On Jul 20, 2017, at 1:11 PM, Noel Chiappa via cctalk <[email protected]> 
wrote:

>> Next stop, I'll pull the cover off the machine and see whether I can
>> spot any spilled battery electrolyte from the old battery or anything
>> else suspicious looking on the logic board in that area
> 
> It probably wouldn't hurt to clean that area with a Q-tip dipped in distilled
> water. (If you get noticeable grup on the Q-tip, repeat with a new Q-tip,
> otherwise you'll just be spreading the ook around.)

        ...re-animating a thread I let get dormant for around 2 months - sorry 
for the delay and thanks for the help so far! 

        The machine in question is an iMac G3, Summer ’01. The presenting 
symptom was odd ethernet behavior and never any packets transmitted over 
ethernet, starting while the PRAM battery was bad (went bad with the machine 
running) and not corrected by replacement of PRAM battery. I had tried multiple 
good suggestions to diagnose and command ethernet status via software, but to 
no effect.

        I did pull the cover and cleaned the area around the battery holder 
(and I/O board including ethernet port) with Q-tip, though humidity was ~100% 
so I elected to use 91% isopropanol rather than water. No obvious contamination 
near the battery or on the interface board; I also cleaned the ethernet 
connector on the outside of the IO board.  I have not re-tested, though - see 
below.

        However, while in the process of doing that, I noticed 3 big 
electrolytic caps on the power board which look failed to me. All 3 are 3300 
microFarad, all 3 have bulging upper ends (the ends with the strain-reliefs, 
away from the circuit board), and one has a brown contamination stain on it. 
One more has a single spot right at the convergence of the strain-relief 
grooves, so it might also be leaking.  I grabbed a few bad photos with my 
phone, which are available on request. I note that the brown stain on the one 
is on the lower end of the cap as installed in the computer when the computer 
is right-side-up, so if that component is leaking electrolyte, it makes sense 
that it would have accumulated and evaporated there leaving some residue.

        My inclination is not to re-start the machine until those capacitors 
have been replaced; I have no good idea whether they could affect the ethernet, 
but whatever they are supposed to be doing, I’m pretty sure they are not doing 
it properly by this point and I think that’s the next thing to pursue. I do 
have an ESR meter but I’ll have to pull the main and power boards out in order 
to get to the terminals, and I shied away from that the other night. I don’t 
have a working O-scope to test noise on the power lines, although that would be 
relatively easy since the power connector has easily accessible wires with the 
cover off.

        Questions for the list:

1) the capacitors are hot-melt glued to a central capacitor and to the circuit 
board. Do I need to replace that hot-melt glue when I replace the caps?

2) What is a good source/brand for replacement capacitors (in the USA, San 
Antonio TX if that makes a difference)?
My priorities for replacing are: same specs, same diameter, better reliability. 
The labels say:

        Su’scon
        3300 (mu) F
        10 V

and on the other side:

        SEK105°C
        07/01
        A3

(or) 

        SD105°C
        0124(V)

        As always, many thanks for any help you can offer; I can’t make any 
more progress this week (work :-P) but hope to get back on this next weekend.

                                                                                
                        - Mark

Reply via email to