exxos's Atari, Amiga & retro forum

Tomswork wrote: 14 Sep 2020 18:47 @rubber_jonnie

Is it a switching psu I would check the windings on the transformer the enameling may getting old if its not in bad a shape then its good but sometimes you can seal them with epoxy to stave off a short. And diodes to go open or turn to resistore

Tom

Back to work this week so may take a few to get back on this, but thanks for the advice 😁

Tomswork wrote: 14 Sep 2020 18:47 @rubber_jonnie

Is it a switching psu I would check the windings on the transformer the enameling may getting old if its not in bad a shape then its good but sometimes you can seal them with epoxy to stave off a short. And diodes to go open or turn to resistore

Tom

@Tomswork I believe the SK3120C voltage dropper IC is the culprit, as the rest of the PSU, transformers etc look pretty good. It's also the last thing in line before the 12V rail. Lack of schematic has made it harder, having to manually trace tracks through.

I've got another post going over here to try and source some or find an alternative: https://www.exxosforum.co.uk/forum/viewt ... =23&t=3379

I have at least been able to hook up my bench PSU to the 12v side and verify the rest of the PSU is working fine, as is the Amstrad once it has a decent 12v supply.

As my replacement tantalum caps arrived for the one that went bad and nearly killed the PC, I dropped one in this evening, and it has a voice again, as it was in the line to the speaker. I get clear loud beeps now :)

I guess in the case of this particular machine, low speaker volume is the incentive to dive in and clean up as a sign the electrolytic is bad.

Well, that was disappointing, the voltage dropper wasn't at fault, so no I have to start working back thru the circuit to see what's next. Got to try and get my Google head on and try to find a schematic, or start tracing the PCB further.

I need to try and measure voltages at the SI3120C too, but it's not a pleasant trying to get to components as it's pretty tightly packed.

what about the enable signal ? or feedback resistors ?

exxos wrote: 19 Sep 2020 23:50 what about the enable signal ? or feedback resistors ?

Trying to figure out what bits are connected to where, but the PCB is double sided and quite densely packed. Plus I'm no expert. I need to try and figure out the whole +12v side. Gonna try and find a schematic, or otherwise I need to start trying to trace stuff, which is going to be a huge pain in the ass. I'd gut it and replace with modern parts, but we are looking at +-5v, +-12v and -24v, the latter being the tricky part.

If you powered the 12 externally and it worked... Then I'd assume all you need to check is if you have a higher input voltage into the regulator such as 20v etc. It sounds like your missing that voltage ?

exxos wrote: 20 Sep 2020 00:21 If you powered the 12 externally and it worked... Then I'd assume all you need to check is if you have a higher input voltage into the regulator such as 20v etc. It sounds like your missing that voltage ?

That's actually next on the list, I'm just trying to work out how to do it easily. The internal power lead is super short and I don't have a load other than the PC itself. Might have to solder some leads to the bottom of the PCB to break out some test points, but that's not a problem.

Impossible to measure from the top.

@exxos Well, I'm now wondering if there is some kind of mechanical problem like a bad solder joint. I stood the PSU on it's end on the low voltage heatsink so I could measure the voltages at the bottom of the PCB, directly at the SI3120C.

Low and behold I have 12v and everything powers up fine and dandy. So I lay the PSU down again and no 12V. time for a bit of a reflow session I think.

rubber_jonnie wrote: 20 Sep 2020 15:00 @exxos Well, I'm now wondering if there is some kind of mechanical problem like a bad solder joint. I stood the PSU on it's end on the low voltage heatsink so I could measure the voltages at the bottom of the PCB, directly at the SI3120C.

Low and behold I have 12v and everything powers up fine and dandy. So I lay the PSU down again and no 12V. time for a bit of a reflow session I think.

It lives!!!


So , just goes to show you should look at the simple stuff first. It turned out that the short 7 wire power cable that plugged into the PSU and mainboard was at fault, which showed itself up when I turned the PSU on its side to troubleshoot and it worked, but then didn't when I put it back into normal orientation.

The problem, was these connections in the plug at the PSU end:


This is a good one, you can see it has a D shape looping back into the main connector body. This D flexes and grips the pins on the PSU, however the 12v one had become brittle and snapped off, so no longer gripped the pin very well, sometimes working sometimes not, hence the erratic nature.

The wires are now soldered at the PSU end, so no danger of this happening there again!

I really wish I'd noticed this sooner, it would have saved quite a bit of time, but I just didn't expect a 7 wire, 1 inch cable to fail. I guess the moral of the story is "Don't assume, check".

Got there in the end though.

:girldance: :girldance: :girldance:

Thanks to everybody for their input :)

:dualthumbup: :yay: