Question about a non-working ZX Spectrum

Nagging hardware related question? Post here!
User avatar
bwinkel67
QL Wafer Drive
Posts: 1187
Joined: Thu Oct 03, 2019 2:09 am

Re: Question about a non-working ZX Spectrum

Post by bwinkel67 »

Thank you...I will have to re-check the RAM. I popped each chip into (a single test socket of) the ZX81 RAM pack and booted the ZX81 to the K cursor but perhaps that doesn't thoroughly test the memory.


User avatar
JupiterJones
ROM Dongle
Posts: 16
Joined: Wed Feb 10, 2021 1:45 pm
Location: Barcelona

Re: Question about a non-working ZX Spectrum

Post by JupiterJones »

bwinkel67 wrote:Some progress. I replaced the Z80 and now I get this. So this seems to be part of the startup sequence where the ROM is checking how much RAM you have (not checking if the RAM is ok) but then doesn't continue to clear the screen and put up the "(c) 1982 Sinclair Research, Ltd" and cursor. It does have some distortion (like maybe that's the splash text trying to print to screen). Does this suggest the ROM is good as well? If so then the only thing left is a couple of logic chips.

Spectrum-screen2.jpg
Hi! This looks like some memory issue... Although you mentioned that you already checked that, I'd double check. If you can get a Dandanator and install Brendan Alford's system check that would help.
Power supplies may also have to do. Just check the two transistors depicted in following link:
https://retrocomputing.stackexchange.co ... y-voltages
I'd also replace capacitors. As explained in previous link it seems that memories need a specific wake-up timing and failing capacitors might affect this.

There is also some gluing logic that may affect. Its construction depends on motherboard issue. For the newer ones, there is a custom chip that has all that logic programmed inside, the PCF1306 / ZX8401, but for the older issues there are a bunch of 74LSxxx ICs performing that gluing function.

You probably already got the schematics. If not this is a very good place to get them: https://spectrumforeveryone.com/technic ... cs-layout/

Hope this helps!


User avatar
Pr0f
QL Wafer Drive
Posts: 1298
Joined: Thu Oct 12, 2017 9:54 am

Re: Question about a non-working ZX Spectrum

Post by Pr0f »

Testing the RAM chips is a good idea, but only tells you the RAM chips are working - if the issue lies on the board itself, then you won't see the fault with the RAM chip out of the spectrum.

Power supplies, glue logic, even just something stuck on the PCB could cause the issue.


User avatar
bwinkel67
QL Wafer Drive
Posts: 1187
Joined: Thu Oct 03, 2019 2:09 am

Re: Question about a non-working ZX Spectrum

Post by bwinkel67 »

I double checked the RAM today...replaced the RAM with other working RAM and the same screen (nothing changed). I've looked closely at the PCB so it's not anything obvious. There are two 74LSxxx logic chips next to the ULA that will likely be my next focus. I'm assuming the ROM, ULA and Z80 are working since I replaced the Z80 and it started to behave differently and seemingly execute part of the ROM (startup to check how much RAM it has) -- and the display seems to be reasonable for working ULA.

I did test all of the caps and they seemed to behave themselves but maybe I should just replace them. I have a regulated 9 volt power supply (and have been advised to use a regulated one for issue 2). Also checked all 6 transistors and they seem fine...even the one on the CPU (not sure why it has one on the CPU...some Issue 2 board pics don't have that...mine is a 16K board.


Maskenlos
Over Heated PSU
Posts: 138
Joined: Sat Nov 03, 2018 12:14 pm

Re: Question about a non-working ZX Spectrum

Post by Maskenlos »

Hi,

I have repaired about 8 Spectrums during the last 20 years, not so many and I am far away from being an expert:

7 of them failed due to missing -5 or 12 volt. 6 out of this 7 has had defect TR4 or TR 5. One has had a dried out cap in the DC DC converter circuit.
The remaining one has had a faulty ULA. This ULA was still able to generate a picture. The attached diagnostic board shows all the time a different bad RAM chip and the picture was quite similar to yours. At this dedicated Spectrum I changed also the Z80 before the ULA.
Also the output of your 5V regulator seems to be to low with 4.85V only, but this is probably not the root cause.
My bet goes to the ULA, but I would do the following:

1. Change 5V regulator, output must be 5V stable
2. Change all electrolytic capacitor there are only a few, its worth the effort
3. Check all address lines with an Ohm-Meter. I heard about an open line

If all this does not help, try to find someone with a working ULA


Stephan


User avatar
bwinkel67
QL Wafer Drive
Posts: 1187
Joined: Thu Oct 03, 2019 2:09 am

Re: Question about a non-working ZX Spectrum

Post by bwinkel67 »

So I've read for issue 2 boards that you have a 10% tolerance with the -5/12/5 volts and I'm showing 4.69/11.73/4.85 which are within that. I've also read that specifically for issue 2 that -4 volts is sufficient. Every video I've seen on repairs of Spectrums has shown a slight drop at the 5 volts coming out of the voltage regulator, even with a non-regulated power supply. But someone from the ZX-TEAM group in Germany who has fixed lots of Spectrums (I think he worked with someone on this forum way in the past on a QL product) also said for issue 2 boards it is wise to use a regulated power supply as there is missing circuitry to handle the initial spike of the unregulated which could eventually lead to a fault.

Good to know about the ULA perhaps being the culprit. That was my initial guess because it gets very hot but others have suggested it is showing that it is working. In the end I may have to find a working one. My next focus will be replacing the caps (they all past the resistance tests flyingly). Then it's those two small logic chips next to the ULA.

I suppose it could still be the ROM as well but it seems to be instructing the new Z80 CPU to do stuff. Also, with the CPU having failed, does that point to other chips in the vicinity failing? In other words, on those occasion when the CPU has been fried, what else is likely to go at the same time?


User avatar
bwinkel67
QL Wafer Drive
Posts: 1187
Joined: Thu Oct 03, 2019 2:09 am

Re: Question about a non-working ZX Spectrum

Post by bwinkel67 »

Maskenlos wrote:Hi,

I have repaired about 8 Spectrums during the last 20 years, not so many and I am far away from being an expert:

Stephan
One question, that transistor on the CPU, is it needed? I only see it with 48K issue 2 boards in online pics. Mine is only 16K (in fact, neither voltage jumper has ever been set next to the ULA).


User avatar
Pr0f
QL Wafer Drive
Posts: 1298
Joined: Thu Oct 12, 2017 9:54 am

Re: Question about a non-working ZX Spectrum

Post by Pr0f »

I think it is a fix for IORQGE - as that signal wasn't good from the ULA for some batches - but it's been some time since I've looked at the old boards


Martin_Head
Aurora
Posts: 847
Joined: Tue Dec 17, 2013 1:17 pm

Re: Question about a non-working ZX Spectrum

Post by Martin_Head »

Any chance that you may have damaged one of the through plated holes in the PCB when removing IC's.

As you have changed the symptoms, maybe you have fixed the initial fault, and just added a new one.


Maskenlos
Over Heated PSU
Posts: 138
Joined: Sat Nov 03, 2018 12:14 pm

Re: Question about a non-working ZX Spectrum

Post by Maskenlos »

One question, that transistor on the CPU, is it needed? I only see it with 48K issue 2 boards in online pics. Mine is only 16K (in fact, neither voltage jumper has ever been set next to the ULA).
Yes it is needed to my understanding but I have never seen a 16k Spectrum

I've also read that specifically for issue 2 that -4 volts is sufficient
This is correct. From my experience even -3.5V would be still ok.

Stephan


Post Reply