SE/30 display crashing to black with a raspberry sound

[CC_333]

To elaborate somewhat on what I *think* techknight meant, the capacitors in the power supply (PSU) are likely failing, as well as maybe other components.

So, to start, the PSU probably needs a recap. And if that doesn't fix it, it might be easiest to find a replacement which is has already been recapped.

c

 

[macclassic]

Thanks, I saw a list of caps for the PSU on the forum so I will price them up, but I’m used to fixing cars so I also like to find and know the root of a problem too and whats bugging me is not knowing what’s happening in this case before I proceed  -  even if I have been barking up the wrong tree for a while

 

[CC_333]

Yes, it is pretty much shotgunning it a bit to just replace the capacitors without testing them, but if you remove one of them, you'll be convinced they're bad, as most likely you'll find a stinky puddle of electrolyte underneath it.

And also, experience has shown that in the vast majority of cases, when things like this happen, replacing the capacitors *usually* solves the problem. And when it doesn't, it's usually some relatively small component that needs to be replaced in addition.

Anyway, if you don't want to deal with recapping, a possible alternative would be to simply retrofit a modern PSU into the old PSU's casing. There's a SeaSonic PSU which is very popular for this purpose as it's very small, but I don't recall the model number (it's posted on the forum in a recent discussion regarding upgrades for the SE/30 PSU, coincidentally). The hardest part is wiring it for soft power, but since the SE/30 doesn't need that, it's just a matter of wiring it in and setting the new PSU to always power on when it's plugged in.

c

 

[Compgeke]

Thanks, I saw a list of caps for the PSU on the forum so I will price them up, but I’m used to fixing cars so I also like to find and know the root of a problem too and whats bugging me is not knowing what’s happening in this case before I proceed  -  even if I have been barking up the wrong tree for a while

Think of it like replacing the old rubber vacuum lines in a 70s or 80s car. They’re all either failed or failing so they should be replaced as a preventive step. A lot of weird issues are solved by fixing the vacuum lines, same with replacing the 30 year old caps.

 

[techknight]

The problem with this stuff is you need to know how to do electronics troubleshooting and have the proper tools, as well as knowing how to use those tools. 

Electronics are different in the fact that you cant physically "see the electrons" moving around and doing there thing, unlike cars and other things in life, which makes understanding electronics really difficult to grasp for some people because you cant put your finger on it. With this kind of stuff, you have to think 4th dimensionally. 

The glitches are being caused by power supply dropouts. What causes power supply dropouts? Tons of things. feedback circuit issues, loss of the oscillator, overcurrent/overvoltage erroneous detection, tons of things. It could even be down to a faulty noisy transistor causing the oscillator to stop and start again. Could be capacitors. 

if you had a good superfast scope with 1 shot capture enabled, you could probably pick up the spike or dropout on the power rails when it happens. Tracing the source of the problem would be difficult. 

 

[macclassic]

I genuinely accept some of that, as I attended a lecture on Graphene presented by Andre Geim at Manchester University,  but most faults have outcomes and the solving of them is of course greatly assisted by knowing how something should function according to spec. as also knowing the underlying principles, maths or science of how an individual part works, or a number of parts work together in a system to produce a designed outcome, and in the field of mechanics some old-school carburation faults can be quite tricky too.