electrolyzing a TB-303 (weird repair story)

[Trevor Page]

> I had a pretty weird repair this weekend.
>
> One of my friends has a TB-303 which got rain water inside after his roof
> started leaking.
> When he came home he turned on his TB-303 which functioned ok for some 5
> minutes
> then it went completely crazy with flashing leds on and off so he turned
it
> off. When he
> picked it up to inspect it he saw quite a lot of water running out of it.
> So he found the hole in the roof and also another synth that was exposed
to
> some rain
> water.
> So when I came around he asked me to look at his broken TB-303. I opened
up
> the
> case and expected to find some white residue close to the microcontroller
> and RAM chips
> and maybe some burnt out chips due to the short-circuits that the water
had
> caused.
> I found that the defec was indeed around the microcontroller and the RAM
> chips but
> instead of finding some white residues due to the rain the whole area
around
> the RAM chips
> was covered in red residue. It looked pretty horrible.
> We set out to buy some stuff to clean the pcb (ethyl alcohol and
> demineralised water).
> After coming back I desoldered the switch board assembly from the main
> board.
> (pretty cool to see the owner's face turning white ;-)
> Then after desoldering one sixth of the main board I cleaned it and the
red
> residue was
> rust. So the next thing was to check where the rust came from. And here
> comes the weird
> part: two of the RAM chips had both their pins 20 eaten away. Since pin 20
> is the plus pole
> the only thing that comes to mind is that the rain water was electrolyzed
> during the 5 minute
> turn-on period. This caused acid to be generated at the plus pole which in
> turn attacked the
> chips' pins and causing the red residue and the failure.
> The reaction would be: 2H2O ----> 4H+ + 4e- + 02.
> I have never seen this kind of "weird" stuff but it seems to me that it's
a
> good hypothesis because
> only the plus pins were eaten (nothing else on the whole board).
> After cleaning, resoldering the whole thing and reassembling it. I plugged
> the wall wart into it and
> it functioned again. And my friend became very happy again instead of
> looking at the wall paper the
> whole time.
>
> Has somebody else seen this defect ? Is this common to electrical devices
> that have been exposed
> to water ? It must be I guess.
>
> Greets and thanks to listening to my story (maybe we should start a
> synth-diy cafe for this kind of bar talk ;-)
>
> Heiko
>
>

Quite remarkable. This is something that happens in TB303's anyway because
there's always a small current flowing through pin 20 on each RAM. Possibly
caused by the fact that they're in close proximity to the batteries
themselves? I replaced one RAM in my TB303 a while ago and recently noticed
when I opened it up that the +V pin on that IC is starting to discolour
slightly.

Trev


=========

Trevor Page BEng (Hons)