Spectrum Computing

Ok

So black is blackish
Blue is a very dark green
Red is fine
Magenta is a little lighter red
Green is bright green
Cyan is bright green
Yellow is yellow
White is also yellow

I have replaced tr9 and tr8 from a donor bored and still the same and have also now replaced tr1 and tr2 still the same

Time to change the ULA.

Mark

Thanks mark

Is it worth change x1 and x2 as well or will this not help?

Also where can I get a replacement?

98bishopmal wrote: ↑Mon Aug 02, 2021 8:29 pm Also where can I get a replacement?

Unfortunately, from another speccy. There are modern reproductions, though, but then you're getting into the whole Theseus's ship thing.

Thanks Mark,

Managed to find one on eBay will let you know how it goes

98bishopmal wrote: ↑Mon Aug 02, 2021 8:28 pm Is it worth change x1 and x2 as well or will this not help?

I don’t think that will help.
If X1 was faulty, you would not get any stable screen display.
If X2 was faulty, as in not working at all, you would be no colours at all, just a monochrome picture. And if it was slightly off frequency, all the colours would be wrong.

It’s the analogue outputs from the ULA that control IC14. But this chip is normally fairly reliable.

Mark

Regarding replacement (original types not 'modern replacements') see:

Here , please scroll down to the " What ULA works in what machine ? " section

and also Here


As I see it, in a brief sentence, as per the first link)

... The 6C001E-7 ULA is backwards compatible with every 48K machine from the Issue 1 right up to the Issue 6A ...

..and the modern replacement that is apparently the currently recommended one.

spider wrote: ↑Tue Aug 03, 2021 8:36 am Regarding replacement (original types not 'modern replacements') see:

Here , please scroll down to the " What
As I see it, in a brief sentence, as per the first link)

... The 6C001E-7 ULA is backwards compatible with every 48K machine from the Issue 1 right up to the Issue 6A ...

Each ULA version can be used in the board issue it was intended for and any earlier issue boards. In other words, it’s backwards compatible with earlier boards.

The 5C102E ULAs were used in issue one and some issue two boards. The ‘dead cockroach’ daughter board is required. The transistor ‘spider’ should also have been fitted, but at the time the problem that this fixes had not been discovered.

The 5C112E ULAs were used in most issue two boards. The ‘dead cockroach’ daughter board was no longer required, the design of the ULA having been modified. The transistor ‘spider’ was however required.

The 6C001E-5 ULAs were used on some late issue two boards. The transistor ‘spider’ was however still required. May also be found in some issue three boards.

The 6C001E-6 ULAs were used on issue three boards. An additional transistor ‘spider’ was no longer required on issue three and later boards, as it became a permanent part of the PCB circuitry. This ULA version may also be found on some issue two boards. All issue two boards still need the ‘spider’.

The 6C001E-7 ULAs were used on issue four and later boards. An earlier ULA version is likely to cause unreliable operation in issue four and later boards due to a design change with both the ULA and the circuitry on the boards. However the 6C001E-7 is still backwards compatible with earlier boards.

Mark

Sol_HSA wrote: ↑Tue Aug 03, 2021 9:03 am ..and the modern replacement that is apparently the currently recommended one.

There is also the NebULA

Mark

Think i have messed up

I removed the old ULA and put the new one in and now get yellow boarder and black paper, reboot a couple of times and get pure yellow screen and white cursor on bottom right of paper.

Any help would be great

98bishopmal wrote: ↑Thu Aug 05, 2021 9:58 am Think i have messed up

I removed the old ULA and put the new one in and now get yellow boarder and black paper, reboot a couple of times and get pure yellow screen and white cursor on bottom right of paper.

Any help would be great

It won't help (me) help you but just for clarity for 'the experts' what was the part number of both the "old" and "replacement" ULA's please ?

Unsure if a pic of "it now" with the "cursor" in the right hand (!) corner will help or not, if its convenient it won't do any harm to post a pic of the screen I guess, never know might give thoughts.

I was going to say try the "old" one back to see if there's any pattern change (as in something else amiss) but I'd say hold off for now until the tech experts appear hopefully.

It may sound stupid sorry but you were careful to not damage or 'fold under' any pins when you fitted it yes ? I have seen this on other machines where ROMS' were swapped over and its dead easy to accidentally get a pin bent under in a moments inattention, been there myself!

Yep, as Spider says, we need to know the part numbers of the original ULA and of the replacement ULA.

DO check that each pin is okay. And yes a photo of screen showing the symptoms would be helpful.

Remember, fault finding is a process of elimination.

Mark

Thanks both,

ULA is the same as the old one 6c001e-7

All pins are connected, I have put the old ULA in and getting the same

This is what I see, sometimes the paper is just black



Many thanks

Thanks both,

ULA is the same as the old one 6c001e-7

All pins are connected, I have put the old ULA in and getting the same

This is what I see, sometimes the paper is just black



I am also getting this sometimes



Many thanks

Is the border always that colour? Power down, wait ten seconds, power up gain. Do this at least five times. Does the border always end up being that colour?

Mark

Hi Mark,

Yes the border is always that colour, I have done as requested and still doing the same.

Before the system was working fine but with yellow boarder and incorrect colours so did the following to try and fix it.

Replaced TR1 and TR2 (same result)
Replaced TV module (same result)
Replaced TR8 and TR9 (same result)
Replaced ULA (now having the problem where does not show copy write message and sometimes gets stuck at black paper)

Both ULAs are 6C001E-7

I have replaced the ULA with the old working one and getting the same.

Many thanks

Ok so I have disabled the upper ram and this now shows the copy write message

But still has yellow boarder and colours are not correct

98bishopmal wrote: ↑Sat Aug 07, 2021 3:10 pm Ok so I have disabled the upper ram and this now shows the copy write message

But still has yellow boarder and colours are not correct

That sounds positive to me. Until the tech members get back to you, now its showing the (c) screen, do the keys work ? Are you able to try any simple programs perhaps to type in or even perhaps load a tape ? If you do the latter it would have to be a 16K title.

But I guess anything would do for a basic "load" test if needed, I'd probably at this stage want to see if the keys worked and what you get with setting different colours (BORDER x PAPER x etc) again at this point.

Thanks spider,

It’s the same colours as before

I asked the question about the border, because one of the first things the ROM code does, is to TR6 to set the border to white. So if the border always turns white (yes, I know yours is yellow) then the ROM and the Z80 are likely okay.

Which you have proven by finding that with the upper DRAM disabled, you are now getting the copyright screen.

So as you have likely worked out, there are two faults; an upper DRAM fault (we will leave that for now) and a colour encoding fault.

I suggest next you renew IC14 (LM1889). But before you do, again with the screen displaying a clear screen, measure the voltages on all the pins of IC14 (with the black meter lead on any convenient 0V/GND point, such as the metal heatsink tab of the 7805 voltage regulator) listed here:
IC14 (LM1889)
Pins 2, 3, 4, 12, 13, 14, 15 and 16

List your results here on the forum.

Mark

Pin 2, 4.95
3, 4.45
4, 4.97
12, 11.67
13, 11.67
14, 11.67
15, 4.31
16, 11.66

Chip has already been change and tested working on another board

If I disconnect a leg from C65 I get normal grayscale so white shows perfect then

98bishopmal wrote: ↑Sat Aug 07, 2021 10:30 pm Chip has already been change and tested working on another board

Ahh, yes, you said that you had changed it in your first post

98bishopmal wrote: ↑Sat Aug 07, 2021 10:30 pm If I disconnect a leg from C65 I get normal grayscale so white shows perfect then

Yes, that’s because C65 is the coupling capacitor that carries the colour carrier signal from pin 13 of IC14.

I’ll have a look at the voltages you posted tomorrow.

Mark

Actually, can you check those voltages again please. Pin 15 is definitely wrong.

Mark