Behringer Xenyx 802 Mixer Repair

A customer brought me two of these to repair – they’d bought them in the US (110 Volt AC supply), and when they got to NZ plugged them into an ADAPTOR (not a step-down transformer!) and then into our 240 Volt AC supply. Er, bang! Nothing to be done with the power supplies (they’re 18V-0-18V AC output), the primaries were toast. I found a replacement at Music Works, PSU3-SAA. There was a burning smell when I plugged it into the first mixer, so I guess the damage wasn’t limited to the wall-wart. I tested components but couldn’t find anything broken, so I CLEVERLY plugged the power supply in again and held my nose close to the PCB. In the middle of the board next to the phantom power switch I found a transistor, MPSA06, which was getting very hot. There was a Zener diode next to it, BZX55C-47, which had cracked. I found replacements for both. I used a BC639 for the transistor (different pinout, check before installing it). I had to buy 250 diodes for $11 as that was the smallest quantity available from RS Online, so if you need one, NZ$0.06 each plus postage. The through-hole resistor next to these components (2.2k Ohm, I think) was fine. After I’d fitted the new components I powered it up, no smells and the power LED lit, but the phantom power switch wouldn’t light the LED – or produce 48 Volts at the mic XLR sockets. I spent ages trying to find the problem, and after realising the other mixer was working fine I dismantled it and compared voltages between the two mixers to try and zero in on the problem…but it was soooo obvious! There’s a 10 Ohm SMD resistor on the underside of the board between the transistor and resistor (I think!) and it had gone open circuit. I scavenged a good one off a TV board and everything worked. Sorry I can’t be more precise, I had to get everything back together for the customer to collect so didn’t have time to take pictures or make better notes. But hopefully this will help if you’ve fried your power supply and the mixer, particularly phantom power, is still not functioning after replacing it.


Hewlett Packard HP41CV

I’ve been trying to get this Hewlett Packard HP41CV calculator working for a customer. I got his Hewlett Packard HP41C working no problem – I cleaned the contacts (they weren’t corroded and the posts were intact) and when this failed I replaced the two electrolytic capacitors. These calculators have a common fault – everything functioning correctly relies on the contacts in the case being held together by screws that go into plastic posts. These plastic posts crumble over time, everything parts company and the calculator stops working. I’ve faffed around with this one for ages. The battery contact posts on this one have crumbled away to nothing, there’s only a tiny remnant into which I’ve tried to screw slightly longer, thicker screws without success. The battery contacts were corroded and I’ve used copper tape and solder to replace one which was rotted all the way through, and a couple of the small contacts on the memory module flex. The keypad posts are a bit cracked, but seem OK. I realise this isn’t quite the full Monty when it comes to HP41 repairs – the internet is full of people reconstituting the posts with plastic shavings and soldering irons, replacing them with lengths of plastic etc. etc. I didn’t want to go the whole hog until I knew if the calculator was even working. I’ve managed to produce weird, partial letters in what look like hieroglyphic patterns and sometimes it comes to life and works properly for a bit. My customer bought it cheap because it had ‘died’. Tonight I was messing around with it for the umpteenth time and found a point on the case where, if I pressed it, the calculator came to life and worked properly. I’ve tried zip ties on the case before, but tonight used a combination of one longitudinally from top to bottom right and one laterally near the display. Bingo! It works properly, turns on and off, resets, does everything (everything I know how to do, which is nothing compared to the potential uses of this calculator).  Here’s a picture:


Ajit, the owner, thought this was funny and may post it on the HP41 Facebook group. I realise this is only a step in the process of restoring the calculator to full operation, but it’s a satisfying one. I’m imagining, as an easy fix for contact-related problems, a custom two-piece perspex cutout that the calculator fits in and which is bolted together, applying pressure to the two halves and taking over the role of the screws and posts. Hmmm…

Mega 328 Component Tester Housing

This is my third Mega component tester – I keep dropping or breaking them. I find them to be a pretty much indispensable tool. I see there are perspex covers available on eBay, but when I got my latest one I looked around for something to put it in. Cassette case! I cut away part of it with my soldering iron (mmmm, fumes!). The 9 Volt battery was too thick to fit in the case, so I cut away part of the rear, padded the interior with draught strip and taped it all together with clear tape. Four stick-on equipment feet, and – voila! So professional…not! Feels remarkably solid, compared to how fragile and vulnerable it is without a case.


Yamaha RX-397 Dead As A Doornail – Fixed!

I got this to repair because it was dead as a doornail. I like that expression. It has an inexplicably surreal quality. I tested lots of things, mostly in the power supply area. Almost immediately after the AC line entered the amp it measured 140V, and I couldn’t figure out why – something must have been pulling it down. I tested components on the two power supply boards and the transformer primary and secondaries were fine. A couple of days later I was having another look at it and a penny dropped – I remembered a Yamaha RX-E810 Pianocraft receiver I’d worked on.  I tested the 22nF 630V metal film capacitor across the AC line, it measured 4.3nF, I replaced it – and the receiver powered up. Amazing. Go to the head of the class…apparently this is a very common problem with these units, as the internet advises.

Pioneer SX-850

I got this to repair because the left channel dropped out when the tone controls were selected – this receiver has a switch to bypass the treble and bass. I (foolishly) took the customer at their word and opened the unit up and started testing things. ALWAYS TRY THINGS OUT FOR YOURSELF! I tested all the components on the tone control and flat amp (that’s the one with the volume control mounted on it) boards and lubricated all pots and switches thoroughly. The thing I usually mess up: I was removing transistors, the pinouts were marked on the top of the PCB, I removed a transistor then realised it was a FET and I hadn’t noted the orientation. G$$%^&mmit! I always do this! But I put it in my trusty (I thought) Mega tester and identified the pinout, matched it with the pinout on the PCB and Bob’s etc. BUT…nothing! No sound from amp…back to drawing board. A couple of days later, I removed the FETs and found that if I tested them, then reversed them in the Mega tester’s ZIF connector – I got the same pinout! I guess there’s a bug or inconsistency in the Mega tester with regard to FETs! I looked up the pinout in my transistor guide (invaluable, and, inexplicably, no longer available), installed them – and – NOTHING! A closer look revealed one of the wires from the power supply to the protection circuit board had broken, so I CHECKED where it went, resoldered it and the mighty beast powered up and sounded AWESOME and NOW I WANT a ’70s silver-faced receiver. For heaven’s sake…NOTE ORIENTATION BEFORE REMOVING COMPONENT! I should have it tattooed…on the inside of my eyelids!

Sansui AU-555A Hissing

This had a blown fuse in the left channel and one of the flying saucer transistors in the right channel was half-dead. There was signal but it was distorted and leaking from the right channel into the left channel output, strangely. I replaced the 8002-1 and 9002-1 driver transistors with BC639 and BC640. All was working well, the owner collected it but I got a call from him that evening. Apparently it had worked well for about half an hour then started hissing from the left channel, not volume-dependent so I assumed it was in the power amp. He returned it, I replaced the two remaining flying saucers but it didn’t fix the problem. I tested, replaced or substituted all the capacitors and semiconductors in the left channel of the power amp, excluding the power transistors – I could hear the hissing in headphones, so I assumed the faulty component was somewhere before or including the driver stage, as the headphone jack is fed from there. A bit of reading suggested the fault might be on the preamp board, but after the volume control. I replaced the famously noisy transistors in the preamp with two BC549 and two 2SC2240 when I ran out of the former. I had to redo this as I hadn’t paid close enough attention to the orientation of the ones I removed, which were in an unusual package. So, consulting the schematic I managed to get them in and the preamp working. This is a theme which will continue throughout the weekend, dammit. I seem to be getting dumber with age! So I powered up the amp, all good for a few minutes – behold, the noise reappeared but in the right channel this time. I pressed on the huge, telephone-part-looking connector from the main board to the preamp, and the noise went away! I resolved to remove the connector and solder the wires direct to the preamp board. I undid the connector, drew a diagram of the layout and started removing the wires. After a while I started to realise that I hadn’t been exact enough about the layout or orientation of the connector, and hadn’t taken pictures. Sooo…a couple of hours later I managed (with the help of the schematic) to get it all connected properly. And fortunately didn’t cook any other components in the process. What’s the lesson here? NOTE ORIENTATION BEFORE REMOVING COMPONENT! TAKE PLENTY OF PICTURES! The thing that tripped me up was, there were two possible ways for the socket to fit into the PCB and I hadn’t carefully noted which way it had been installed. There were two pins which had been cut off, even though there were wires soldered to the tags for these pins. I had soldered these wires to the corresponding holes in the PCB, but on the second-to-last attempt there was still no signal coming through. I realised these wires were just shorting the 12V preamp supply voltage to ground, so I lifted them and all was well. I sometimes wonder if I just do things carelessly to make more interesting problems for myself. The way I ended up doing the job, I gained a much closer understanding of how the circuit works and got practice at reading a schematic. I suppose the only concern is, if I’m trying to make money doing this, unproductive time is best avoided. I’ve seen a few internet postings about noise in one channel of the AU-555A, but nobody has mentioned the possibility that the connector is one of the culprits. I’d tried using contact cleaner on the pins and refreshing the solder joints, but the hissing continued. Best to remove it completely, I think.

iPhone 5S Battery Drain Woes…

So I buy broken iPhones, fix and sell them. I recently purchased a white 5S which was supposedly working fine except for a broken screen, NZ$71. The housing was dented quite badly near the sleep/wake button (to use the correct Apple terminology) and the touch ID didn’t work. I took the logic board out of the housing and tried straightening the housing, but despite being careful fractured the plastic link between the main housing and the top end. I always manage to do this! Actually I’d rather replace the housing than attempt to straighten it when it’s quite bent, sorry Louis Rossman. I hunted through my used housings and chose the best one, a vivid gold colour, transferred the components, reinstalled the OS – actually updated to iOS 11, d’oh! – and put it on Trademe. I noticed that it would be dead by morning, then it wouldn’t charge, so I bought a new battery and it charged fine. Meanwhile I sold it, and then noticed it was still running down completely after 4 hours. I told the guy all this when he came to collect it, he left it with me and I put my SIM in to test the phone out. I didn’t think the problem was symptomatic of U2 disease. It would charge properly, or so it seemed, but it kept going flat after 4 hours use, and it would last all night on standby but be at 5% by morning. I bought a black 5S with a broken screen, and rather than order a new screen I put the logic board into the gold housing, and the same thing happened. The phone would run completely flat after 4 hours of moderate use. I did all the turning off people recommend on the internet – background app refresh etc. – but no joy. The buyer passed on the trade after I gave him an update, and tonight I tried putting the black 5S logic board back in its original black housing. I was just reading something which talked about poor cellular reception contributing to battery drain, and my workshop is in an absolute black spot, so I’m wondering if that has anything to do with it. But I’m noticing the black 5S has been sitting on 72% for the last half hour and wondering if the housing can contribute to battery drain. I changed the charging flex in the gold housing because I’d read that a short in that can cause battery drain. I wonder if something like a burr on the interior of the housing or something like that can cause a short. I made sure all the connector foam pads were present (they weren’t) because somebody said that contributed to battery drain. So I’ve turned the place upside down trying to fix this. I guess there could be some component on the logic board causing drain, but the same thing happening with two different logic boards? The housing is the only thing in common – different boards, batteries and charging flexes. Or maybe it’s iOS11…watch this space!

So a couple of weeks have gone by, I put the original logic board into a new aftermarket housing and used the phone with a SIM installed for a couple of days. I’d get about 3 1/2 hours use before the battery was flat. I realised on Sunday that the housing wasn’t the only part I hadn’t changed, so swapped the front camera/proximity sensor/ambient light sensor/ear speaker flex, the ear speaker and the metal screen shield – and, lo and behold, the phone now lasts all night on standby, a solid 8-9 hours of use on a charge – amazing! It’s been sitting on around 94% for the last 4 hours, Wifi on, background app refresh off and no mail account installed. Which is not a lot of drain, but in the past the charge would have gone down like a cartoon cigarette. Amazing. It may now be saleable. I realise this isn’t on the same order of phone fault diagnosis as “short on VCC main” or “replace Tristar” but the proof of the pudding, as they say, is in the eating…I thought it was Tristar for a while, but it has none of the symptoms of Tristar fault, as far as I can make out – no fake charging, it will charge from a drained battery etc.



Adventures in old tech…until your mind reels…