, then solder each of the four wires to the pads on the Module. Add two more lead offcuts to the two large bottom holes of the OLED and solder them to the PCB pads below. The OLED should light up if you apply 3.3V and GND (via the ICSP header or pins 9 and 20 of CON12). That’s about as much testing as is possible for now. length of 20-way ribbon cable and fit it with one 20-way IDC plug along its length for each module you have built. They don’t have to be in a specific order, as it is all a single bus. Pin 1 of each plug must align with the marked pin 1 of the cable (usually red). Otherwise, it doesn’t matter too much. The sockets can sit above or below the cable; the endmost sockets should have the cable looped back through their locking tabs to secure them. It’s best to use a designated IDC crimping tool such as Altronics’ T1540, but it is possible to use a bench vise with some care. Keep the cable square to the headers and use some pieces of timber on the faces of the vise spread the load. Proceed carefully to avoid cracking the IDC plugs. Ribbon cables Now connect all the modules together with your ribbon cable and wire up your AC supply of choice. A single 12V AC source can connect to CON7 or between the GND and one of the AC phases on CON8. For a 24V AC supply, connect its centre tap to the GND of CON8; the outer 12V taps go to the other terminals of CON8 (it doesn’t matter which). Now we must join all the modules with a custom 20-way ribbon cable. The exact arrangement depends on how you plan to arrange your modules within your enclosure, so we don’t have a specific assembly diagram of such a cable. Fig.11 shows how a typical cable might look. You should use a single Commissioning Fig.11: this is only an example of a possible ribbon cable; you might have different requirements depending on your choice of modules. As long as the pin 1 markings align with the same edge of the cable, the cable should work. Note how the keys on the headers on one side of the cable face the same way, opposite to the keys on the other side. siliconchip.com.au Australia's electronics magazine Power on the Multi-Channel Volume Control and verify that the LCD panel shows the expected screen. We still need to perform one last setup step for each of the Volume Modules. Take a multimeter and confirm that there is 5.5V between TP1 (GND) and TP2 (5.5V) of each Volume Module. If so, adjust trimpot VR1 on each to get 2.75V at TP3. This completes the hardware setup. We’ll now delve into the firmware settings to complete the configuration and then work through the operation of the controls. OLED Module If you have an OLED Module fitted, you should be able to operate its controls and see that both displays update together. Screen 2 shows a typical OLED Module display. There is no configuration needed for the OLED Module. The OLED Module will show three dashes when powered up until it receives data on the ribbon cable. If the dashes persist for more than a few seconds, the OLED module may not be receiving data correctly. In that case, check the ribbon cable and connectors, especially that the IDC plugs are fully clamped around the ribbon cable. Screen 2: the OLED Module display will show this on the screen (depending on the MUTE state). If you see three dashes then the OLED Module is not receiving data from the Control Module. January 2024  79 If you find that the operation of the rotary encoder is backward, reverse the connections from the two outer pins to the PCB using short lengths of insulated wire. We haven’t seen this happen, but it is an easy fix. Setup The default settings for the Multi-Channel Volume Control are to drive 16 channels with an OLED Module connected and the last op amp stages fitted to each Volume Module. If you have fewer than 16 channels, the ‘phantom’ channels will not respond, so you won’t need to change the settings even if you only have six or eight channels. The default IR code settings allow the Volume Control to respond to the Jaycar XC3718 IR remote control unit. Use the “−” and “+” buttons to change the volume and the PLAY/PAUSE button to mute and unmute. The LCD screen should show a red circle when a signal is received. If you don’t see a red circle when operating your remote control, its batteries could be flat, or the IR receiver may not be connected correctly. Screen 1 shows the IR telltale. To enter SETUP on the Control Module, press and hold the SETUP button on the LCD touch panel until the screen changes and you see Screen 3. In general, the “>” button cycles between the different settings, while the “+” and “−” buttons adjust them. The first four parameters set the IR device code and IR command codes. All the commands must correspond to the same device code. While these can be set manually, the option to ‘learn’ a code is also available. Press a button on your transmitter of choice and see that the value in brackets changes; these are the device and command codes the IR receiver detected. You might need to press another button and then your chosen button again to confirm this. Pressing this area of the screen (around the IR codes) will set the last received device code or command code as the current code. The values are stored in EEPROM and used immediately, so you can easily check that the Volume Control responds to the new IR code as expected. We have also found a set of codes that can be used with the Altronics A1012A Programmable IR Remote Control. Program the A1012A to use AUX code 0724 (which is for a Yamaha amplifier). This corresponds to device code 94 and command codes 216 (DOWN), 88 (UP) and 56 (MUTE). You could use the code-learning feature instead of having to enter these manually. Many other Japanese manufacturers use NEC codes. If the Yamaha code conflicts with existing equipment, a few other codes (from the Altronics A1012A list) that start with 07 also give valid NEC codes that the Volume Control can receive. The MAX VOLUME setting limits the highest value that the volume can be set to in dB. This can be set as high Screen 3: during setup, part of the screen is turned over to the setup parameters and buttons. Press and hold the SETUP button for five seconds to get to this screen and start the setup process. 80 Silicon Chip as 20dB and defaults to 5dB. Disabling the OLED Module is also possible by setting the SLAVE IN USE parameter to 0. If your OLED Module is not responding, check that this is set to 1. The LEVEL OFFSET parameter provides an adjustment to the overall gain. If you have omitted the last op amp stage on the Volume Modules, set this to -6 to account for the loss of the last ×2 gain stages. The next parameter changes the number of channels in use; this is the number of channels driven by the Volume Control. This should be a multiple of four and match the number of Volume Modules you have installed. If in doubt, set it to the maximum possible. Say you have two Volume Modules and are using six channels; in that case, set it to eight to ensure the two spare channels are set to safe levels. It can’t be set higher than 16 if the OLED Module is enabled. For these settings (apart from the IR codes), the values in brackets show the lower and upper limits of what these parameters can be set to. The remaining settings are offsets (in digital potentiometer steps) that can be applied to each channel. This can be used to adjust the balance between different speakers. A short press on the SETUP button returns to the normal display. Screen 4 shows what the display looks like when MUTE is active. The EEPROM text is also yellow, indicating that the current state has not been saved to Screen 4: when MUTE is active, the screen changes to look like this. The yellow EEPROM text means that there are unsaved changes. After 10 seconds of no activity, the state (volume and mute) is saved and will be reloaded if the Volume Control is switched off and then on again. Australia's electronics magazine siliconchip.com.au LCD screen backlight modification There are two common variants of the 2.8in LCD touchscreen panels. The main difference we noted is that the touch panels register differently, requiring different calibrations. As we mentioned in last month’s installment, the Multi-Channel Volume Control is programmed to handle these variations. Another difference is in the circuitry of the LED driver for the LCD panel backlight. The two variants we have seen are marked v1.1 and v1.2, as shown in our photos (adjacent and below). Both versions have an XC6206 3.3V regulator to power the LED controller from the panel’s VCC pin and an XPT2046 touch controller IC to provide an interface to the resistive touch panel. Fig.a highlights how they differ in their connections to the LED control line (one of the pins on the 14-way header). For the v1.1 boards, this line connects directly to the LEDs and then ground via a series ballast resistor. The later v1.2 boards use the LED control line to drive a low-side NPN S8050 transistor (Q1). The LEDs are wired to the VCC line, so when the transistor switches on, current flows via a ballast resistor and the transistor to ground. The v1.1 board design lends itself to dimming by an extra series resistor in the LED line. For example, the original Micromite LCD BackPack (February 2016; siliconchip.au/Article/9812) used a trimpot for manual backlight adjustment. The v1.2 boards do not allow that, so we have tended to use it less and less. While both arrangements can be driven by a high-current PWM signal (which could be provided by Q1 and Q2 of the Control and Power Supply Module), we have avoided using PWM in this project due to the resulting digital noise. So we looked into how to modify the LCD panel to adjust the backlight current linearly. Fortunately, changing the LED ballast resistor works well enough, which is what we did. Figs.b & c show a v1.1 board before and after modification. The green circle in Fig.b shows the resistor in question, originally 3.9Ω and designated R6. The original resistor was an M1608/0603 (1.6 × 0.8mm) part, but we replaced it with a larger M3216/1206 (3.2 × 1.6mm) part by scratching back some of the nearby solder mask to allow the larger part to be soldered. This must be done carefully as the surrounding copper area is connected to ground, and a bridge here will short the incoming LED signal to ground. We used 110Ω resistors for our tests Fig.a: the LED control lines for the V1.1 and V1.2 LCD modules. Figs.d & e: for the V1.2 LCD modules, we needed to scrape some of the solder mask so we could fit a larger resistor for R5. siliconchip.com.au Figs.b & c: a V1.1 touchscreen LCD module before (left) and after (right) replacing R6 with a 110Ω resistor to reduce the backlight current. Australia's electronics magazine because we had a few left over from building our prototypes. Figs.d & e show the v1.2 LCD panels before and after the changes. Here, the resistor is marked R5 and is 8.2Ω. We did the same thing, scraping some of the solder mask back to bare copper before soldering in the replacement resistor. You might even be able to solder in an axial leaded resistor by bending its leads back until they are nearly touching. Resistor value The 110Ω resistors were great at keeping the noise and heat down but the resulting backlight brightness is too dim for a well-lit room. We suggest 22Ω as a good compromise. A 100Ω trimpot in series with a 10Ω resistor would be a good choice if you want to tweak the brightness to suit your specific conditions. Our Control Module kits will include a 22Ω M1608/0603 SMD resistor so you can make this modification with a direct resistor swap. January 2024  81 The Power Supply and Control Module mounts to the LCD module using the 14-pin header and two Nylon M3 spacers. EEPROM; that happens automatically after 10 seconds of no further activity. Installing the modules To help you fit the modules into your desired enclosures, Figs.12 & 13 are cutting diagrams of the display cutouts for the Control & Power Supply Module and the OLED Module. The cutout for the Control & Power Supply Module is essentially the same as for the 2.8in LCD module. You could even consider using one of our laser-cut acrylic lids, such as SC3456 (siliconchip.au/Shop/19/3456), as a bezel for neatly mounting the LCD panel. This acrylic piece is intended to fit onto a UB3 Jiffy box and is 68mm tall, so it will be too tall for a 3U rack unit. Otherwise, refer to Fig.12 for the dimensions of the square cutout and screw holes to suit the 2.8in LCD. You will also need to create a hole for the IR receiver if you are using it. Its exact position depends on how you have fitted it. If mounting the LCD inside a metal enclosure, we recommend using a plastic bezel or foam tape to prevent the LCD pins from shorting against anything. Fig.13 shows the outline for the OLED Module. The outermost dimensions (76.5 × 51mm) are the outline of the Module, so you can start by marking these onto your enclosure. Use something erasable or work inside the enclosure, as these will be visible once the Module is fitted. Now add another set of lines 4mm inside these and yet another set of lines 4mm inside these; thus, the second set of lines is 8mm inside the Module’s border. These twelve lines will allow you to drill four holes and cut out the panel, as shown in Fig.13. Note that the inner cutout area does not need to be precise. You should leave enough material for the screws to hold. If the panel is metal, it is worth attaching some foam tape around the perimeter at the back, where the OLED Module attaches. This will prevent the case from scraping the solder mask and possibly shorting against the PCB traces. Completion The Multi-Channel Volume Control is intended to be a ‘subsystem’ within a system such as a multi-channel amplifier, so it is up to you how you connect it to your equipment of choice. As for the RCA sockets, the white upper connections are the inputs, and the red lower connections are the outSC puts of each Volume Module. Fig.12 (left): to mount the 2.8in LCD and thus the Control Module, you’ll need a large rectangular hole and four small round holes. You might also need another small hole for the IR receiver to ‘see’ outside (like the one marked “B”). Fig.13 (right): the exact dimensions of the cutout for the OLED Module are not critical, as the shape overlaps the edge of the hole by about 4mm. Still, you might need to use foam tape or similar to protect the back of the PCB if you are using a metal enclosure. 82 Silicon Chip Australia's electronics magazine siliconchip.com.au