News:

Printed Amstrad Addict magazine announced, check it out here!

Main Menu

Dead PcW 9256?

Started by squelch41, 23:55, 12 March 24

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

squelch41

Hi,
I got a PcW9256.
I got it to boot once into locoscript but it just seems to have died.
Originally, if I turned it on with no disc would produce 3 beeps after a few seconds (which is normal)

However, now I get a white screen and the disc spins but there seems to be no attempt to load and I don't get the three beeps to indicate bad disc.

I get the sync pulses and the video signal on the video connector.
I have tried taking the system board out and powering it from a 5v power supply but same problem.

Clock signal on the z80 is fine and steady at 4MHz. I can see activity on the data bus and address lines.
Reset line behaves normally.
No chips get hot.

1 capacitor had leaked but it was on the 12v line and I have replaced it - unsurprisingly no difference.

Jumper is set correctly for 256k ram.

Can anyone help point me in the right direction?

robcfg

I'm not an expert on the 9256, but while one pops up, maybe the service manual can be of help: https://habisoft.com/pcwwiki/lib/exe/fetch.php?media=manuales:servicio:manual_servicio_tecnico_pcw_9256.pdf

And in case you don't have, here's the user manual too: https://habisoft.com/pcwwiki/lib/exe/fetch.php?media=manuales:descargas:manual_amstrad_pcw_9256.pdf

Good luck!

JohnElliott

Unusually for 3.5" drives, the Citizen drives on the second-generation PCWs are belt driven, so it's possible that the belt has stretched and needs replacing. (Or the entire drive to be replaced, perhaps with a Gotek).

squelch41

#3
I think the problem is the logic board - it no longer gives the 3 beeps to say no disc recognised and same with the drive disconnected.

So, I think the system isnt getting to the point where it is looking for the disc.

squelch41

Is it normal for the IRQ pin (pin 16) of the Z80 to be held low?
It briefly goes high as the reset signal is high, then drops and is held low - seems to imply the Z80 is waiting for an interupt to come? It is connected directly to the gate array with just a pull up resistor a far as I can see from the schematic.

NMI (pin 17) is high

IROQ (pin 20) looks like it's doing normal things? - screenshot attached

Could this point to a bad gate array if it is holding the IRQ pin low all the time?

squelch41

Still none the wiser, but now have an 8256 motherboard (as couldnt find another 9256 on ebay!)


On the working 8256 motherboard, there is activity on all address lines all of the time.
On the non working 9256, there is no activity on the upper lines (Think A10-15 from what I remember)  most of the time, though at power on, there is a second or two of activity just as the reset line has deactivated.

Also, A8 on the RAM is much quieter on the 9256, there are brief strobing like signals but they are every second or so, vs all the time on the 8256.

Desoldering the link between the A8 drive on the gate array and the ram results in no change on the A8 pin on the gate array (ie doesn't seem to be bad ram holding up the gate array)

Only the 8256 ha a socketed Z80 but if you remove the pins of the higher address lines (the ones that are quiet on the 9256) a) the system error beeps and b) the pin of the z80 shows activity and there is on activity on the connected pin on the gate array - implying that the signal is driven by the z80.

BUT.... I cant figure out why the 9256 wont 'post'

Can you see what I am missing?

I am wondering if the 9256 has a bad z80 -- but then I'd have though there would be no activity at all on the higher address lines on a floating address line.

The 9256 z80 is soldered so just trying to think through before I risk damaging the mainboard by desoldering it!

Tried piggybacking the other z80 but that (unsurprisingly!) didn't work

squelch41

For anyone who finds this in the future - a bad floppy controller chip seemed to be the issue.

Didnt see how it could be so swapped z80, and printer controller chips 1st with those from a 8256 board - no change.

disconnected the 5v from the floppy controller chip to see if it was corrupting the bus - no change, still dead.

Thought might as welll swap it out and hey-presto, 3 beeps

Connected the floppy drive and am booting into locoscript

Bryce

It might be still a capacitor problem on the mainboard and the controller chip just happens to be the first chip that has an issue with the dodgy power rail.

Bryce.

squelch41

Quote from: Bryce on 19:30, 20 March 24It might be still a capacitor problem on the mainboard and the controller chip just happens to be the first chip that has an issue with the dodgy power rail.

Bryce.
I replaced the electrolytics prior, just in case.
the 12v one was leaking but the one on the 5v rail looked fine and tested fine once removed (I put a new one in it's place)

Bryce

Quote from: squelch41 on 19:47, 20 March 24
Quote from: Bryce on 19:30, 20 March 24It might be still a capacitor problem on the mainboard and the controller chip just happens to be the first chip that has an issue with the dodgy power rail.

Bryce.
I replaced the electrolytics prior, just in case.
the 12v one was leaking but the one on the 5v rail looked fine and tested fine once removed (I put a new one in it's place)

How did you test it? Most hobbyists don't own an ESR meter unless they are seriously into electronics.

Bryce.

squelch41

#10
Quote from: Bryce on 21:23, 20 March 24
Quote from: squelch41 on 19:47, 20 March 24
Quote from: Bryce on 19:30, 20 March 24It might be still a capacitor problem on the mainboard and the controller chip just happens to be the first chip that has an issue with the dodgy power rail.

Bryce.
I replaced the electrolytics prior, just in case.
the 12v one was leaking but the one on the 5v rail looked fine and tested fine once removed (I put a new one in it's place)

How did you test it? Most hobbyists don't own an ESR meter unless they are seriously into electronics.

Bryce.
an ESR meter and also measured capacitance on my multimeter.
Plus, having changed the capacitors, didnt fix the fault.

That said, for a future finder of this post, is certainly something to check as a lot easier than desoldering a 40 pin IC

Powered by SMFPacks Menu Editor Mod