Electrolytic capacitors and other picture postcards.
There is one type of ecap that is rather sensitive to
negative spikes and overvoltage. These are called tantalium
capacitors.
Tantalium cap have that very distinct drop-shape !
Best way to protect them is to solder an 1n4001 antiparalel
between then ground and the positive powerline such that it will short
out any negative spikes on the board. Install likewise on all other
power feeds. Overvoltage protection needs a specialised cirquit with a
crowbar to blow the fuse in case of an overvoltage condition.
----+------- Positive plane
|
_
^ diode (note orientation)
|
----+------- ground plane
|
_
^ another diode
|
----+------- negative plane
Sipke de Wal
Bill Sudbrink wrote:
>
> Well, my OSI C3D is currently non-functional. Here's
> what happened:
>
> I've been testing and repairing a number of OSI 48-pin
> bus cards using the C3. This has required a large number
> of power cycles. Also, several cards caused the PS fuse
> (2A, 250 volt, glass cartridge) to blow. In each case, this
> problem was corrected by replacing the electrolytic caps on
> the offending board. OSI used 47 microfarad, 16 volt ecaps
> (ecaps... is this a good abbreviation? I'm getting tired of
> typing "electrolytic") pretty much exclusively. I could not
> find a source for these, so I used 47mf, 36 volt ecaps as
> replacements. This seemed to work OK. I finished working
> on the other cards and tried to bring up the machine with
> just the original cards. The fuse blew. Grrr... I pulled
> all of the cards, replaced the fuse and tried them one at a
> time. Sure enough, the CPU card (the model 510, triple CPU
> card with 6800, 6502 and 8080) was the culprit. I replaced
> its two ecaps as per above. It came up, rebooted once OK,
> started to get flakey and now won't come up at all. Is my
> understanding of ecaps correct? That is, as long as the
> capacitance is correct, the voltage is greater or equal to
> the required load and it is installed "right way around"
> it should be OK? There is so much logic on this board, a
> lot of it 8Txx, I hope I haven't blown a chip. It would
> take me a month to find it and who knows if I could get a
> replacement.
>
> Thanks,
> Bill Sudbrink
Received on Tue Feb 08 2000 - 14:09:44 GMT
This archive was generated by hypermail 2.3.0
: Fri Oct 10 2014 - 23:32:52 BST