Odd signs of a failing QL??

[RWAP]

My main test QL is showing various signs of going wrong...

It could be the QL, the power supply or the Trump Card I use...

For quite a while mdv2 has not worked for reading or writing - even if I swap the unit for a known working unit.

I didn't bother investigating further, as I just tend to use mdv1 for reading cartrridges (for the preservation project), as this has always been very reliable and good at reading poor cartridges.

I also sometimes format a cartridge in mdv1 and copy files onto it (from disk), it appears to work and FORMAT ram1_mdv1 gives me 255/255 sectors with no corrupt files (marked with * for those who are not used to this command). However, for some unknown reason, every now and then when I reset the QL and try to run the program on the newly created microdrive, the program may appear to load correctly but then display a blank screen.

The same cartridge shows the same problem when loaded on another QL setup (no trump card).

If I make 5 copies of the same cartridge, this may only affect 2 of the 5 cartridges (and the two can be in any order).

The WREN command can also lock up the QL!

Then today, some even odder behaviour - I was preserving some cartridges and was doing quite well - but then I noticed the PAUSE command and the CLS command both returned BAD NAME error and did not get converted to upper case...

WCOPY, FORMAT and other file commands seemed to work ok.

I reset the QL and PAUSE and CLS were working again.

However, having copied the files to disk with no errors, when I did a directory of the disk, it only had the original files on the disk...


Any ideas.....

[1024MAK]

Have you run any RAM test? If the first run does not show any errors, leave it going for a while...

Tried swapping ROM's? Tried swapping the CPU?

Could be a dry joint - known to give all manner of strange intermittent faults...

Mark

[tofro]

Rich,
could be anything....

the latter symptoms hint a bit towards corrupted memory, the former more towards a power supply oddity. But maybe (1) is caused by (2) or some other issue, like bad contact in sockets or connectors.
I'd remove the TC and check whether the symptoms are gone, then investigate from there. Memory corruption in internal memory could also be a defective or weak ULA.

(I suppose it's not a heat problem, we still have like 38° over here during the day, but it's likely to be different in the UK, I guess?)

Tobias

[RWAP]

I tried it with a Minerva and it was working (not got my Minerva at the moment)...

I have changed the ROM chips and CPU but all to no avail...

I think it is probably a dry joint somewhere but how you find it is beyond me!

Removing the Trump Card seems to fix the problem but then most of the problems have always been linked to stuff I need the Trump Card for (ie. copying to / from disk). The Trump Card seems to work ok in a second QL...

It could of course be a dodgy pin on the expansion port but without a spare Trump Card, I can't test it!

[1024MAK]

I think it is probably a dry joint somewhere but how you find it is beyond me!

Only one way I know of...

Carefully examine each and every joint using a good light and a magnifying glass. Examine the joint from at least two or three angles. Mark any that look poor with a marker pen. Once you have checked the whole board, resolder all the poor joints (remove the old solder before resoldering). Then repeat...

If you find that you have resoldered the whole board, it's not a dry solder joint

Mark

[jose_leandro]

Hi
I've seen some faulty QLs that gave me a bad impression about QL hardware.

The last two faulty QLs had similar symptoms:

They started well but they hangs after few minutes. If you made a reset they didn't start or they started more slowly or showed start messages two times,etc,etc.

If I changed the two roms with a eprom or put ( in my case) a qubide , I saw that they worked a bit different than if they didn't have these elements. I also see that one of them worked better without keyboard.

At the end of they life, one didn't start anymore and the other hangs at few seconds.

And the worse thing is read the service manual and found this:

5. how to cure a crashing QL
5.1 you may have a QL that "freezes" after a few hours of operation. We don not yet fully understand why this happens with some computers.

........ ok, no problem.

Anyway, these two QL started to work well after change the CPU (both), one 8049 and one CLA.

I have the impression that some signals are a bit critical and can cause problems if some chips are not in best conditions (electric characteristics). Also I thing it can be a little problem with "current" and it can explain why there are more fault when you connect a interface.

But It's only my impression. I don't see these things with a faulty QL and my scope.

bye

José Leandro