Another A5000 restoration....

Arc/RPCs, peripherals, RISCOS operating system & ARM kit eg GP2x, BeagleBoard
lcww1
Posts: 23
Joined: Wed Mar 15, 2017 11:16 pm

Another A5000 restoration....

Postby lcww1 » Sun Nov 05, 2017 11:34 pm

I'm trying to restore an A5000 which has suffered battery damage to the PCB.

I received the machine after someone else had done a bit of a clean up of the PCB, and had replaced the CMOS battery. At that point, the machine was non functioning. There was no video output (though perhaps that's not strictly true - my NEC multisync monitor detected a video signal, but displayed only a black screen), no floppy disc activity at all, no POST output, and no keyboard lights (other than a brief blink on power on, and on pushing the reset button). Holding delete whilst powering on resulted in no difference. After allowing the machine some time to start, pressing F12 followed by ctrl-G was met with silence.

So, I checked the PSU output - all seemed well. The PCB, whilst superficially clean, still had evidence of plenty of corrosion on surface mount passives, and ICs in the vicinity of the battery. So, I removed the recently fitted battery, and cleaned the PCB. I removed the socketed ICs (the ROMs, and ICs 14 and 38 - these latter are both GALs, I think). I then used a dilute solution of vinegar, to try to remove as much of the crusty material as possible from the passives and IC legs by gently brushing with a soft toothbrush. I followed that with some water, to clean off the acid, then followed that with plenty of isopropyl alcohol, and then followed that with some Maplin contact cleaner. I then left the PCB to dry. This left the PCB looking much better, though far from perfect.

Using a multimeter to test PCB tracks in the vicinity of the CMOS chip and battery, and other obviously affected areas didn't turn up any problems, so the PCB itself appears to have survived quite well (this work was greatly aided by the info on this forum, and the links to the A5000 circuit diagrams on Chris's Acorns). The main damage to this A5000 seems to have been to the exposed IC legs and the solder joints of the SMD passives. Lastly, I checked that all the PCB links were correctly set to their default positions as shown in the A5000 TRM.

So, I then powered up the cleaned PCB, without much hope, and lo and behold, I was met with the supervisor prompt! A del-power-on to clear the CMOS later, and I was at a functioning RISC OS 3.10 desktop! I can run the ROM based apps normally, video and sound seem ok, and the machine reports the correct 2MB of RAM. However, I have not been able to get the floppy drive working at all. I know that the floppy drive is ok, having successfully tested it in a A3010. The A5000 just reports "bad drive", and there is no floppy drive icon on the iconbar (there are also IDE problems, but I'll save those for another day).

I've used the A5000 circuit diagram to check the tracks from the PL10 floppy connector, and these all seem to be ok. I can trace the signals to IC42, IC38, and IC26. However, I'm beginning to think that there may be a problem with IC38 itself. The legs of IC38 had a lot of battery corrosion, though they've cleaned up ok, and if there's a problem then I think it'll be inside the DIL package. The nearby IC42 looks remarkably good, so I think that chip may be ok. IC26 also looks fine, and is on the far side of the backplane connector, so was presumably shielded from the battery attack...

I'm therefore wondering how to go about testing IC38 (and IC42). I'm also wanting to confirm that, as IC38 is a GAL, there's no easy way of replacing this chip?

I'm actually amazed to have got this far with this machine. Having now got this far though, I'd really really like to get the rest of the system up and running. I do have some tools over and above my multimeter available, though I'm a novice with them. I've got a 2 channel bitscope oscilloscope, and an 8 channel selae logic analyser.

Any thoughts and advice would be greatly appreciated.

steve3000
Posts: 1711
Joined: Sun Nov 25, 2012 12:43 am

Re: Another A5000 restoration....

Postby steve3000 » Mon Nov 06, 2017 12:10 am

Well done for getting so far! It's easy to give up on these and I know from experience they can be really frustrating at times.

I repaired an A5000 with faulty floppy drive in the past, cause was IC42 in that case, but I'm sure the symptoms were different to yours. I vaguely recall the drive was detected at startup, but the motor did not spin up when accessed. So you're probably right looking elsewhere.

What IDE problems do you have? The IDE controller IC is also the floppy controller - could the problems be linked?

lcww1
Posts: 23
Joined: Wed Mar 15, 2017 11:16 pm

Re: Another A5000 restoration....

Postby lcww1 » Mon Nov 06, 2017 7:53 am

Thanks for getting back to me steve3000! I’m really enjoying working on this A5000, especially now that I’ve made some progress in getting it functioning again...

re the IDE problem - as the floppy and the IDE systems appear to be the only non-functioning systems I’d wondered whether they were linked by more than just physical proximity - I’d thought to tackle them separately but now that I know that they share a controller I’ll look into the state of the IDE traces, though superficially they look ok.

At the moment configuring idediscs to 1 with a known working ADFS formatted CF card in a working IDE adapter gives an icon on the iconbar, but also results in apparently continuous disc access according to the leds on the IDE to CF adapter. The machine hangs if I click on the HD icon

I’ll check the IDE PL9 connector traces but unless I find an obvious defect, I’m tempted to give the PCB some further cleaning, particularly around ICs 37 and 49 the legs of which which are both still a bit crusty....

User avatar
DutchAcorn
Posts: 1631
Joined: Fri Mar 21, 2014 9:56 am
Location: Maarn, Netherlands

Re: Another A5000 restoration....

Postby DutchAcorn » Mon Nov 06, 2017 12:00 pm

After checking all the tracks it is time to move attention to the logic ICs. I'd start with replacing IC49 as that is closest to the battery, but if you need to order parts you might as well get two 74HCT245s and a 7406 to replace IC37 and IC42 if needed.

You could also remove IC38 temporarily if you are worried about interference by faulty floppy drive logic on the shared IC26, and focus on the IDE interface first.
Paul

philb
Posts: 118
Joined: Sat Aug 05, 2017 6:05 pm

Re: Another A5000 restoration....

Postby philb » Mon Nov 06, 2017 8:47 pm

Does RISC OS 3.1 even check for the floppy hardware before registering the drive? I thought it was only based on the *Configure Floppies setting. But equally I would have expected a delete-power-on to reset that to 1, so I guess there must be a bit more going on than that.

ajw99uk
Posts: 108
Joined: Fri Sep 29, 2017 3:00 pm
Location: W Yorks, UK

Re: Another A5000 restoration....

Postby ajw99uk » Mon Nov 06, 2017 9:36 pm

The floppy LED flashes and there is sound of some activity from the drive on power-on. A5000 and A3000 do/did this (when I was booting from floppy in 1993, a friend suggested a wait until after that moment to push the disc home, forget why - also forget whether there was any 2.00 vs 3.10 difference).
Running RISCOS: A5000, A540, R140, RiscPC, RPi B
Running *nix: SGI Fuel & Indigo2, RPi2, x86, amd64, RiscPC, A540, R140

lcww1
Posts: 23
Joined: Wed Mar 15, 2017 11:16 pm

Re: Another A5000 restoration....

Postby lcww1 » Tue Nov 07, 2017 12:04 am

DutchAcorn - many thanks for your advice! Sounds like a good plan - looks like I’ll need to get set up for dealing with surface mount rework....slightly daunting but also quite exciting at the same time...

philb - after my first successful del-power on, *st. showed floppies set to 1, just as you thought - any attempt to interact with the floppy drive from the CLI just results in a bad drive error - so far I’ve not seen any led or motor activity from the floppy drive at all whilst attached to this A5000, but hopefully that will change!

User avatar
flaxcottage
Posts: 2797
Joined: Thu Dec 13, 2012 8:46 pm
Location: Derbyshire

Re: Another A5000 restoration....

Postby flaxcottage » Tue Nov 07, 2017 9:37 am

Just a thought ...

You have checked the HDD and FDD cables? I have had cables that were faulty. :(
- John

Currently running Level 4 Econet with BBC B, BBC B+ 128K, Master 128K, 4Mb A3000, 4Mb A3020, 4Mb A4000, 4Mb A5000 dual FDD; UK101; HP41CX setup; Psion 3a, 3mx and 5mx; Z88; TI-58c, TI-59 and printer, HP-16C programmer's calculator

lcww1
Posts: 23
Joined: Wed Mar 15, 2017 11:16 pm

Re: Another A5000 restoration....

Postby lcww1 » Thu Nov 09, 2017 8:26 pm

I had the same thought as you, flaxcottage! I’ve been testing the floppy and IDE with some new(ish) cables that work fine with my other bits of Acorn kit, rather than with the somewhat less new cables that arrived with my A5000 - I do continue to suspect that it’s some of the crusty ICs that are behind my floppy/IDE problems...... I’m hoping to make some progress this weekend......

philb
Posts: 118
Joined: Sat Aug 05, 2017 6:05 pm

Re: Another A5000 restoration....

Postby philb » Thu Nov 09, 2017 9:38 pm

Some things to look at, perhaps:

If the IDE adapter activity LED is stuck on then that might mean it is seeing itself permanently selected. Check pin 37 of PL9 which should be high except when the IDE space is being read or written.

You might try reading or writing the registers in IDE space directly so you can observe the effects on the hardware without having to worry about what the drivers are doing. For example, write a tiny program that just reads from the first IDE register (which I think is at &301007c0, but you should probably check that) in a loop and see if you can see PL39 pin 37 changing state. If that pin isn't moving, see if IC26 pins 4 and 24 are doing anything.

When you say that trying to access the IDE drive causes the machine to hang, is it absolutely and completely locked up (mouse pointer doesn't even move) or is it just slightly stuck? If the former, that might suggest that IORDY* is stuck low and IOEB is waiting forever for the IDE drive to complete an I/O cycle.

It seems a bit unlikely that problems with IC38 could cause "Bad Drive" since it looks like that chip just drives control signals out to the floppy drive and there is no way for the software to tell whether it's working. A bad drive error seems like it must be due to ADFS having decided that the floppy controller isn't there at all, which would seem to point more in the direction of IC26 itself or the signalling between it and IOEB. Again I would start by trying to read from the floppy controller address space at &30100FD0 and see what happens.

lcww1
Posts: 23
Joined: Wed Mar 15, 2017 11:16 pm

Re: Another A5000 restoration....

Postby lcww1 » Sat Nov 11, 2017 4:08 pm

Many thanks for your advice philb - much appreciated - I'll follow up on all your suggestions and report back here in due course.....


Return to “hardware”

Who is online

Users browsing this forum: helpful and 3 guests