Fullscreen HTML5Med Res (??MB)HTML4

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 may edit and has the right to reproduce in electronic form and communicate these letters. This also applies to submissions to “Ask SILICON CHIP”, “Circuit Notebook” and “Serviceman”. Microbridge serial interface connector pinout is correct I read the comment from Mike Flor in the Mailbag section of the July 2017 issue regarding the pinout of the serial connector on the Microbridge (May 2017; www.siliconchip.com.au/ Article/10648). His comment and your response both miss the point; the TX and RX pins have been purposefully swapped compared to the BackPack boards. When interfacing two serial devices, the TX pin of one goes to the RX of the other and vice versa. So the Microbridge can be plugged directly into the console connector of the LCD BackPack and it will work fine. I designed it that way! Geoff Graham, Perth, WA. Praise for analog circuitry In this day and age of all things digital in electronics, there’s one area that still fascinates this electronics enthusiast when it comes to technology. That is the confluence of the human ear and electronics. The war was fought; solid-state outdid all analog devices and then complete digital systems outdid their analog counterparts... or did they? I’ve been amazed recently at the Flash support will end soon, regardless of the consequences I read your reply to Terry Ives’ letter in the August 2017 mailbag regarding Flash and the Silicon Chip website, and I couldn’t help wonder if your stance has changed since Adobe’s big announcement (after the issue went to print), that Flash support would be discontinued in 2020. I can completely understand your aversion to re-engineering your production processes and website code around a new format but you may not have much choice. See: http:// siliconchip.com.au/l/aag3 What makes this even more serious from a content-creator’s point 4 Silicon Chip abundance of new valve amplifiers and analog audio technology appearing on the market in the last couple of years. (Not to mention the rise of the turntable and vinyl records again). It has always been there in the background (so to speak) but when I take a wander into my local hifi dealer and they’re recommending I go for a valve style pre-amp for example, surely this marketing amidst a multitude of other very good reasons has got to show that analog is clearly alive and well. Today I was looking through the pages of a magazine I purchased purely because it had a nice picture of an Australian manufactured aircraft. This was the “Jindivik”. This aircraft was an unmanned aerial vehicle (UAV). Well, UAV’s are the future of Aviation and the next big step so that would be a good article to include. But here’s the surprise: this issue is dated June 1969 and is Electronics Australia. Australia has actually been looking into this technology for a very long time now. Back to analog electronics and my reason for this letter: inside that magazine, I found a great circuit for a Theremin. I have already built one using op amps with great success but if there’s one thing I really love in the electronics world, it’s simple circuits that can achieve the exact same results. of view, is that the major browser developers have all banded together to ensure that Flash Player will no longer function after 2020 either. So it’s not just a case of ceasing development, but a case that existing content will stop working then too! See this URL: http://siliconchip. com.au/l/aag2 Content developers are going to be forced to adopt the newer HTML5 and WebGL technologies or face their sites not working any more. Sticking with only Flash means it all breaks in 2020. I would expect that well before then Adobe will have added the HTML5 export function to their InDesign product. I hope this info helps plan your Celebrating 30 Years I refer to a tasty little circuit on page 98 that most enthusiasts should be able to build simply on a good old protoboard, requiring only seven transistors and one FET. It should be possible to do this in one day. And who wrote the article and even provided a nice enclosure plan for the circuit? The now legendary Leo Simpson. So thank you, Mr Simpson, for a great circuit which you actually designed around 48 years ago. It’s just as valid for music creators and sound effects enthusiasts like myself today as it was first designed. By the look things, it will remain so for the future as well. Perhaps an updated valve-style version for a new Silicon Chip article might be worth considering! Sean Curtin, via email. Editor’s response: we are considering updating that Theremin design to use commonly available parts and producing a PCB for it. Regarding the resurgence of valve and vinyl equipment, there are few successful migration strategy well in advance. Pete Mundy, Nelson, NZ. Response: it seems unlikely that there will be good HTML5-based alternatives to Flash by 2020 given that they are still lacking as of late 2017. After all, it’s less than three years away. We expect there will still be ways to view Flash content after 2020. However, we cannot expect our readers to jump through hoops to view the online issue. So it seems likely that we will ultimately need to come up with a completely different method for producing and/ or distributing online issues. siliconchip.com.au objective reasons to prefer these over solid state analog electronics and CDs. Solid state amplifiers do a much better job of reproducing the original audio and vinyl records suffer from many problems that CDs don’t, including stylus pinch effects and inner groove distortion, wow & flutter, rumble, dust, scratches, fungal growth, etc. Perhaps the resurgence of valves and vinyl is a justifiable, if misguided, backlash against the decidedly poor sound quality produced by MP3s and other digitally compressed files. It’s certainly hard to consider anything that has been compressed in a lossy manner as “hifi”. Whether or not it makes sense, valves and vinyl do seem to be making a significant comeback. Anyone who wants the best audio quality should build one of our UltraLD series amplifiers and pair it with our CLASSiC DAC and/or a good quality CD/DVD/Blu-ray player. That will beat the pants off even the best record player/valve amplifier combination in terms of pure sound quality. You shouldn’t have to periodically reboot routers Regarding the automatic NBN modem rebooter published in Circuit Notebook, September 2017 (www. siliconchip.com.au/Article/10786); I commend Les for an excellent solution to his problem. He obviously put a lot of thought into his design. But I find it very sad that we as a country are spending over forty billion dollars to have the latest FTTP technology, which has in many parts of Australia been downgraded to FTTN and he has to design a homemade circuit to be able to use it to its full potential. Geoff Hansen, Littlehampton, SA. Comment: the NBN roll-out has had many serious problems but we don’t really think it’s fair for it to cop the blame this time. Broadband modems have always been lousy and the fault lies with the manufacturers. Remember the horrible days of slow dial-up internet? At least the good modems were programmed to “re-train” periodically in order to adapt to changing line conditions. These days, DSL/cable/NBN modems only seem to adapt in one direction. When they detect interference on a given frequency, they stop using siliconchip.com.au it and as a result, over time they get slower and slower until eventually you get frustrated and reboot it. Then it is OK for a while until more random interference comes along and it slows down again. Worse, many of them have dodgy firmware with memory leaks and other problems which causes them to either randomly reset or freeze up and require rebooting. So it really comes down to the cheapness of the modems and their poor programming, not the network itself. Note that there is a mistake in the relay wiring of the circuit. See Notes & Errata on page 104. Observations on Power Supply for Battery Radios Having just read your article on the above excellent project in the August 2017 issue (www.siliconchip.com.au/ Article/10751), I have made some observations. It is commendable that we do everything we can to discourage the wrecking of battery radios and that includes the widely-held misconception that they can only be converted to mains sets. That might have been the only practical approach in the 1950s and 60s but these days, we have ready access to simple components that make power supplies easy and logical. The circuit does not have the resistors and capacitors numbered, to complement the text and overlays. This makes it difficult (especially for the less experienced enthusiasts) to follow. As printed, D3 & D4 in the text refer to D3 & D13 in the circuit! In hindsight, they probably should have been labelled D3 & D4 in the circuit. A good numbering practice is as follows. Where there are two or more distinct sections to a design, one section should be numbered C1, C2...R1, R2...D1, D2... etc. The next section is numbered C11, C12...R11, R12...D11, D12... etc or C101, C102...R101, R102... D101, D102... etc. In the text, where it says “...and so the voltage doubler output will be about 85V, neglecting the voltage drop across diodes D1 & D2.”, I think you intended this to read: “about 45V, neglecting...”. One other point, not mentioned, is that when using the “battery” sockets on the front panel, care must be taken Celebrating 30 Years Want to work for Australia’s Electronics Magazine If you live, breathe and sleep electronics you could be just the person we’re looking for. While formal qualifications are well regarded, don’t let a lack of letters after your name put you off, if you have the experience we’re looking for. The right person will certainly have skills in the following areas: Analog and digital circuit design from concept to completion Circuit analysis and debugging PCB layout (we use Altium Designer) PC software development and embedded programming Operating electronic test equipment Mechanical design But most of all, you’ll have the ability to write interesting articles (in English) describing what you’ve built and how SILICON CHIP readers can reproduce what you’ve done. You will have seen the style of SILICON CHIP articles – you’re almost certainly an existing SILICON CHIP reader. If you have skills in other areas which would help SILICON CHIP appear each month, tell us about them too: skills such as sub-editing, desktop publishing/layout, circuit drawing, photography, image processing, technical support/customer service (via telephone), project management, parts ordering and management, database administration, website design/programming and operating CNC equipment. We don’t expect you to have all these skills – but we’ll help you to develop them as required. You’ll need to be highly self-motivated and able to work well by yourself as well as in a small team. Being able to work to the rigorous deadlines of a monthly magazine is vital. Candidates will be given a six-month trial with a permanent position at the successful conclusion. If you think you have what it takes, email your resume/CV (along with contact details!) to silicon<at>siliconchip.com.au October 2017  5 to ensure that the plugs carrying the +90V and B(-) wires are connected to the correct sockets. Careless plugging can result in the joining link between the plugs actually shorting out the supply with less-thandesirable results. Plug-in batteries came into being in approximately 1940-46; 1939 designs had pin tips and errors could be easily made. Power adaptors were offered by set manufacturers from the late 1940s and 50s, using sockets as you have done. But the accepted way of setting up these sockets was always to feed the 90V to one socket and provide the other with a shorting link to complete the circuit, thus it mattered not which plug went to which socket. Geoff Trengove, Maryborough, Vic. Comment: you are right that the text incorrectly referred to D13 as D4. We’re guessing that D13 was added to the design later, at which point D4 already referred to another component. When we design a circuit, we sometimes renumber components after the final revision to avoid this sort of confusion but renumbering can also lead to errors. Your point with the socket is a good one and the solution you have given is ingenious. If we revisit this design, we will wire the sockets up this way (it could possibly be done as a running change if we sell our existing stock of PCBs). Finally, we think the reference to 85V regarding the output of the voltage doubler is correct. The input is 30VAC, which is around 42.5V peak, hence doubling this (via D1 and D2) should give around 85V. Consider that D1 and D2 are effectively connected across the B- and B+ 90V outputs. Positive feedback from a happy reader I want to say thanks to Leo Simpson as I spoke to him a couple of weeks ago when I rang about a Micromite kit I had purchased that was not working. We worked out a series of substitutions and determined the LCD touchscreen was not working. I received a replacement promptly and all is well now, thanks Leo. Also, the article you published about Incat in the July 2017 issue was great. We were hiking in Tasmania in early April this year and I was wondering if Incat still existed. They seemed to not be all that well-known, even by 6 Silicon Chip the Spirit of Tasmania crew (some of them, at least). It’s great to see and hear stories that Australian companies can and do compete on the world market; even better when they are market leaders! We’re not that backward after all! Whilst on the subject of the magazine, I would like to say that I look forward to receiving my copy of Silicon Chip each month. The only complaint I have is that I read through it soon as I get it then I have to wait patiently another 4-5 weeks to get the next one. Perhaps I should not be wishing away the time waiting, maybe I need to learn some more patience! I’ll try. Another thank you is due here also. A long time ago, I emailed Leo Simpson about the Micromite and how it would be useful if some sort of tutorial could be published as well as the projects so people like me (microprocessor illiterate) could learn and get to understand the BASIC code used to program the devices. You and Geoff Graham have answered my prayers! Geoff Graham deserves a medal for the massive amount of work he has put into the Micromite projects. He is absolutely brilliant and it is fantastic to see him so willing to pass on his knowledge to us not so brilliant. Just one other point about the projects you publish. I love the Micromite projects you have published and have built several of them. This month you published an Arduino audio playback/ recording shield, that’s good. It shows that your magazine is not biased, but as I am addicted to the Micromite I would love to see a similar project for the Micromite, I’m sure I am not alone here. Keep up the good work! George Wundele, Belgrave, Vic. Privatisation partly to blame for SA’s electricity supply woes I agree with your Publisher’s Letter in the August 2017 issue, except for the part about the effect of CO2 on climate which I am not competent to comment on. The energy problems we have began years ago when energy production was privatised. State governments got an injection of money but lost the ongoing return from the production of energy and also lost any control over it. I don’t believe that our SECV would Celebrating 30 Years have shut down Hazelwood without arranging for replacement generation. Can you really envisage any private company building a power station where there is a lead time of five years or so and billions of dollars involved? And then there is the stupid way that the government manages generation where various generators bid their prices and the generation on line is determined from these prices. It’s no way to run a power supply system, in my opinion. Then there is the 100MW Tesla battery being built in SA. This is half the size of one of the 200MW Hazelwood generator units and would last for a bit over an hour before it goes flat! Batteries are worthwhile in domestic situations in order to get better value out of the PV system after the Sun goes down but not appropriate for the state-wide high voltage power supply system. I don’t know what the answer is to our energy problems but I don’t have much faith in the current lot of our politicians to sort anything out. Alex Brown, Ashburton, Vic. Editor’s comment: companies regularly invest billions of dollars in projects which will not have an immediate pay-off but only if they expect to make their money back with a profit. We’re not sure it’s fair to blame privatisation. If you ran a power company, given the present regulatory environment, would you make the decision to invest in a new base-load power station? Micromite serial problem resolved I wrote to you quite a while ago because I was having some trouble communicating with a 44-pin Micromite that I built from short-form kits that you supplied (see the August 2014 issue; www.siliconchip.com.au/ Article/7960). I purchased and built three; two worked fine but with the third, I had to set the FTDI chip to use a baud rate of 32,500 (rather than the correct value of 38,400) to establish communications. At the time, you gave me some suggestions but I was unable to resolve this. I have since gotten to the bottom of this problem. I have determined that the FTDI chips are fine and other USB/serial adaptors gave me the same problem in communicating with that chip. siliconchip.com.au Silicon-Chip--More-New-Products.pdf 1 8/30/17 6:01 PM C M Y CM MY CY CMY K siliconchip.com.au Celebrating 30 Years October 2017  7 Given the fact that I had to set the USB/serial baud rate 18% low, I tried instead setting the Micromite’s baud rate 18% high, to 45,288. This resulted in correct communications at 38,400 through the FTDI chip! I then used the Microbridge (May 2017; www.siliconchip.com.au/ Article/10648) to see if that would program this particular Micromite differently to the PICkit 3. It reported that the CPU ID was wrong and refused to program the Micromite at all. The PICkit 3 reported “Valid but unexpected CPU ID, do you want to proceed anyway?” Answering yes allowed me to program the chip and it verified OK. I therefore decided that the problem must be in the PIC32MX150 chip. Everything else seemed to work OK but the internal oscillator frequency seems to be 18% below what it should be, causing the serial communication problem. I subsequently replaced this PIC32MX150 with the enhanced PIC32MX170, programmed it without a hitch using the Microbridge and tested at 38,400 baud. It now works fine. Ingo Evers, Higgins, ACT. Getting on-soapbox about going off-grid I felt compelled to write after the May 2017 Publisher’s Letter about going off-grid maybe being a bad idea. First, there’s the scam of grid-connected solar. You (and the taxpayers) pay thousands of dollars for the small scale (expensive) infrastructure, yet you are rewarded with only 6c/kWh (in Victoria) for renewable energy. They get a 25% cost reduction compared to what they pay for coal generation, plus they eliminate millions in infrastructure costs due to much of the peak load being supplied by solar. How much did they contribute? Almost nothing. OK, they pay you a few cents for it and they let you use it all yourself if you want. Nice. Who got the better deal here, really? The entire scheme was designed primarily to benefit them while throwing you a few crumbs. This is yet another reason to go off-grid. If you’re going to spend thousands on solar make sure you get 100% of the benefit and they get a fat zero. I do agree with using all your generated energy instead of giving it to 8 Silicon Chip them dirt cheap. But then let’s stop and think what we are doing here. We are basically being forced to increase energy use just so that we don’t give it to them! Is that the way we should be going? We have been trained to consume far more than we really need. Yes, trained. We should be cutting back. Most of us are so addicted to energy consumption that our lives would be miserable without all our modern appliances we think we need but really don’t. If you are willing to cut back on the excesses of our modern living then going off-grid is not that hard to do, nor prohibitively expensive. For example, if you need a clothes dryer, disconnect the heating element. So it takes three times as long to dry the washing – who cares? Now ask yourselves why the appliance manufacturers have not implemented such a huge energy saving initiative. Answer: they don’t want you to get anywhere close to detaching from your dependence and bondage to grid. Need more proof? Look at modern washing machines: many have only a cold water inlet and the default programs all use 30, 40 degree and higher temperatures. That’s an awful lot of energy unless you manually override and select cold each time. How about a project which connects directly to solar panels and has variable output voltage to a heater element? (see www.easywarm.co.nz). I’d say pool heating would be more of an energy issue to most pool owners than a couple of pumps running. For those that don’t have heating, they could extend their season by installing one of these. This kind of project could also be used as a direct connection to electric hot water heaters. Now that would be a real energy saving initiative by Silicon Chip. The power companies would hate you for it though. Robert Hatvani, Noble Park, Vic. Comment: All electric clothes dryers can be set to run at half power but the corresponding increase in drying time means that not much energy is likely to be saved. And trying to dry clothes without any heat input during wet cold weather simply won’t work. On the other hand, most people are aware that cold water washing works well and does save energy. There is Celebrating 30 Years no need to select cold water operation each time you turn on the machine; the setting will be remembered from the last time it was used. Trying to heat a swimming pool with the average domestic solar installation is likely to be a futile exercise – far more power is needed and the pool would need to be covered every night to avoid heat loss. If you want to heat a swimming pool, dedicated roof collectors are the most effective solution. Problems compiling Arduino sketches on a Mac Just a note to advise on the use of a Mac running OS X and the Arduino IDE. As suggested in your article, I installed the latest Arduino IDE software, version 1.0.8.3, running on Mac OS X 10.10. There appears to be a problem with the LiquidCrystal_I2C.h library with this version of the Arduino IDE; the lcd.print(“string” or number) command will not print strings beyond the first (left-hand) character. Repeated lcd.print(char) commands work OK, but of course this is not very useful. I checked the hardware, even substituting an older ATmega328 board I purchased some years ago. I also tried earlier versions of the IDE, V1.0.8 and V1.0.6.13, but to no avail. The printing problem was resolved using the Arduino V1.0.6 IDE. Strings, numbers and assembled strings as you have used in the sketch print perfectly. Using this version of the IDE, however, raises another problem – the included EEPROM library does not support floating point numbers or the 4-byte EEPROM.put and EEPROM.get commands. Since these occur in the NUDGE section of your sketch only, I am not too concerned, but I did rewrite the code there so that if nudging had been performed, there would appear a printout of the new value of CF (once the nudge switch was in the neutral position) while LK1 was in place. That new CF value would then have to be entered and saved in the global declarations part of the sketch, and the sketch recompiled. I can forward these changes to you, if you wish. Since this situation is unlikely to occur too often, it’s not really serious. But any of your readers using the latest version of the IDE on a Mac could be frustrated by the puzzling performance siliconchip.com.au of the I2C print library. Fortunately, the IDE V1.06 for Mac is still available. I am sending you my modified sketch code so that you can provide it to other readers who run into the same issue. Anthony H. Goodman, Worrigee, NSW. Missing text in Radio Telescopes article The article about Radio Telescopes in the August edition seems to be missing a section after the short paragraph in column 2 on page 16. Would it be possible to publish this missing bit in the next edition? It was an interesting article. As an aside, on page 21 column 1 second paragraph, the past tense of “to lead” is spelled “led”. Alex Danilov, Naremburn, NSW. Publisher’s response: Thanks for bringing this to our attention. The mistake happened because there was a slight change to the layout just before we went to press. The missing lines are “… is set at a lower altitude. There is a talk about ALMA by Australian, Anthony ...”. Later HMV valve sets had 457.5kHz IF I am currently burn testing a HMV E43F valve radio. While looking at the article on using a DDS module for IF alignment in the September issue (www.siliconchip.com.au/Article/ 10799), I noticed that the caption for Fig.4 may not be correct. The HMV’s peak response could well be correct if its IF is not resonant at 455kHz. Most HMV IFs in the latter years, including the one I just aligned, are designed for 457.5kHz. Marc Chick, Wangaratta, Vic. Response: you’re right that the HMV 6452 was made with an IF of 457.5kHz. However, given that we’re measuring a peak at 453.6kHz, that suggests it still needs tweaking for the optimum response. Clarification on modifying Valve Radio Power Supply I am very interested in the Mains Power Supply for Battery Valve Radio Sets published in the August 2017 issue (www.siliconchip.com.au/Article/ 10751). The project has some excellent features which I would recommend to anyone involved in this area of interest. siliconchip.com.au What I need to convey is that the overlay as produced in the magazine does not reflect some comments in the article about component numbers. For example, my requirement is usually for 90/45V operation but the lack of component numbering on the circuit or published overlay has caused some difficulty in understanding how this was to be achieved. It took a while but I can now see that the two lower 150kW resistors in the loading chain can produce the 45V if it is supplied through a 470W resistor from the anode of D3. I built a “mock up” of the 1.5V circuit and was pleasantly surprised to see a very low noise level, around 1mV peak-to-peak on this line. Robert Forbes, Forest Hill, Vic. Response: the text on page 39 describing the modifications assumes you have the PCB in front of you, which has the resistors and capacitors labelled. We should have labelled the relevant components on the PCB overlay and circuit diagram to make it more clear. As you surmised, the 470W resistor added is between the junction of the two 220µF capacitors at lower left in the circuit diagram and the junction of the two 220µF capacitors to their right. It essentially provides a low-pass filter for the existing 45V present at pin 8 of transformer T1. For the modification to provide a 4V output for the A+ filament, R1 is the 100W resistor from the ADJ terminal of REG1 to the A- pin of CON1. As you point out, this type of regulator is very good at rejecting 50/100Hz ripple when the ADJ pin is bypassed so the A battery output should have low noise. BASIC as used in Micromite lacks error checking P. C.’s problem with the Micromite code (“Quirks encountered with Micromite tutorial”, July 2017 Ask Silicon Chip) could well be due to a problem with the BASIC interpreter and language. DO and FOR are both block open statements which need to be closed with a LOOP or NEXT statement respectively. These can be nested (placed one inside the other) but once nested, the first LOOP statement closes the last opened DO block irrespective of what the code layout implies. Nice looking, Celebrating 30 Years Helping to put you in Control NFC Temperature Data Logger sealed temperature logger for monitoring temperatures of products during transportation. NFC wireless interface and Windows software for configuration, download and charting. SKU: NOD-052 Price: $59.00 ea + GST Button Control Box The green and yellow pushbuttons have 1 NO contact while the red emergency pushbutton has 1 NC contact. SKU: HEE-025 Price: $27.50 ea + GST IP watchdog monitoring module TCW122B-WD is an IP watchdog monitoring module, specially designed for a failsafe monitor system. A relay is activated if there isn’t an ICMP echo for a certain time. SKU: TCC-004 Price: $114.50 ea + GST Current Transducer Split core hall effect current transducer presents a 4 to 20 mA DC signal representing the AC current flowing through a primary conductor. 0 to 30 A primary AC current range. SKU: WES-0550 Price: $143.00 ea + GST TECO Starter Kit SG2-20HR-D Starter Kit. includes PLC, HMI and programming cable, with a 15% saving on the regular price. SKU: TEC-081 Price: $393.00 ea + GST Mean Well DC-DC Converter 100 W Isolated DCDC converter module accepts 9.5-18Vdc input and gives 24 VDC out at up to 4.2 A. SKU: PDC-010 Price: $83.60 ea + GST LoopPowered Temperature Sensor This is a simple 4 to 20 mA output loop powered temperature sensor with measurement range from -10°C to +125°C designed for monitoring RTU and PLC cabinet temperatures. SKU: KTD-267 Price: $54.95 ea + GST For Wholesale prices Contact Ocean Controls Ph: (03) 9782 5882 oceancontrols.com.au Prices are subjected to change without notice. October 2017  9 correctly indented code will not save you from this error. In some compilers/interpreters, there is no checking as to the correctness of the syntax of the block closing statement, which means a LOOP statement might close a FOR block and a NEXT statement might close a DO block. You won’t necessarily find this error until you try to run the program and it actually reaches these lines (which may not happen very often). The problem with IF is that, depending on the exact syntax used, IF can be a block open statement and END IF is (or should be) its matching block closing statement. In the code as printed, with no syntax checking, LOOP (the last line) closes the IF block (second last line), leaving the DO block still open, hence the error message: “DO WITHOUT LOOP” The fix is to replace the last two lines of code with the following: IF TOUCH(X) <> -1 THEN END END IF LOOP Chris Simpson, Glenbrook, NSW. Comment: it’s true that IF can be a block open statement but it’s also possible to have a single-line IF or IF/ELSE statement and in that case, you don’t need the END IF. The code as published (Fig.6 on page 28 of the February 2016 issue) is correct and will work if entered exactly as shown. But even slight changes or typos can mean that it won’t work and may produce the “DO WITHOUT LOOP” message. Comments on a range of topics Reading Serviceman’s Log reminded me of a recent repair I made. A few years ago, a product came out that was supposed to scare away snakes. I thought they were outrageously priced at the time, so I didn’t buy one. Instead, I thought about how they might work and made one from an old tape recorder motor with a small nut cable tied to the shaft, to unbalance it, and powered from a small solar panel from a junked solar light. The whole lot was cable-tied to a star post and shook two or three posts either side of it when in operation. Two years later, it is still working. Anyway, a friend recently turned up with a set 10 Silicon Chip of solar-powered snake scarers and requested that I repair them, as they had cost $50 each a few years ago. Upon disassembling one, I found it consisted of a single 300mA Nicad cell, a small PCB with an IC and some discrete components, plus a USB socket where the transducer plugged in. Further testing revealed that the batteries were all dead and replacing these brought one to life, emitting a rather annoying chirp or three every few minutes or so. Plugging and swapping around the working transducer revealed that the other units were also now operational but the transducers were faulty. Pressing apart the plastic spike revealed a small 1.5V DC vibration motor screwed to it. What a let-down and a rip-off for $50 each! Anyway, I had a couple of old code practice buzzers in my junk box that were exactly the same as the dead ones. It only took a matter of minutes to replace them and then all three worked. I estimate the cost of the components in these products to be less than $5. Do they work? I don’t know but I haven’t seen a snake in the area where I installed my homemade one. In the Ask Silicon Chip pages of the June 2017 issue, on page 108, P. W., asked about synthesising stereo from a mono recording. I re-record old recordings, some from the 1800s and have found that using the free audio program Audacity, a reasonable sense of presence can be created easily by doing the following. After getting rid of the scratches, pops etc, select the entire recording and copy it to the clipboard, then make an empty second channel. Expand the time scale out so you can see a 50ms interval. Select a point between 10 and 20ms from the start of the file and paste the copy of the original track into the second channel. This creates a stereo expansion effect with the second channel delayed from the first by about 20ms. It’s simple and quite effective. All you need to do then is rename the original channel to “left” and the new one “right” (or vica versa). Note though that with some music, you may find the resulting effect disorientating when listening with headphones. Leo Simpson’s comments regarding solar tariffs (Publisher’s Letter, May 2017) stirred some memories of over 20 years ago when all this grid-feed stuff was being thrown around. Celebrating 30 Years How binocular beam strain gauges really work The Circuit Notebook item entitled “Measuring weight using Arduino” in Silicon Chip, April 2017 (www.siliconchip.com.au/Article/ 10618) caught my eye because I have in mind a project involving a strain gauged pressure transducer. With a little modification, the circuit and programming will provide a useful starting point. However, incidentally, I noticed that there is somewhat of a misdirection in the brief explanation of how the strain gauges and the “binocular” beam function as a load cell. A side view diagram of the binocular beam was provided along with the circuit. The accompanying text explained that: “When a load is placed on the free end of the beam, the beam flexes and the upper two strain gauges are in tension and the lower two are placed in compression.” For a beam of the proportions shown that is not correct. The given explanation would apply if it were a solid beam and if the strain gauges were connected so as to sense the overall bending moment in the beam. However, in that case, the output signal would vary substantially depending on the load position. That is because, the further the load is from the support, the larger the bending moment in the beam, the higher the bending stress and the higher the resulting signal from the strain gauges. Such sensitivity to load position is undesirable in weighing applications such as platform scales. Rather than sense the overall bending moment, the binocular beam is actually configured to sense the overall shear force in the beam. The advantage of that approach is that the shear force in the beam only depends on the magnitude of the load, not its distance from the support. So how does the binocular beam sense the overall shear force? The key is to understand the function of the four thin sections of the member that remain above and below the two “binocular” holes. These thinned-down sections effectively operate as flexural hinges, siliconchip.com.au albeit with a certain elastic stiffness that resists rotation of the “hinge”. Once they are recognised as hinges, it can then be appreciated that the binocular beam actually constitutes a parallelogram mechanism, with the four flexural hinges located at the corners of the parallelogram. The notional parallelogram is illustrated by the green dashed line in Fig.1, which shows the binocular beam with exaggerated deflection under load. The beam’s mechanical resistance to being distorted in this way is mainly dependent on the elastic bending stiffness of the four flexural hinges. When a load is applied, substantial bending stresses are induced in the thin sections that form the hinges and that causes change in the electrical resistance of the strain gauges that are bonded to the hinge sections. (A strain gauge’s electrical resistance varies according to the strain, that is the stress induced contraction or extension occurring along the surface to which it is bonded.) The strain gauge on the top of the beam, nearest to the support, will indeed undergo tension strain, not so much from the overall bending moment in the beam but rather due to the local bending of the hinge. However, in a beam of the proportions shown, and contrary to the explanation given in the text, the gauge on the top of the beam nearest to the loaded end will undergo compression strain due to the dominant effect of local bending of that hinge. The opposite pattern applies to the gauges on the bottom of the beam. It is easier to understand that by reference to Fig.1, and remembering that when a part of a structure is deformed by bending. In this case, the deformation is most pronounced at each of the flexural hinges; the bending causes tension stress and accompanying tension strain on the outside of the local bend and compression stress and strain on the inside of the bend. A key feature of the parallelogram mechanism is that when the free end of the beam moves down under the effect of the load, it stays parallel to the fixed end of the beam, which in a weighing application, ideally should be horizontal. siliconchip.com.au Importantly, that also means that the bending moment and stress induced in the flexural hinges and hence the signal is not affected much by the where the load is placed on the load end of the beam. That concept can be appreciated by thinking about the work done on the structure by the load. In an ideal device of this type, the gravitational potential energy lost by the load when it deflects the beam downward must equal the elastic energy stored in the distorted flexural hinges. Thanks to the parallelogram mechanism, for a given degree of distortion of the hinges, the load moves downwards by the same amount irrespective of the load’s exact position on the end of the beam. The corollary of that is that a given load placed anywhere on the load end of the beam will produce the same amount of distortion of the hinges, irrespective of load position, and hence produce the same signal from the strain gauges. That leaves the matter of the unwanted signal due to overall bending in the beam, which indeed will vary with load position. Two factors help reduce this unwanted signal relative to that caused by the parallelogram action, which senses the overall shear force. The overall bending causes tension generally along the top of the beam and compression in the bottom. However, one strain gauge on the top is connected into a tension leg of the Wheatstone bridge and one is connected into a compression leg, which means that the signal from overall bending is partially cancelled. The cancellation is only partial because the overall bending moment varies along the beam, so the pair of gauges at the binocular hole near the load end generates a smaller signal from overall bending than does the other pair. The closer the binocular holes are together, the better the cancellation but that reduces the parallelogram action so it is a design trade-off. The other factor that helps reduce the unwanted signal from overall bending is that the bending of the hinges by the parallelogram action induces much greater stress than is induced by the direct forces in the hinges caused by overall bending of the beam. Thus, the peculiar form of the beam, together with the electrical arrangement of the gauges, allows it to sensitively measure the overall shear force in the beam, and as far as possible, excludes and minimises the effects of the overall bending moment. The parallelogram mechanism facilitates that by causing the overall shear force in the beam to manifest as localised high bending stresses and strains at the four flexural hinges. A limitation of these devices is that they can be permanently deformed by overload. This can be avoided by having a mechanical backstop located suitably close under the load end of the beam so as to limit excessive deflection. Descriptions of other variants of this type of load cell, which have many diverse applications, can be found here: www.sensorland.com/ HowPage005.html Thanks for the great magazine. I have been reading it since its inception. Nigel Beal, BE FIEAust MIStructE RPEQ, Chapel Hill, Qld. Editor’s response: thank you for a highly enlightening letter. The erroneous description of the operation of this type of load cell was introduced during editing; we can’t lay the blame on the contributor for this one. Fig.1: side view of binocular beam showing exaggerated deflection under load. Celebrating 30 Years October 2017  11 How common are rat attacks on cars? I was completely immersed in reading the Serviceman story from B. Y., of MacKay, Queensland regarding the encounter with the dreaded scourge of the gnawing rat (August 2017). It’s almost a shame that it was in this section and perhaps, this letter may give that section more prominence. I have been a victim of this scourge not once but on several occasions with consequent huge repair costs and inconvenience. Maybe no-one really wants to hear my hard-luck story and if it were only me and B. Y., then fair enough. Except that when ever I bring this subject up in casual conversation, I never cease to be amazed at the sheer number of people who have similar tales of woe to tell. I suspect that this is just the tip of a very big iceberg. Apart from being a curse to the vehicle owner, I believe that this is a hidden and serious public safety issue. This is especially the case with the pervasiveness of modern embedded electronic control systems – drive-bywire (electro-servo throttle, all-electric steering, electronically controlled braking, etc). With the increasing number of hybrid and all electric battery powered vehicles on the road, the risk of fire could be significant, which would be disastrous for those with internal garages! I am sure the insurance industry is sitting on a bundle of statistics that would give the full breadth of scale to this issue but I am particularly incensed as to why, in the 21st century, car manufacturers seem completely incapable of designing and building vermin-proof vehicles without resorting to the application of chilli sauce! A few large public liability law-suits might get the ball rolling in the right direction! In the meantime, I would be most interested to hear more tales from other readers on this subject but in particular, stories from automotive electricians who by far, would be most likely to encounter the worst of it. You know, I’ve always believed that if you could put a rat in charge of the training of a US Navy Seal you would produce a truly unbeatable, “universal” soldier! Andre Rousseau, Auckland South, New Zealand. We engineers argued for a net tariff, where the meter basically went fowards when you were using power, and backwards when you were exporting power. At the end of the billing period, if the meter was positive, you paid up, if negative, the supply authority paid you, at the same rate. This is called parity pricing and would have been a fair way of paying for your energy. This was rejected by the powers-that-be because they wanted to keep track of the energy being generated to raise the solar renewable energy certificates (RECs) and metering at that time didn’t allow recording of input and output power independently. Also, the case for different tariffs for grid feed-in was raised. This is of necessity a simple explanation of a very complicated system and now we have the situation where you may get paid say 6c/kWh for your power, whilst getting charged something like 50c/kWh to use energy from the grid in peak periods. I have seen some systems that were not correctly wired where the customer was getting charged 50c/kWh for their own generated power, whilst getting only 6c/kWh for power fed into the grid. This could explain why some people find their electricity bills are still so high, even after the installation of a solar system. I suspect that this has been done on purpose in some instances, to make more profit for the retailer. As to Leo Simpson’s comments regarding battery systems and maintenance costs, yes, the initial purchase cost of batteries is high but maintenance costs are nearly zero with sealed deep-cycle batteries and correctly adjusted regulators etc. Indeed, 20+ years’ life from this type of battery is the norm, at 10% maximum discharge. I suppose it comes down to demand; if people want to have batteries for secure power (and load shedding at peak times to save money from the demand tariffs), then the market will respond appropriately. Finally, I would like to point out that a diesel-electric submarine will always be quieter than a nuclear submarine, and that’s because the nuclear submarine needs coolant pumps operating continuously to cool the reactor. Nuclear subs are also detectable by other methods, such as looking for a place in the ocean where there is more hydrogen in the sea water than there should be (the process of making oxygen from sea water leaves hydrogen and a waste product which is dumped back into the sea). Other methods for detecting subs are looking for a quiet spot in the ocean, as modern submarines are insulated to absorb noise, analysing the sea water to detect traces of shaft seal oil and looking for a large, moving magnetic anomaly in the sea. Peter Laughton, Tabulam, NSW. Editor’s comments: while you are right that a good quality battery bank could last 20 years or more, limiting depth of discharge to 10% means that if your maximum daily consumption was say 5kWh (including aircon, fridge, laundry, etc) you might need a bank of around 50kWh to keep the average depth of discharge at 10%. You may possibly need more to account for multiple days of poor weather, which would make for a huge and very costly battery bank. Replacing this every 20 years (or so) would work out to a very large maintenance cost when amortised over that period. You would also need to factor in the panel life-span (maybe similar to the batteries) and inverter(s), which can fail after just a few years’ service. These all need to be factored into a proper cost/benefit analysis. By the way, one reason lithiumbased storage batteries like the Tesla Powerwall are so attractive is that they can handle a much greater depth of discharge without shortening their lifespan dramatically, so you don’t need nearly as much capacity, making them more cost competitive; even occasionally flattening them should not harm them. Regarding your comments on nuclear submarines, some modern submarine reactors can operate at low power levels without active coolant pumps so they aren’t necessarily much noiser SC than a lurking diesel sub. 12 Silicon Chip Celebrating 30 Years siliconchip.com.au