• Please review our updated Terms and Rules here

Display problems on a Commodore Pet 8032

mrmacca said:
My collegue did look through all these chips and draw a blank. H'es got a good 15 years more experiance then me.

I can't rummage through them now as everything is at work.
Click to expand...
Never mind; if Carlsson or someone sends you a replacement 8032 CG ROM you'll probably be in business.

But we're curious: what *are* those pretty purple ones with the round labels?
 
Very Very Clever Mike

smiley-bow.gif


So Fab Forum'ers as a novice (but one who works at a Uni and obviously have stuff around which may or may not help) where do we go from here?

:) I'm really pleased to see the error replicated above Mike :)
 
Last edited:
MikeS said:
Never mind; if Carlsson or someone sends you a replacement 8032 CG ROM you'll probably be in business.

But we're curious: what *are* those pretty purple ones with the round labels?
Click to expand...

I'll relook at work and get any numbers off them. Theres a yellow, orange and blue circular ones :p

Shall I take my camera into work next week and take a pic of some of the old bits we have on display or is that just "done to death on here"?
We've got everything from Galaxy 2000's to punch card readers lol
 
mrmacca said:
Very Very Clever Mike
Click to expand...
Hey, your intuition was right on! There's a future in troubleshooting PETs waiting for you...

So Fab Forum'ers as a novice (but one who works at a Uni and obviously have stuff around which may or may not help) where do we go from here?
Click to expand...
Well, if you have the facilities to erase and (re)program EPROMs and somewhere in those drawers you find a 2716, 2532 or a TI 2516 (check the chips with 24 pins and a window) then you download the appropriate image (901447-10) from one of several sites and program a new EPROM with it. Failing that, you ask one of the generous and helpful folks on here like Anders and Steve to send you one (I would and Dave probably would as well, but it'd take a while; I assume someone closer can help).

AFAIK those three EPROM types should work (as well as an original ROM of course); if someone knows otherwise, please correct me.
 
Last edited:
mrmacca said:
I'll relook at work and get any numbers off them. Theres a yellow, orange and blue circular ones :p

Shall I take my camera into work next week and take a pic of some of the old bits we have on display or is that just "done to death on here"?
We've got everything from Galaxy 2000's to punch card readers lol
Click to expand...
We all love to drool over pictures of neat old stuff, but if there are more than two or three it might be better to put them on one of the numerous free photo sites and just post a link here. Looking forward to seeing what you've got there!
 
ok dokey, I shall ask a few guys around work if we have the technology to Programme one of the chips. That would be the coolest solution wouldn't it :)

But if they couldn't I might please come and beg one from one of you. I guess I am in no real rush, the machine is 30 years old already hey.
Next week I'll try and snap some old stuff for you and I'll load it up on photobucket, perhaps make a new thread.

thanks to you all for the help so far, Loggin out as its the end of the night.
 
mrmacca said:
where do we go from here?
Click to expand...

Alasdair,
Using the Private Message (PM) capability of this forum, send Anders your shipping address and ask for how he wants the 2UK paid (paypal, etc or maybe in EPROMs if you have the kind he wants). Address the PM to "carlsson".

Best regards,
Dave
 
Yup. I'll have a look tomorrow. Almost all my PET stuff is in the basement, both complete machines and spare items. Fortunately I have two programmers of which at least one handles 2532 so I'll be able to read any chip I can find. Quite possibly though these will be 2532 EPROMs loaded with a Swedish font, but it would be a quick fix for me to reprogram one of my blank 2532's with the International font instead.
 
I still think it'd be worth taking two minutes to try another ROM in the CG socket - as far as I can tell, any of the other ROMs on the board would do. The reason is that a broken shift register could cause exactly the same symptoms. If Mike wants to get his soldering iron out and lift a couple of inputs on the 74166 I'm sure he could reproduce the above screen shot yet again ;-) My money is also on the ROM being toast, but it would be a shame to fit a new one just to find out that it's the shift register that needs replacing (and that the original ROM was probably perfectly good from the start!).
 
cosam said:
I still think it'd be worth taking two minutes to try another ROM in the CG socket - as far as I can tell, any of the other ROMs on the board would do. The reason is that a broken shift register could cause exactly the same symptoms. If Mike wants to get his soldering iron out and lift a couple of inputs on the 74166 I'm sure he could reproduce the above screen shot yet again ;-) My money is also on the ROM being toast, but it would be a shame to fit a new one just to find out that it's the shift register that needs replacing (and that the original ROM was probably perfectly good from the start!).
Click to expand...
I second the motion. The ROM is more likely to be the cause of the problem, but it could indeed be the 74166 at UA2; removing the CG ROM doesn't really prove anything since it just lets all the data lines float high including the two that are probably stuck high already, but another ROM would most likely pull those lines low somewhere in the CG table so if the bars disappear that would indicate that the SR is OK and it must be the ROM.

Try the irrelevant UD11 or even UD10. I assume that since it failed while just running, a broken or bent pin is not the problem; be careful not to bend/break any pins now though!
 
Great thread guys!

Although I can't contribute much, these deductive exercises are educational. The solution appears close.

Tez
 
Guys,
One possibility we did not consider is a cold solder joint under the Character Generator ROM socket. But I think the "MikeS test" indicates that it would take two cold solder joints which would lower the probability of that being the problem. So I guess we all agree that if no other tests are performed, we will replace the ROM?

Tez, yes this has been fun.
 
MikeS said:
I second the motion. The ROM is more likely to be the cause of the problem, but it could indeed be the 74166 at UA2; removing the CG ROM doesn't really prove anything since it just lets all the data lines float high including the two that are probably stuck high already, but another ROM would most likely pull those lines low somewhere in the CG table so if the bars disappear that would indicate that the SR is OK and it must be the ROM.

Try the irrelevant UD11 or even UD10. I assume that since it failed while just running, a broken or bent pin is not the problem; be careful not to bend/break any pins now though!
Click to expand...

Don't forget my earlier attempt to remove each socketed chip at a time I'll repost incase the line with UD11 was relevant

I removed and reseated all the mid sized chips to no avail.

I'm leaving out UD11 as it seems to be irrelevant, (and its still out)
Take out UA3 and the whole screen fills totally green. Put it back in same two lines per character.
Take out UD 10 - The Green Lines remain but I loose the basic 4.0 output leaving just the lines.
Take out UD 9, 8 or 7 and you get nothing at all from the machine. (and the screen is totally empty without the green lines either)


I'm concerned about that chip that was at UD11, it had a soldered pin on it and some hand written writing. The machine obviously has had more owners then myself in its time. I'm a novice but do worry about moving chip locations incase it ends up with the whole board going pop!

I'd be really happy to take Anders kind offer and will send him a short PM.
 
Last edited:
No to worry about the PROMS in UD11 (address hex A000) or UD12 (address hex 9000), they are for user installed machine language programs and are not part of the operating system. The code there can not be executed without being called with a SYS command. Mike was just suggesting that the chip at UD11 would be a good candidate to drop into the character generator spot as it is pin compatible and is not needed by the operating system. The actual code in the chip was irrelevant for this test. But if you think the chip may have a problem, do not use it. It is OK to leave it off the system board.

The soldered pin may have been attached to the chip as a test point to clip-on a scope lead.

I agree, at this point in the testing, it is best to make arrangements with Anders. For a few quid, it is a very good deal. The 2532 EPROM is worth more than that without the programming.
 
dave_m said:
No to worry about the PROMS in UD11 (address hex A000) or UD12 (address hex 9000), they are for user installed machine language programs and are not part of the operating system. The code there can not be executed without being called with a SYS command. Mike was just suggesting that the chip at UD11 would be a good candidate to drop into the character generator spot as it is pin compatible and is not needed by the operating system. The actual code in the chip was irrelevant for this test. But if you think the chip may have a problem, do not use it. It is OK to leave it off the system board.

The soldered pin may have been attached to the chip as a test point to clip-on a scope lead.

I agree, at this point in the testing, it is best to make arrangements with Anders. For a few quid, it is a very good deal. The 2532 EPROM is worth more than that without the programming.
Click to expand...

If it helps the diagnosis I can give it a go... If there are any chips I could help with alternatively if you think I may have some here that people could use? I'll try and note down some I have.
:D
 
Actually I went through my boards this morning and noticed they all have 2716 EPROMs. This makes sense since the PET character set is only 2K and the reverse half is generated by the CPU, unlike later VIC/C64 which has a 4K character set. Since I already have piles of 2516 and 2716 chips, I will take one of my loose ones and reprogram with the proper character set instead of pulling a chip from a board - in particular since that chip would hold a slightly different character set than what our friend ultimately needs.

And yes, I got a PM which I will handle shortly.
 
Great news, all went smoothly with the PM :)

Thanks Anders, makes sense to swap for exactly the right character set hey. Its exciting breathing life back into these old girls isn't it :cool:
 
carlsson said:
Actually I went through my boards this morning and noticed they all have 2716 EPROMs. <snip>
Since I already have piles of 2516 and 2716 chips,<snip>
Click to expand...
That's why I said to look for 2716s and TI 2516s as well as 2532s; this discussion seemed to assume that the CG ROM is a 4K 2332/2532 type instead of the much more common 2K 2716 type (probably because of the suggestions to replace it with a 4K ROM from elsewhere for testing).

I doubt that you have any, but watch out for TI's non-standard 2716s; on the other hand if it weren't for their non-standard 2532 we wouldn't have straight replacements for the 2332s in the PETs and elsewhere...

http://www.tomshardware.com/forum/80532-13-what-diff-eprom-eprom
 
While we're on the 2K/4K topic:

Has anyone ever read or done anything about exploiting the alternate character set that's provided for in the 4032/8032 PETs?
 
You must log in or register to reply here.
Back
Top