"(c) 1982 Sinclair Rese" only on boot

[rich_r]

Hi there - not the best first post, but I guess a lot of people end up here when they run out of ideas and need to ask those with experience. Long story short, I had a ZX81 when I was 8, then moved on to a Spectrum in 1983 when I was 10. It's been in a box since I went to university until Covid when I decided to get all my old computers running again. I have a degree in electronics so might as well use it, and I taught myself Z80 assembler on the Spectrum before moving on to 68000 on the ST and Amiga and using that again at university. I still have my Philips 15" TV from 1988, so no issues with display.

ZX81 just needed a new keyboard membrane, Amiga needed a floppy drive replacement (and will need a recap soon), Atari ST just seemed to work with no problems. But my trusty Spectrum Issue 3 wasn't so happy, so I put it aside. Last month I decided to give it another go. Bought a known faulty Issue 3 board off eBay to practice on (burnt a few tracks discovering my solder wick was rubbish. Bought decent stuff). In the end I think my original has too many upper RAM faults and I didn't fancy unsoldering all the chips given how easily I damaged the practice one - so I bought a known good Issue 6A board off eBAY that had already had the electrolytics replaced. I replaced the 7805 with a modern switched mode regulator and also a DivMMC Future to save my patience loading from tape.

It's been working beautifully for a few weeks, and I've been enjoying using it again. But this morning I turned it on and got the following on the screen.



It appears to initialise every power on (or reset via the DivMMC reset button), white border, black paper that then drops to white, but the copyright message is messed up, there's a low beep from the speaker and the keyboard doesn't respond. I checked the voltages and all looks good there. Same results with the DivMMC removed.

This doesn't look like a lower RAM fault as the initial boot looks OK and the text seems fine, but as it doesn't boot I'm not sure how to test. My initial thoughts are ULA as it looks like it's messing up reading the screen memory. But I guess it could still be display RAM being corrupt due to a failing 4116.

What should I check next? I'm guessing I can't use one of my Issue 3 ULAs in an Issue 6A board or can I?

[1024MAK]

ULAs are backwards compatible (you can use a later ULA in an earlier board), but not the other way round. Sinclair discovered and fixed a timing problem in later issue boards. That doesn’t mean that the older ULA won’t work. It’s just that it’s not certain to work.

Back to your fault. I don’t think your ULA has failed. When a ULA does fail, the symptoms are normally completely different to what you are seeing. Ignore all the bull on the internet, ULA chips are more reliable than people think (although they do fail from time to time, especially if the edge-connector is abused). The most common failure in 1970s, 1980s and 1990s computers are the RAM chips.

The ROM does a test to see how much RAM is available. This is far from a proper test of the RAM. Hence certain failures will not be detected. So it’s possible for the machine to think it has 48K of working RAM, and put the Z80 machine stack in at the top of memory.

But faulty “upper” RAM will corrupt the data on the stack, which will then cause the ROM code to crash at some point.

This is what I think is the most likely scenario for what you are seeing. But there are other possibilities. A partial ROM fault being a possible but unlikely cause (assuming it’s a normal mask ROM chip, that is flat plastic package with no window).

I think it may be worthwhile disabling the ‘upper’ RAM, if only to eliminate it to narrow down the problem.

The ‘upper’ DRAM chips are IC15 to IC22. These do not always run hot when faulty.

You can disable the upper DRAM. Disable upper RAM. Worth a try...

In regards to removing chips, other than irreplaceable parts that you may want to reuse, the easiest safe way if you are not good at desoldering, is to cut each leg of the chip. Then you can gently pull the individual pins out while heating the joint. But never use any force or you will damage the through-hole plating or lift PCB tracks/pads.

I find a solder sucker / desoldering pump better for clearing pads on multilayer boards.

Mark

[rich_r]

Many thanks Mark.

I tried the 6C001E-6 ULA from my Issue 3 in the Issue 6A (6C001E-7), as I thought it worth a try (having done a bit more research and finding that it probably isn't compatible but might be enough for testing), and yes, it boots but to the same fault. I thought I was reasonably competent at soldering, but it appears that I'm not as competent at de-soldering. I have a mechanical solder sucker, but don't seem to have developed a reliable technique of removing solder from holes. More practice needed I guess.

So next step will be to disable upper RAM, if I can find how to do that on an Issue 6A (no separate decoding chips) and see what happens.

[1024MAK]

Ahh, I was rushing to answer earlier.

There is no easy way with an issue 6A board unfortunately.

Unless the ‘upper’ DRAM chips are in sockets…

Mark

[rich_r]

Sadly nothing is socketed other than the ULA (otherwise I'd have whipped the RAM out and tried replacements I have a couple of spare 4164s). I guess I'll just have to unsolder them one at a time and replace with sockets until I find the fault. Good desoldering practice anyway

[1024MAK]

Before you do that, do you have any expansions or interfaces that have/or replace the ROM?

Or if you don’t, are you happy to buy a diagnostic card?

Using a diagnostic ROM is by far the quickest and easiest way of testing the system. As long as the Z80 is running (which it must be on your issue 6a board).

Mark

[rich_r]

Apologies for the long delay - work and stuff got in the way

Anyway, I have managed to borrow a diagnostics card from someone at work, it appears to be the one myretrostore.co.uk sell, based on the Dylan Smith design. It has eight LEDs and a couple of jumpers. I believe I can change it to Brendan Alford's 0.38 diagnostic rom, but I need a working Spectrum for that Unfortunately I've struggled to find any documentation for what the tests carried out are and what the LEDs indicate.

So plugging this board in, I get the following:

LED 1 flashes 4 times, with a blue border and black screen.
LED 2 flashes 4 times, with a blue border and black screen.
LED 3 flashes 4 times, the screen changes to a blue border with flashing black and white vertical stripes.
LED 4 flashes 4 fines, with a blue border and white then black then white screen.
LED 5 flashes 4 times, with a blue border and random characters on the screen, then changes to a red border with random characters.
LEDs flash 11001101 four times.
LED 6 flashes 4 times, with a red border and random characters, then changes to a black screen
LEDs flash 10101010 four times, then show that pattern solid until powered off.

Does this shed any light?