Acid Damage Issues

The Last few months have been strewn with synths with Battery acid damage issues, we’re not entirely surprised by this, as the price of any analogue synth climbs, the more dross comes in from attics, sheds and garages with totally unbelievable sob stories attached as to how it got in this condition. You aren’t fooling anyone guys, you got it for next to nothing and now want it sorted so you can cash in. If it isn’t too bad and the desired result is obtained reasonably quickly, then you may get your wish, but if it is bad and a lot of tedious work has to be done then the costs will rise very sharply and to high levels and your returns may not meet your expectations, especially for a rare but undesirable item that has an unknown inflation path.

The list of equipment that we have seen over the last few months with NiCad battery backup has seen several new additions, one unexpected item was the Cheetah Master Series controller keyboard, in this case the 88 key version. It has a very nice Fatar hammer action keyboard and a lot of buttons for a lot of control and is a very well built piece of equipment but the NiCad killed it.
We had it back in September for a first crack at it and there were a lot of tracks to rebuild and a couple of IC’s to replace but it worked for a week after it left us, but now the CPU has died, and it is a weird one only available according to our best searches from, It is a 64 pin chip that is the same size as a normal 40 pin IC, the pins are of necessity closer together, (the beginnings of the miniaturisation that plagues us today), how many synth service agencies can handle large BGA chips?
The old CPU will have to be cut out, there is not much chance of desoldering a 64 pin chip with some acid damaged pins without further risk to the pcb, a suitable 64 pin socket is available but can’t be used due to height constrictions as the CPU is mounted on the same side as the front panel switches, and space is very limited. Checking out the Internet seems to confirm that very few of these Cheetah’s remain in the wild (just like the feline version!) It’s probably due to the battery problem and the complete lack of any service documents for what is a very sophisticated controller.
As of early March 2015, the new CPU has been fitted and the Cheetah finally working properly and back with it’s owner, the last hurdle was that controller wheel 3 (for MIDI volume control) wasn’t working whereas it did before, and this turned out to be yet another acid eaten PTH (plated through hole). The whole thing had to be dismantled yet again to sort this. A new switch mode regulated Wallwart was converted for use with the machine, as the original was a big transformer with a Euro two pin mains power connnector which was a pain to plug into an Irish (or UK) socket.

The Teisco SX400

We got one of these in a couple of weeks ago which was traded on a local auction site. It’s provenance is not certain but it’s complete lack of functionality is beyond doubt. It’s dead, it has ceased to be, it is an ex synth!

This synth did have a Ni-Cad battery originally but has a Coin Cell in it now that the recent seller thought might be enough to bring this monster back to life. There are some signs of acid damage on the pcb but nothing as severe as you would see on a Polysix or Poly 61. We suspect that this synth was dead a long time before the battery got a chance to leak, and that may explain it’s lying up in storage for so long. Its front panel was also covered in the residue of very old Gaffa (or Duck) tape across most of it’s width, which had dried out to become a horrible brown scab over much of it. White Spirit seems to be the only thing that works on old Gaffa glue, Isopropanol just won’t make a dent, and the only scraping tool that does no damage is your own fingernail, you can’t use screwdrivers or anything metal, it will scratch the paint permanently. Other possible scraping implements are wooden cocktail sticks or large cactus spines, We have tried all of them and unfortunately the fingernail is the quickest and best. The picture below is obviously after several hours of cleaning, and a severe need for a manicure!


This synth is definitely a strange beast, very clever in some ways and extremely dumb in lots of others. It’s keyboard has Aftertouch using circuitry derived from an ARP patent, but it only has 4 Oscillators which are assigned in ways more reminiscent of the Korg Monopoly than a polysynth, but it does have 4 filters and 8 envelope generators making it a bit more like the Jupiter 4. It also has 8 preset tones, and 8 programmable ones, again similar to the Jupiter 4. Unlike the JP4, none of the Oscillator functions are stored, while almost everything else is. The other intriguing thing is the tremendous amount of electronic circuitry used to achieve so little result.

Gordon Reid lambasted this synth in a Retro review in Sound on Sound magazine as a complete turkey, but one that he owned and had a grudging like for.
It has discrete oscillators using uA726 chips (as does the JP4), discrete 24dB ladder filters using the same transistor types as the TB303, the only other Polysynth we can think of that has discrete ladder filters is the Memorymoog. According to Gordon Reid, the SX400s’ filters don’t sound good, but if this beast ever comes close to working we have a few ideas to improve them. It also has discrete Voltage controlled Envelope Generators, and uses loads of 3080 and 3280 IC’s for it’s various VCA functions.

So is it worth spending a huge amount of time on? Logic says no, the values of its rare components are more than enough to recompense everyone, but it is a unique if weird machine, and quite rare these days on Ebay, says the last one sold in 2012, which is a long time ago in synth bartering terms but there were a few more in 2011, but none in a perfectly working state.


The above picture is obviously of the CPU board, the little white stickers indicate the first candidates for further testing and/or replacement. None of the CPU board behaved anything like it should. The Z80 CPU had clock and activity on all it’s Data and Address lines, as well as the 4 main control lines. For the time being we had to give the Z80 the all clear, an EPROM image obtainable from the Internet was burned to an EEPROM, and tested but made absolutely no difference, so the original EPROM was reinstated. The initial issue was as follows, the only function of the synth that seemed to work was the Preset and Memory buttons on pcb Syn 36. These weren’t flawless but the right buttons lit up the right LED’s, so a hint again that the CPU and EPROM were not culprits.

Unfortunately nothing else does, the ADC and DAC are completely non functional, the multiplexers for the sample and holds are not getting any control signals at all, the keyboard didn’t work and none of the analog switch drivers (used for selecting waveforms, modulation options etc.) worked. The way all the external circuitry is driven is a little unusual. The CPU’s address lines mainly just talk to the EPROM and RAM, only 12 of the 16 available address lines are used, almost all other addressing is done by a few address lines and the I/O and Read and Write lines of the CPU along with a large and complicated system of decoding IC’s to scan and drive all the functions of a Polysynth.

This is March 2015, but for continuity the story will continue here for a bit. It seems as though the synth was stored on it’s end with the pitch bender towards the ground, and the battery acid has wiped out every IC in its path to the left of the CPU board which includes all the Pot and Slider reading, and all the Voltages for the Sample and Holds. We replaced a total of 15 IC’s on the CPU board which were obviously damaged and/or behaving suspiciously. This is a quarter of the IC’s on the CPU board and the only net gain has been that the keyboard scanning now works. Each of these new IC’s lives in a Gold plated turned pin socket, so the time and costs are already mounting. Some of the IC’s are not common items and will have to be obtained from various and far away sources. This synth has gone back to it’s owner due to time and space considerations while the replacement IC’s arrive, and the next batch of 15 IC’s is ready to go into it, which will be to replace everything in the path of the acid leak.

We think that the acid itself does not kill the IC’s but power being applied to the synth when battery acid is present on the PCB certainly does. Our reasons for this are twofold, the behaviour of Polysixes and Poly61’s while being used and the rapid and almost logical progression of failures until the machine is dead. The other was a GEM S2 Turbo which had it’s battery replaced while the synth was seemingly working fine (it had been in storage for a time) and within a day the chip that buffered the Display died, cutting the old chip out revealed a tiny droplet of acid under it which normal cleaning could not get to. The S2 worked fine after that for another couple of days and then we noticed severe Digital distortion which we think was due to problems with the Sample address decoders which live very close to the previously replaced chip, another drop of acid under another chip? Probably!

Battery acid problems are certainly some of the nastiest and difficult to diagnose and fix of all the problems that synths can have, and are becoming the principle reason to declare a synth dead, and not fixable at all or not economically viable to fix . This Teisco is perilously close to the latter. When it comes back and the next round of chip replacements is installed, that will be crunch time, if there is still no noticeable improvement in functionality, then we will have to “call it” and dispose of the body parts to save other synths with a better chance of survival.


This picture is obviously of the right hand side control PCB, the sliders had taken a battering over time and were very stiff and notchy in their action, so they were all removed from the PCB, rebuilt, straightened out, cleaned and lubricated with new grease and then put back in. They measure perfectly and feel fine, the only reason the slider caps are not back on them is in case one or both of the Multiplexer chips on the board has failed, and somebody in the past Superglued the slider caps on, making them very difficult to remove without risking damage to the sliders themselves. When (and if) we are happy with the results, they will go back on with a soft (Evostik type) adhesive if necessary.


The above image is of a Poly 61 CPU board, the synth doesn’t work and there is a green patina over most of the components from one end to the other of the board, the pins of several of the plug-in connectors have been eaten away and it seems a complete loss. It is possibly fixable given enough time and money, but would a €1200 repair bill for a 61 seem like good value?


The above picture is of the worst condition Roland JD800 we have ever come across. The infamous red goo has not only leaked from the black keys (as is usual) but from all the white keys as well, leaving an incredible mess all over the keyboard and base of the synth. We went to a great deal of effort to rebuild the top octave and a half of the keyboard to ensure that the keyboard’s contact systems were still OK, and no they weren’t. It took many hours over a few days to clean the sticky keys thoroughly, apply new Epoxy to the weights and to clean the rubber contacts, and the keyboard still just doesn’t work. A replacement keyboard from Roland has been quoted to us at €850 from Roland Ireland, and to add insult to injury most of the tact switches don’t work either, so they will all have to be replaced, the Editing sliders have problems too, quite a few of those would have to be taken out, stripped and rebuilt. This synth is economically unviable to fix, with no guarantee of perfection even if a large amount of money is spent, so like many others lately, this one is beyond the call of duty!

Even before the process of trying to rebuild the keyboard, the output Jack Board had two smashed output jack sockets which were replaced and a dual op-amp in the DAC section which was seriously gone shorting out the analog voltage rails. We got it to make sounds, but only over MIDI, and as for the rest, it’s just too much.