17th Apr 2022, 3:53 pm | #61 |
Nonode
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,587
|
Re: Perth PET
The chequerboard patterns evident in both screen photos are certainly indicative of problems with the video RAMs or associated logic/buffer chips which you have already socketed and replaced. Poor connections/contacts again?
Incidentally if I'm looking at the right schematic the video RAMs are UF7 and UF8. Alan |
17th Apr 2022, 4:21 pm | #62 | |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,813
|
Re: Perth PET
I shouldn't be trusted with a schematic.
I'll check the continuity on the UF7/8 chips/sockets and look back on what we did with them in the big thread. Incidentally, if I switch the Tynemouth to Basic 4, I get the attached screen who looks to me like it is trying to put the normal Commodore Basic 4 message on the screen but it's getting corrupt (although I could and have been wrong before). Colin. Quote:
|
|
17th Apr 2022, 11:22 pm | #63 |
Pentode
Join Date: Oct 2021
Location: Falkirk, Scotland, UK
Posts: 140
|
Re: Perth PET
The screen shot is very similar to what I found on my Pet (before the power connector issue which took out the processor). On mine it seemed (as Alan has mentioned) to be the video RAM which was at issue. I did replace mine, and it seemed to fix that problem.
Tim. |
18th Apr 2022, 10:47 am | #64 |
Nonode
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,587
|
Re: Perth PET
For ease of reference I’m attaching copies of the 2001N board layout diagram and schematics. Also these are the chips Colin has previously socketed & replaced: UA7, UA8, UB2, UB3, UC3, UC4, UC9, UE7, UE8, UE10, UF7, UF8, UF9, UG5, UH1, UH3 & UH6
Alan |
18th Apr 2022, 11:33 am | #65 |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,813
|
Re: Perth PET
OK - for simplicity, I have removed the Tynemouth board from the equation and put back the previously known working 6502. In that instance I get the garbage screen ,not the checkerboard screen.
With power off, I have checked continuity both in and out of UF7 and UF8 and everything checks out OK. I will now go back to the tests we did of UF7 and 8 with my scope to see if there is any difference to what we found before. Please shout if there are other tests I should do. Colin. |
18th Apr 2022, 12:28 pm | #66 |
Nonode
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,587
|
Re: Perth PET
Removing the Tynemouth board for now is a sensible move but I think we need to make a distinction different types of garbage screen. The first image below is the random character garbage screen. Notice how the characters appear completely at random with the full range of different characters displayed. The second image is your first photo from post #60. It’s certainly garbage but is quite different from a random character screen and is indicative of a problem with the video RAM circuit. Worth checking the associated chips UE7, UE8 & UF9 for continuity as well. The scope checks are definitely worthwhile too. Alan
|
18th Apr 2022, 12:35 pm | #67 |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,813
|
Re: Perth PET
Here's a photo of the screen. I will go check UE7, 8 and 9.
Colin. |
18th Apr 2022, 4:39 pm | #68 |
Nonode
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,587
|
Re: Perth PET
The characters still don't seem to be entirely random as there are chequerboard blocks here and there. How did you get on with UE7, UE8 and UF9? Note that it's UF9.
Alan |
18th Apr 2022, 7:40 pm | #69 |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,813
|
Re: Perth PET
E7, E8 and F9 all check out ok from a continuity pov in and out.
It's interesting that the top half of the screen does look random and the bottom half has far more checkerboard in it. So - if continuity is OK, do you have a view on what waveforms I should start with using my scope? Colin. |
18th Apr 2022, 8:29 pm | #70 | |
Nonode
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,587
|
Re: Perth PET
Quote:
In the meantime can you confirm that the character screen is more or less unchanged? If it is that would suggest there's still an issue with the video RAM circuit. As a double check you could try Daver2's PETTester EPROM again. If things are in order PETTester should initiate the VDU test described in the manual. Alan |
|
18th Apr 2022, 9:05 pm | #71 | ||
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,813
|
Re: Perth PET
Similar at each boot, but not exactly the same - attached.
Sadly the PET will not boot past that screen with either DAVER2 EPROM or Slothie EPROM fitted so I'm back to something more fundamental. Colin. Quote:
|
||
18th Apr 2022, 9:58 pm | #72 |
Octode
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,297
|
Re: Perth PET
Hi Colin
How does the tynemouth board connect to the pet? Is it pin headers into the 6502 socket? Could you post a picture of the type of header pins on the board and the type of 6502 socket fitted on the pet. I’m thinking the header pins might have opened up the contacts on the socket, causing intermittent connections. Mark |
19th Apr 2022, 9:34 am | #73 |
Nonode
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,587
|
Re: Perth PET
Worth having a look at the 6502 socket as Mark suggests. Also are you satisfied that you’ve resolved the connectivity problems you’ve been having with UE10? Looking back at the epic thread many of the symptoms being exhibited now are very similar to those which were resolved by replacing UE10 back then. Alan
|
19th Apr 2022, 11:37 am | #74 |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,813
|
Re: Perth PET
The Tynemouth adapter has a turned pin socket for the 6502, which in turn has another turned pin socket to fit into the PET. See attached photos. I removed the 6502 socket many months ago during the mammoth thread and fitted a turned pin socket onto the motherboard.
I'll test continuity at the 6502 'top' - ie the pins next. Colin. |
19th Apr 2022, 8:52 pm | #75 |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,813
|
Re: Perth PET
Some Deoxit on UF10 has helped a little. With the Tynemouth board installed and replacing ROM and RAM set to Commodore Basic v4, I get a slightly corrupted but recognisable screen. A reset (there's a press switch on the board) clears the screen better.
But no keyboard. And Commodore Basic 1 and 2 just give me blank screens. The simple test on the Tynemouth board runs if I select that and says all is good... If I replace the Tynemouth board with a bare 6502, I get the screen on post 72, and neither the Slothie EPROM nor the DaveR2 EPROM will boot and start testing. I am now seriously confused. I have rechecked UF9 continuity and that's OK, but I have ordered a replacement chip just in case. I will go to work on UE10 and the 6502 continuity. I'm sure there's some logical reasoning for all of this, but I'm beggared if I can see it. Colin. |
19th Apr 2022, 10:16 pm | #76 | |
Octode
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,297
|
Re: Perth PET
Quote:
If the pins of a turned pin socket are plugged into the turned pin socket on the pet this will open up the contacts in the socket on the pet and might then cause a poor connection when plugging the 6502 in directly. It is possible to get male round pin headers for plugging into round pin sockets, but you need to make sure the thin end of the male header is plugged into the socket and not the thick end. Those male round pin headers are still slightly different shape cross section than an IC pin. |
|
19th Apr 2022, 10:20 pm | #77 |
Octode
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,297
|
Re: Perth PET
|
19th Apr 2022, 10:59 pm | #78 |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,813
|
Re: Perth PET
Apologies. Photos attached.
And the screenshot I was referring to was post 71. Colin. |
20th Apr 2022, 12:00 am | #79 |
Octode
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,297
|
Re: Perth PET
That was the type of male header pin I hoped they were using. I didn’t know you could get them in 40 pin DIL, I only have the 40 pin SIL strips. I’ll have to search for some of the DIL type now.
I still wonder if pressing the reset button is an improvement only because of physical pressure to the connections so definitely worth checking continuity, but take care that applying the probes doesn’t improve the connection through added pressure and give a false pass. Also inpect the turned pin sockets to make sure the inserts are not missing or damaged. There should be a small sping contact inside every turned pin socket, sometimes they get pulled out of the socket when the IC is unplugged. One other thing you could try is apply a little gentle side pressure on the 6502 in its socket while you press the reset button. |
20th Apr 2022, 2:16 pm | #80 |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,813
|
Re: Perth PET
OK - I've done as much continuity checking as I can including of UE10.
I'm going to need to go back to basics I think and start to track things just like before. Any and all help appreciated. I think I'm stuck. Colin. |