RISCPC not booting, scratchy beep

Hi Folks,
I hope you can help me here. I have a RISCPC600 that was operating quite well up to a day ago. It was working ok, when I switched it off for a reboot. The usual start up seemed to be happening, but instead of a nice crisp beep, I got a scratchy,crunchy sound. The video flashed the first few times I retried it but now I have no video at all.
The Caps-Lock button on the keyboard still toggles. The Hard Drive light appears to show some booting activity. I have taken out the memory and video boards and cleaned contacts, they look quite ok. I have removed the power supply and checked the +5V, +12V and -12V were all correct. I have checked the CMOS battery (now a 1.2NiMH) has plenty of charge. I have removed/reseated the StrongArm card, ROMS and I/O podule backplane.

I have tried doing a power on/Delete start up, and blind typing to try to recover the video with configuration commands, but to no avail. The mother board has had some damage, mainly cosmetic I think) from a leaky battery (curse them!) and has been cleaned up about 10 years ago. I have been using it regularly the last 6 months to get some retro-fun vibes and enjoying the challenge, until now.

The scratchy sound, accompanied by keyboard flashing, is regular and somewhat delayed by normal start up sequence, is quite regular, ie it does the same every time.
Any further debugging suggestions or better still a solution would be greatly appreciated. I have come this far with getting it back up and running a hardware fault like this was always my greatest fear. I live in Australia so dropping around to a RISCPC guru is not really likely.
Cheers and thanks,
Rob

The first step is disconnect all cables, remove processor cards, RAM and VRAM, inspect for any corrosion on contacts, then put it all back together again. Sometimes just action of doing this clears some faults. Next check with a different power supply.

OK David, I confess I haven’t checked the back of the board recently, so I will take it all out tomorrow and give it a once over for corrosion. I will take it to the Mens Shed, so I can do it properly. I don’t have access to an alternative power supply. I will try to measure the voltages while under load.
I have cleaned the RAM and VRAM sockets about 4 months ago when I realised the amount of very old goo that was being produced on the contacts. I have since bought new RAM (2*64Mb, with gold contacts this time) and 2Mb VRAM. The machine was booting fine with these in since I installed them. The previous RAM was a single 128Mb card and the solder contacts were way beyond repair, hence the replacement. I had not considered the battery fluid getting into the base of these sockets the first I cleaned them about 10 years ago. (Bad decision).

That part of Australia I live in WA if you are close I have two risc pc in storage if you need help just ask.

Hi Rob,I live in East Gippsland near Lakes Entrance, we are about a continent apart. Two in storage sounds like heaven at the moment to me.

I have inspected the back of the PCB mother board and the damage does not appear very bad. I have cleaned around tracks with “corrosion fuzz” on them to check integrity of tracks and they appear ok. I will post some pictures and more detailed comments when I get home tonight.

Thank you very much Rob for your offer of help, it is very much appreciated. I may very well need some help.

Rob

Well further investigation has produced the following observations:
1. The damage under the board is minimal, the acid may have run down vias and lifted the solder mask in places, but not much else.
2. Pressing the chips around the battery area has got the machine going again, but if left for a while it reverts to no start. I have not had time yet to methodically push each chip in turn to analyse the problem further.
3. The CMOS chip is getting the battery voltage (1.35V) when the power supply is switched off and when switched on it reaches (2.5V) so would appear to be powered ok. It is keeping settings like modes, monitortype etc over cold restarts. However the very bad palette mashup is continuing and needs to be corrected with Display/Mode selection to correct it. Only a minor mashup occurs with a recycle start up from the reset button.
4. What has definitely changed though is the clock is totally resetting to Jan 1 1900 etc on a cold power up. I tested the continuity between the crystal and the CMOS chip and both leads appear to be connected. What does this indicate?
5. There is some small, but critical PCB damage, some where that is/was intermittent and may well be where the colour cast is coming from as well.

Does anyone have a hi-res photo of the CMOS/Battery area of a RISCPC they could type in measure voltages please so I can compare it with the a working system? This is probably a big ask as there are quite a few points. At the moment I am clutching at straws to save it. Or can anyone suggest a more systematic debugging procedure than my poking around in the dark might reveal. Are there any published comprehensive RISCPC circuit diagrams?

Any help gratefully received.
Cheers, Rob

If you haven’t already, you should probably try asking about this on the stardot 32bit hardware forums , lots of info there about this kind of issue.

Ok, thanks Matthew I will give them a try too. If I make any progress I will report back to tidy this thread up. If anyone here has an idea, I would still be grateful.
Cheers, Rob