Fullscreen HTML5Med Res (??MB)HTML4

SERVICEMAN'S LOG Fixing a guitar amp is an enjoyable task Which would you rather do, solve a problem with a laptop PC or fix a large guitar amplifier, and in the process maybe play a few riffs? It was a pretty easy choice and involves work which is almost my hobby. I’ve said it before and I’ll say it again; the computer repair business is a sunset industry. The golden years of computer-repair guys skilfully assigning IRQs and IP addresses are long gone and those of us trying to eke out a living doing computer work really only have two choices: give it away altogether or diversify into a similar trade and hope that we can make some use of the skills and tools we’ve amassed over the years. To that end I’ve started taking on musical instrument and amplifier repairs in an effort to shore up the bottom line. I also assemble kits and troubleshoot projects for builders who have trouble getting their stuff working. I’ve done this sort of work as a hobby for 58  Silicon Chip the last 40-odd years anyway, so my thinking was that I might as well go ‘pro’ and try to make a living out of it. When I say musical instrument repairs, I’m not just talking guitars, although as a guitar player, naturally that has been the focus of work I’ve done previously. However, since I’ve Dave Thompson* Items Covered This Month • • • Guitar amplifier repair Car battery charger Westinghouse oven repair *Dave Thompson runs PC Anytime in Christchurch, NZ. Website: www.pcanytime.co.nz Email: dave<at>pcanytime.co.nz been advertising as doing this type of work, I’ve had several different instruments through the workshop, from keyboards to saxophones as well as the usual busted guitars, faulty amplifiers and effects pedals. This variety makes things very interesting and to be honest is a welcome respite from the same old computer gripes I’m more used to dealing with. Most amplifier issues I’ve encountered revolve around flaky valves or siliconchip.com.au dodgy input sockets. So tightening up the sockets and replacing the valves is often all that’s required to get things humming again. It isn’t surprising that input sockets give out; guitar cables typically have 6.5mm mono plugs at either end and these can exert an awful lot of strain on the sockets mounted in the chassis of an amplifier. They can get especially strained when the guitarist gets carried away and runs out of cable, or tries those fancy moves where the player throws the guitar back and over their shoulder, relying on the strap-locks to hold everything together until the instrument completes the circle and ends up back in the playing position. YouTube is full of videos where this manoeuvre fails, usually spectacularly, with the guitar either flying off out of shot and landing off-stage somewhere or worse, ending up taking out one of the other musicians on-stage or tangled up in the drums and cymbal stands. Getting smacked with a flying guitar is not something to be brushed off lightly; it is only luck that none have hit me over the years! This is exactly why I didn’t attempt any of those showy tricks as a guitar player. For one, I didn’t fancy two grands’ worth of my guitar sailing through the air to the inevitable (and expensive) smash-landing and two, it just looks stupid, whether the player pulls it off or not. If the cable suddenly runs siliconchip.com.au out during these stage shenanigans, it tears at the sockets at both ends and you end up damaging the amplifier and the guitar. Input sockets (and speaker sockets, many of which are also simple 6.5mm mono sockets) get a hammering even in normal road use, so tightening or replacing these and replacing dead or dying valves are the sort of breadand-butter jobs that keep guitar and amplifier repair guys going. Occasionally though, there is a problem outside the square and it is these jobs that make the day more interesting. A few weeks ago, a client brought in a 6-year-old solid-state 100W guitar combo amplifier complaining of two faults; one was very noisy controls and the other was a dodgy reverb effect. As part of my job booking-in procedures, I powered the amp up while he was there, partly to assess these problems so we both know exactly what I am expected to fix and also to make sure there aren’t any other problems the client might have forgotten to mention. I’ve been around the block too many times to fall for those old: “well, it was running perfectly when I dropped it off to you” routines. The best way to make sure there are no surprises is to fire it up and take the time to check it properly. I can recall a few instances over the years when I went to run up a machine and the client suddenly remembers there are other, more serious faults. Nice try, but not on my watch. After making sure the volume controls were all set to minimum before switching the amplifier on, a wise precaution with all solid-state amps that typically power on instantly, I flicked the switch. It was very quiet for a guitar amp, even with the master volume controls wound up half way, but merely touching one of the two channel volume pots caused some very loud and aggressive-sounding static to come from the on-board 12-inch speaker. This wasn’t just some minor crackling; this was the sort of boneshaking, full-volume amplified noise that you just knew could do some serious damage to the speaker or even the output components. The control, labelled ‘Growl’ on this particular amp, also didn’t feel right and would likely need to be replaced. The only way we could get it to settle down was to isolate that input channel and use the second input. That would definitely need to be looked at. While it was running properly I wound in a bit of reverb to test that effect and the resulting sound was, well, just not right. Usually, the builtin spring reverb circuits on guitar amplifiers give reasonable effect depth and sound output quality but this one sounded like there was something physically wrong with the reverb itself, being very muted and with a muddy sounding output. While some modern amps boast numerous reverb sound types as part of a digital effects chain, older and more traditional methods involve the use of a spring tank, typically mounted in the bottom of the speaker cabinet. The theory of how it works is relatively simple; a small transducer sits at each end of a 12-inch long (300-mm) spring. These transducers work somewhat like the voice-coil of a speaker, with a wire coil wrapped around a suspended and movable centre core. The body or coil of the transducer is physically fixed to each end of the tank while the spring attaches to the movable inside parts. When an audio signal is fed into the input transducer, the sound is converted to movement by the transducer and these physical waves travel back and forth, up and down the spring and produce corresponding signals at the other end, the output of the spring, March 2017  59 Serr v ice Se ceman’s man’s Log – continued which are then fed back into the amplifier and mixed with the original signal. Since the signal that travels along the spring is somewhat delayed compared to the original, and it has several reflected components as the sound waves bounce up and down the spring, a realistic room reverberation effect is produced. Mixing in more of the delayed signal increases the overall depth of the effect and while it sounds quite primitive, the system works very well; good spring reverbs sound remarkably natural and are often preferred over digitally-created reverb effects. Once again, YouTube has some very interesting videos of DIY spring tanks made from speaker voice coils and all manner of springs, including a very large one made from a Slinky! Editor’s note: for a comprehensive description of spring reverberation, have a look at the project article in the January 2000 issue: www.siliconchip. com.au/Issue/2000/January/ Spring+Reverberation+Module One of the main disadvantages of the spring reverb is that the tank system is somewhat microphonic. That is, bumping the amplifier with reverb dialled in on the controls usually results in a corresponding “boing” from the system as the lightly-tensioned springs are 60  Silicon Chip physically moved about in the tank and strike the sides. This usually isn’t an issue, as most instrument amplifiers sit on-stage and don’t usually get whacked by anything (except when the guitar player tries that flying guitar trick!). Whatever was causing this amp’s reverb issues, I’d have to dig deeper and look into it. Most instrument amplifiers are heavy beasts, especially the combos (those with built-in speakers) and this one was no different. I’d conveniently forgotten about that side of the job! Oh well, it’d give me a good workout lifting these things up and down from the workbench. Removal of the amplifier part of it is relatively simple; four long screws hold the metal chassis in and these go down through the top of the cabinet. A portable drill-type screwdriver is a necessity when removing these long screws. Once removed, the chassis slides out to the front of the cabinet. Inside is what you’d expect from any highpowered audio amplifier. The heavy bits are the power supply transformer, which in this case was a large toroidal type, preferred for their lower hum signature, and a rather significant heatsink, required to keep the output amplifier cool under heavy use. Being solid-state, there are no large output transformers like you would get in a valve-based amplifier. This is why solid-state amps are often significantly lighter than their valve counterparts. You may not think it makes much difference but to a jobbing musician, who has to pack his or her own gear up and down of a night, it can make all the difference. Lugging 50 kilos of guitar amp around at 2am, especially after a few cold ones with the bar owners, is not the rock-and-roll ideal. This is why the Rolling Stones need five jumbo jets – to cart all their guitar amps and other gear around! In this amplifier (back in the real world), a single PCB held all the relevant components. All the potentiometers and input sockets were mounted along the front edge, while the output and switch-pedal sockets were mounted along the back edge. The preamp is typically mounted on this board as well and depending on the amp’s size and architecture, this board can also hold the output transistors or modules as well. In this case however, the output module was mounted on its own small PCB and this was secured to the large aluminium heatsink by a couple of small bolts and copious amounts of heatsink compound, which appeared to have been applied with a trowel. Why the people assembling these things or the quality-control engineers in the Chinese factory can’t take a bit more pride in their work is one of the reasons they are so behind the eightball in global engineering standards. While in this case it is purely cosmetic and doesn’t have any effect on the sound or operation of the amplifier, it does make a difference to me. My thoughts on the noisy potentiometer would be that I’d hit it with some contact cleaner and if that didn’t clean it up, I’d simply replace it. As the chassis comes out all in one piece, the controls are all still in place and labelled so locating the suspect pot was easy. It also didn’t take much skill to see the cause of the issue. The back cover of the pot had parted company from the front, making it next to useless for controlling anything; more like whimper than Growl! Surprisingly, all the pots were high-end components and not the bargain-basement types I was expecting. Perhaps my assumptions siliconchip.com.au were a little harsh on this easternmade amplifier. I have to say the PCB and internal components were very well-made and professionally wired up, using bestpractices to reduce hum and interference. A replacement pot for this would likely cost a few dollars, so as always, I looked for another solution. I remember back in the day taking pots apart when they got a bit noisy to clean them out; this was before the widespread use of aerosol contact cleaners and besides, my pocket money didn’t quite stretch to such luxuries. These pots, like those of yesteryear, were assembled and held together with four clasps that are part of the back cover. When mated with the front half of the pot, these clasps are then folded over to hold the thing together. This pot looked to have taken a bit of a hit, which had driven the knob and shaft backwards into the back housing and popped a couple of the clasps clear. This is why it made terrible contact, as there was no tension holding the wiper to the carbon track in some places, and too much in others. No wonder it didn’t feel right. I eased the remaining clasps clear and pulled the back free. The shaft had pushed back through, popping a circlip, with only the knob itself stopping the shaft and wiper from coming through further. I pushed back on the shaft and with a bit of pressure, forced it back into position. The circlip clicked into place and the control now moved quite smoothly. I gave all the controls a good squirt of cleaner while cranking them around; easy enough to do when the openings are accessible inside the case. After sitting the amp on the top of the cabinet and wiring up the speaker and a power cable, I fired it up and touched the pot. Nothing, that is, no noise. I plugged in a guitar and gradually wound it up; the static had once again become a growl. A quick check of the other controls confirmed they were all functioning correctly. That was one issue down; one to go. With the amplifier chassis out of the cabinet, accessing the spring tank was a lot easier. In this amp, as in many others, the tank is screwed directly to the floor of the cabinet and is enclosed in a leatherette bag, mainly to keep out dust, cigarette butts, roaches, beer and broken glass. This one was held in with siliconchip.com.au two rather long wood-type screws and was easily removed. Two shielded cables connect it to the rest of the amp via RCA plugs and sockets; one input and one output. The leads protrude through the folded flap of the bag and with the flap open the tank slides straight out. Once again, it didn’t take a brain surgeon to spot the cause of the problem. Half way down the top of the tank was a large dent. When I turned the tank over, I could see the springs were fouling on the dent and this is why the reverb sounded a bit weird. The bottom of the tank is open, and as I didn’t really want to disturb the transducers at either end, I simply held the springs apart either side of the dent with a small piece of plastic cut from an ice-cream container (what would we do without them?). From there it was a simple panel-beating job to flatten the top of the tank and restore sweet reverb once again. The open back of a combo amp is an inviting repository for power cables, guitar leads, effects pedals and other gig-related detritus. Someone had dropped something a bit too heavy into the space and had impacted into the top of the tank. I advised the owner to be careful of what he carried in there from now on. Rock and roll! Car battery charger When switchmode power supplies fail, they can often generate a string of faults. K. G., of One Tree Hill, SA methodically tracked them down in a faulty battery charger that came his way . . . This repair job involved a 12V, 14A battery charger with a switchmode supply circuit. Designated model CC1214, it was assembled by Wialki Electronics in Perth, WA and was about the size and shape of a PC power supply. The internal PCB was branded MeanWell model ESC-240N-R7. This brand is frequently seen on power supplies and their website shows a huge variety of models and types, ranging from open frame units to complete bench supply units. This particular unit was bought on spec by a friend of mine at a garage sale, the seller advising him at the time that it didn’t work. Not much money changed hands and my friend subsequently opened the unit up, hoping that it might be an easy fix. He discovered that the mains fuse had blown but when he replaced it and applied power, the fuse immediately blew again. That was as far as he was prepared to go with the investigation, probably due to the high voltages which he knew existed in this type of power supply. And so he handed it on to me saying I could have it if it was of any use to me. I’ve had quite a bit to do with switchmode power supplies (SMPS), mainly involving modifying cheap PC supplies to deliver a single output of 13.6V at 20A or so for amateur radio transceiver use. And over the years, I’ve accumulated a few test equipment items which make working on these units easier and safer. These include a variable auto-transformer (or Variac) and an isolation transformer which enables the negative side of the highvoltage DC supply in an SMPS to be grounded. This is a great help if you want to look at waveforms in that part of the circuit with an oscilloscope, for example. Another useful device is an electronic load. Mine is home-built and will sink 50A or more for short periods. I also have a plastic box with a 40W incandescent light globe mounted on it, wired so that it can be placed in series with the mains supply. A switch is provided so that the globe can also be shorted out, allowing the full mains supply to be applied directly to the power supply as normal. With the globe in series with the mains, you can tell immediately if a fault in the power supply is causing a high current to flow. The globe comes on at full brilliance with a dead short. This saves on fuses and possible damage to other components. Getting back to the charger, my friend had removed both its lid and the screws holding the PCB inside the case (he had thoughtfully put the screws in a zip-lock plastic bag). With the unit on the workbench, I could see that the quality of construction was of a high standard, with a double-sided PCB and good quality components. This is in contrast to the average PC power supply made to a much lower budget. I began by checking the fuse and it was indeed blown. I then replaced it and connected the device to the mains via the aforementioned light globe unit. As soon as I applied power, the globe went to full brilliance, confirming the presence of a short circuit. March 2017  61 Serr v ice Se ceman’s man’s Log – continued Westinghouse GGP475WNG gas wall oven Recently, John W., from Hillarys, WA, was looking forward to freshly baked muffins but the cook reported that there would no muffins unless he could do some repair magic. He managed to conjure up a cure . . . When I got the call from my daughter about her non-working oven, I thought it would be as simple as turning the power off for 30 seconds to reset the electronics but it was not to be. I then pulled the oven out from the cupboard, rested it on a chair and removed the top cover. I found a circuit diagram on the top cover that seemed rather simple except for the section marked “ignition module”. This was a small PCB with a microprocessor and looked rather complicated for the job it was supposed to do. The board was labeled Tytronics DSI230 so I investigated on the net to find some basic information but could find no circuit diagram. Using the circuit from the top cover, I proved that the thermostat was working and that it was providing 230VAC to the thermostat pin on the PCB when the oven was turned on. The other terminals on the PCB went to mains Active, Neutral and the gas solenoid. After removing the PCB I traced out some of the circuit that was connected to the input pins and found that the 230VAC was fed via two separate capacitors, 100nF (C12) and 1.5µF (C1), to a switchmode power supply which provided 10V DC. A partial circuit is shown in Fig.1. Fig.1: This partial circuit of the supply on the microprocessor PCB shows that the DC supply could be derived from two capacitors, depending on whether the oven was in standby (C12) or operating (C1). I removed the mains plug from the wall socket and waited a minute or so for any capacitors to discharge, though with a short-circuit in evidence not much voltage would have been applied to any capacitors. I then gave the unit a close visual examination. The first thing I noticed were bulges in the tops of the two low- voltage electrolytic capacitors in the charger’s output section. In each case, the bulge wasn’t severe and there was no sign of leaking electrolyte but they were clearly faulty. I removed them and tested their ESR (effective series resistance) using my trusty “Electronics Australia” ESR meter. They each gave an ESR reading Servicing Stories Wanted Do you have any good servicing stories that you would like to share in The Serviceman column? If so, why not send those stories in to us? 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. 62  Silicon Chip When 230VAC was applied to the oven there was no DC present from the supply but when I bridged across to the thermostat terminal, there was 10V across the 1000µF capacitor and a LED was flashing, possibly indicating an error code. So should there be 10V when the oven had power applied or did the circuit only require the thermostat input to be live? I measured the value of capacitors C1 and C12 to find that C12 which was marked 0.1µF was in fact only 9nF. Thinking this must be the problem, I replaced C12 and put the PCB back in the oven. But I still had no gas valve operation or spark to ignite the gas. I decided that the muffins would have to wait and went out for a coffee with the family. Next morning I rang Westinghouse and found that a replacement board was $180 with a wait of two weeks; not acceptable. I then found a business that advertised secondhand oven parts so took the board there and purchased another for $110, on condition that if it did not work I could bring it back. Well it did not work and on the next trip to the shop I brought home three boards to see if any of them would work. One did work, so I now had a working oven and returned it to its spot from the middle of the kitchen. I resolved to return the two of about five times the expected value and so they were replaced. Then I did another quick test, although I didn’t really expect the short circuit fault to have been cured. Sure enough, the globe again lit to full brilliance when the battery charger was powered up. Next on the list of suspects were the two high-voltage switching transistors. They can be tested in-circuit but I like to remove them so that no other parts can confuse the test. It was easy enough to remove them and I then checked them on my semiconductor tester and this indicated a “short circuit” between all three leads on both transistors. siliconchip.com.au Above: this part of the circuit diagram located on the top cover of the Westinghouse oven was used for troubleshooting. Left: this microprocessor board seemed to be more complicated than needed considering its simple functions. Capacitor C12 is the grey block above the 5-way connector while C1 is the large blue block immediately to the left of the connector non-working boards to the shop the next day but overnight the penny dropped after seeing the oven working as it should. I realised that when the mains was applied the circuit was activated via C12 and the micro performed some tests on the circuit including, as I found, testing the DC resistance of the gas solenoid. The result of the tests was displayed by the LED flashing on the PCB. I had an idea that I might have a bit more luck with one of the boards if C12 was a common problem. I rigged up a test circuit with a 220W resistor instead of the solenoid and a temporary spark plug. I then replaced C12 in each PCB. The first one did not work but the second one was a success so I went back to the shop and they gave me my $110 back in exchange for the working board. I had a chat with the owner and found that he had another seven such boards and he did not know if they were working, so I measured C12 on each one and found them all to be low, under about 30nF. I took them home and tested them with a good capacitor tacked across the faulty one and found that five of seven worked after C12 had been replaced. The owner of the shop was delighted that he now had six boards that were tested and working and he paid me a nominal sum for replacing five capacitors. So my oven was repaired for free and I made a bit of pocket money along the way. It seems that there could be a lot of these units from gas ovens and heaters that are being thrown out simply because a $1.00 capacitor has become faulty. The transistors were both type 2SC3320 and a search on the net revealed that this device is rated at 400V and 15A, a very conservative current rating for this power level. I didn’t have any exact equivalents but I had some with a 400V 9A rating that had been salvaged from a 300W PC power supply. This current rating was still ample and so I decided to give them a try. I soldered the replacement devices into the board but initially left off the heatsink, as I planned to run the power supply at only a light or with no load until I was sure that it was working. Unfortunately, when I applied power, the lamp again immediately went to full brilliance, so there was still a problem lurking somewhere. The next thing to test was the bridge rectifier, something I really should have checked before previously applying power. As it turned out, this single package device had a dead short across its AC terminals. I replaced it and checked the resistance between the AC terminals. This was initially low but quickly rose to a higher value as the electros charged, so that was encouraging. 12V was then applied from a bench supply to the mains input terminals. The power supply also passed that test and I felt justified in applying mains power again but still with the light globe in series. This time, the globe lit only briefly and the charger’s DC-powered fan began running. I checked the output voltage and it measured 13.8V, so the unit was back up and running again. A check again on the net revealed the original transistors were available quite cheaply and so I ordered a pair. My aim was to return the battery charger to its original state with the conservatively-rated devices rather than leave the lower-rated devices I’d substituted to get the unit running. The replacements duly arrived and were installed, after which I gave the unit a thorough test before putting it away for future use. SC siliconchip.com.au March 2017  63