Rob, replace c21 because of the bulge. If you want to check an electrolytic cap with a ohmmeter ,(kinda hard with a digital one), connect the meter across the ends. You should get either infinite resistance or a momentary reading of zero going back down to infinite resistance as the cap charges. Then reverse the leads and it should be infinite from the start. On an old analog meter with a dial, you can quickly tell by see the needle deflect full scale in one direction then settle back down to infinite. I keep an old meter around just for this purpose.
Sent from my iPhone > On Aug 30, 2025, at 23:01, Rob Jarratt <[email protected]> wrote: > > >> >> -----Original Message----- >> From: Wayne S <[email protected]> >> Sent: 29 August 2025 22:57 >> To: [email protected] >> Cc: General Discussion: On-Topic and Off-Topic Posts >> <[email protected]>; Robert Jarratt <[email protected]>; Scott Baker >> <[email protected]> >> Subject: Re: [cctalk] Re: Repairing an Olivetti M24 PSU >> >> Hookup your scope across the 5v and gnd on J254 and see what the trace >> looks like. Also do +12 v. >> And -12v on j100 >> That will tell you if one rail is bad. >> You should have not much ripple. >> If there is a lot, trace that voltage rail back. >> > > Thanks for the suggestions, Wayne. Something changed in the behaviour. I am > getting some very erratic behaviour and I wonder if I am making a mistake. > > After getting your suggestion I checked the outputs again, because, to my > surprise, I got some output. With no load all the outputs seemed correct, but > as there was no load I switched on the PSU for only the briefest of periods. > Then I added a test load and the outputs were generally just about in spec > except for the 5V output, although +15V was also just fractionally below > spec. With the test load, the 5V output only produced +3.9V. I used a 0.5R > load for a 10A current, the spec says min current 6.2 and max 16.8A, so I > wasn't overloading it. I was wondering if the fact that I had removed and put > back some parts might have cured a dry joint, and in fact I had found one > connector was not well attached. The +5V ripple seemed OK, the spec says > 50mVp-p, but I was getting 400mV spikes every 16us. This is what the scope > showed me: > https://rjarratt.wordpress.com/wp-content/uploads/2025/08/5v-output-0.5r-load-1.png. > One of the output caps, C21, has a slight bulge, so I will replace that. > > Then I noticed that I had forgotten to put back some of the components that I > had partially removed for checking. I thought that putting them back might > resolve the low voltage on the 5V output. But, instead, it stopped working > altogether. The +5V output is at 0V and +12V is at 0.75V. The components I > had lifted were R60 and D66. This made me think that maybe the SCR was being > triggered. I lifted them again, but now the outputs are still stubbornly at > 0V (or thereabouts). I tried removing the SCR altogether but this made no > difference. Then at other random times the +12V will go to about 7.7V, while > the +5V output will stay at 0. This is all quite baffling. The signal on the > rectifying diodes seems unchanged, so something is wrong on the secondary > side. > > One possible candidate is C21. I have removed it and tested it out of > circuit, but I think my instruments may be struggling to cope with the > relatively high capacitance value (10000uF, not 1000uF as marked). My ESR > meter gives wildly varying ESR values, my DMM can't measure its capacitance, > even though it is rated up to 9999uF, but if I measure resistance, it shows > open circuit which is what I would expect. At the very least C21 is suspect > and as it has a very slight bulge, I am going to get a new one. Otherwise, I > am not really sure what could randomly affect the output so radically. It > would seem though that something is shorting +5V to GND because the > rectifying diodes are getting a signal from the output transformer. > > I am going to get a replacement for C21, and in the meantime I am continuing > to investigate, but any suggestions very welcome! > > For reference I am referring to the parts in the marked up schematic here: > https://rjarratt.wordpress.com/wp-content/uploads/2025/08/m24_psu_schematic-marked-up.jpg > > Thanks > > Rob > >
