9" CRT vc 12" CRT

[giobbi]

Hi all,


I have a spare, working 8032sk without keyboad and a faulty 3032 I'm one step away to fix it use gasoline and a match, since I'm trying to fix its garbage screen problem since 2001 (yes, I mean 12 years!) with no luck...

I was trying to find a way to use the 8032 working board inside the 3032 case. Yes, I'm trying to make the first 3032 at 80 colums


As first test, I tried to connect the 8032 board to the 3032 9" monitor. The image appears but it's quite messy (like when you drive a CRT above its max resolution: green background instead of black, text shifted, some horz. bright lines, etc...). I imagine (but of course I'm not sure!) the problem could be the 9" crt driver board. I'm assuming (better: I'm hoping!) the crt itself can support the 8032 resolution and the problem is just the 9" crt driver board...

So here's the question:

CAN I USE THE 8032 CRT DRIVER BOARD (the one inside the monitor, directly connected to the 8032 crt) ON THE 3032 9" CRT?
I was thinking to extract the board from the 12" CRT and use it to replace the one inside the 9" CRT.

- Will it blow, or will it work?

- Will it drive the 9" CRT at the 80 columns resolution of the 8032 board?

- Should I use the original crt transformer or whatever it is (I mean the one attached on the side of the crt with a "sucker" connector)
OR
Should I simply extract the "sucker" connector from the 9" crt and connect the one originally attached to the 8032 12" crt?

every suggestion is welcome!

thank you!
Giovi

 

[MikeS]

...every suggestion is welcome!
Giovi

Well, my first suggestion would be to repair the 3032 and find or adapt a keyboard for the 8032SK.

Exactly what have you tried to repair the 3032 in those 12 years? Contrary to some peoples' expectations, they generally do not repair themselves if you leave them sitting long enough ;-) Whether it's the combined skill of the 'experts' here or just plain luck (mostly the latter I suspect) almost everyone who brought PET problems here for discussion ultimately ended up with a working unit.

As to what you plan to do, connect an 8032 to a 9" monitor: I don't know of anyone who's actually tried it, but I suspect that the answer is no, it won't work, at least not without some hardware and/or software changes. If you just replaced the CRT I would expect the high voltage to be too high and you might have trouble getting the display to focus properly. Also, the yoke might not match the CRT correctly.

I don't understand what you mean by "Should I simply extract the "sucker" connector from the 9" crt and connect the one originally attached to the 8032 12" crt?"; if you are talking about swapping the LOPT (flyback transformer) from the 12" board to the 9" board I would think that you're asking for trouble.

The simplest thing would be to just reprogram the CRTC in the 8032 to match the 9" monitor's characteristics, which might of course be incompatible with some software.

But it all sounds like a lot more work or inconvenience than just repairing the two systems...

 

[giobbi]

Well, my first suggestion would be to repair the 3032 and find or adapt a keyboard for the 8032SK.

Exactly what have you tried to repair the 3032 in those 12 years? Contrary to some peoples' expectations, they generally do not repair themselves if you leave them sitting long enough ;-) Whether it's the combined skill of the 'experts' here or just plain luck (mostly the latter I suspect) almost everyone who brought PET problems here for discussion ultimately ended up with a working unit.

Eh eh eh of course everybody at least once in a lifetime tend to believe in the "night-fixing spirit", or the "well, I will go to sleep and maybe tomorrow will be ok again" ;-)

BUT.... answering to your question, I can tell you I replaced *every* chip on the damn'd board with new one or working ones: RAM, video RAM, PIA, VIA, CPU, ROM (i bought an eprom programmer to test them), logic chips and some electrolitic capacitors too. The frustrating truth is: nothing changed. I have a more or less always-the-same garbage screen: a screen full of chars, both normal or inverted.

Trying to extract some key-chips like ROM, RAM or even the 6502 CPU or the PIA and VIA chips (I mean the two 6520 and 6522) it doesn't change.
The only way I've seen something different was when I extracted the char ROM (of course!): instead of a screen full of chars then I saw a black and green pattern (green where normally it shows inverted chars).

Of course I didn't work at it for 12 years, 8 hours per day, 365 days per year: simply I works on it for a while, then I get frustrated enough to say "this was the last time I spent on it, there's no way to fix it", I forget it for some weeks/months and then I come back on it.

It seems the garbage screen is quite a common problem with many causes: sometimes are RAM chips, sometimes ROM, sometimes other uncommon causes. And many times there was no way to fix it.

Since I already have another complete and working 8032-SK and it's quite hard (better: quite impossible) to find a spare keyboard for this model, I was trying to find another way to make the 3032 to work, at least in some way.

If you have any idea about what I can do to fix the 3032 board, I'm really interesting to know: sometimes you think you did all you can do but you're misconsidering something. Every suggestion, every idea will be really welcome.

As to what you plan to do, connect an 8032 to a 9" monitor: I don't know of anyone who's actually tried it, but I suspect that the answer is no, it won't work, at least not without some hardware and/or software changes. If you just replaced the CRT I would expect the high voltage to be too high and you might have trouble getting the display to focus properly. Also, the yoke might not match the CRT correctly.

Well, I received some minutes ago an email from Ruud Baltissen (www.Baltissen.org) where he told me:
"You already mentioned the idea I used when building C64 inside a brooken PET 2001 in 1989: I replaced the circuit of the original CRT with the one of a Philips video monitor. Had to adjust some resistors IIRC because of the smaller tube but it worked!"
That's giving me some hoping...

I don't understand what you mean by "Should I simply extract the "sucker" connector from the 9" crt and connect the one originally attached to the 8032 12" crt?"; if you are talking about swapping the LOPT (flyback transformer) from the 12" board to the 9" board I would think that you're asking for trouble.

The simplest thing would be to just reprogram the CRTC in the 8032 to match the 9" monitor's characteristics, which might of course be incompatible with some software.

But it all sounds like a lot more work or inconvenience than just repairing the two systems...

Yes, I was wondering if I can connect the LOPT (I didn't know the right term) of the 12" crt directly to the 9" crt, or if should to swap them or none of this two options.
And when you say "But it all sounds like a lot more work or inconvenience than just repairing the two systems...", well that's the sound of my desperation... ;-)
Of course it would be better and fine if I can fix the original 3032 card.... simply I'm quite frustrated and I'm really starting to think I will never find a way to fix it...

Well, thank you for your answer and your time!

 

[MikeS]

...Trying to extract some key-chips like ROM, RAM or even the 6502 CPU or the PIA and VIA chips (I mean the two 6520 and 6522) it doesn't change.

Well, yes, often just swapping chips does fix the problem, but sometimes it needs a little deeper investigation and a systematic troubleshooting approach. Do you have any test equipment? Can you burn EPROMs? It's sort of like trying to repair a car without any tools and saying you've tried everything...

The only way I've seen something different was when I extracted the char ROM (of course!): instead of a screen full of chars then I saw a black and green pattern (green where normally it shows inverted chars).

To be expected; removing either or both the video RAM chips should also change what's displayed.

It seems the garbage screen is quite a common problem with many causes: sometimes are RAM chips, sometimes ROM, sometimes other uncommon causes. And many times there was no way to fix it.

There is ALWAYS a way to fix it, although sometimes (rarely) it can be difficult to find the trouble. PETs are really quite simple and uncomplicated systems, as computers go. But yes, the garbage screen on a pre-CRTC PET just means that it is not running its normal program, which could be caused by just about anything; however, it usually comes down to only a few common problems.

Since I already have another complete and working 8032-SK and it's quite hard (better: quite impossible) to find a spare keyboard for this model, I was trying to find another way to make the 3032 to work, at least in some way.

There was just a discussion in another thread here about using a C64 keyboard on an 8296; the 8032 might be useful for some testing if you can (temporarily) adapt the 3032 keyboard or even another one:
http://www.vintage-computer.com/vcf...265-Pet-8296-w-o-keyboard&highlight=color+pet

If you have any idea about what I can do to fix the 3032 board, I'm really interesting to know: sometimes you think you did all you can do but you're misconsidering something. Every suggestion, every idea will be really welcome.

Again, do you have any tools (i.e. multimeter, oscilloscope, EPROM programmer, etc.?)

Simple things to check are the reset line (should be high) and looking at/measuring the address lines while running a NOP generator.

Well, I received some minutes ago an email from Ruud Baltissen (www.Baltissen.org) where he told me:
"You already mentioned the idea I used when building C64 inside a brooken PET 2001 in 1989: I replaced the circuit of the original CRT with the one of a Philips video monitor. Had to adjust some resistors IIRC because of the smaller tube but it worked!"
That's giving me some hoping...

Yes, I read your post on CBM-hackers and Ruud's reply, but I don't see how it's relevant; a C64's video is quite different from a PET, and although the 3032 can be persuaded to display on a 'normal' monitor, the 8032 uses a higher non-compatible frequency. But like I said, that can be adjusted in software so if that's good enough then that's the simplest way to use the 9" display with an 8032. Find the discussion threads on CBM-hackers about Steve G's 'Color PET'.

Yes, I was wondering if I can connect the LOPT (I didn't know the right term) of the 12" crt directly to the 9" crt, or if should to swap them or none of this two options.

I still don't get this; you want to remove the two flybacks from their respective boards and mount the 12" one on the 9" board?

And when you say "But it all sounds like a lot more work or inconvenience than just repairing the two systems...", well that's the sound of my desperation... ;-)
Of course it would be better and fine if I can fix the original 3032 card.... simply I'm quite frustrated and I'm really starting to think I will never find a way to fix it...

Where are you? If you don't have any equipment maybe someone nearby can help you with that.

 

[giobbi]

Well, yes, often just swapping chips does fix the problem, but sometimes it needs a little deeper investigation and a systematic troubleshooting approach. Do you have any test equipment? Can you burn EPROMs? It's sort of like trying to repair a car without any tools and saying you've tried everything...

My biggest trouble is my lack of knowledge about electronics. I just have some basic skill (i.e. I can work on a pcb without to destroy it...sometimes...) but I'm not an expert...
Yes, I have both EPROM burner and EPROM eraser, some 2732 and 2716 and an adapter 27xx to 25xx/23xx that *should* work (I always use the word *should* when I'm talking about something I personally did ;-) )
I have a multimeter but not an oscilloscope.

To be expected; removing either or both the video RAM chips should also change what's displayed.

Yes, of course you're right, I forgot to mention the two 2114; without them, I get a huge chessboard pattern; with just one of them I get some garbage pattern.
I tested them on a VIC-20 and should be ok.

There is ALWAYS a way to fix it, although sometimes (rarely) it can be difficult to find the trouble. PETs are really quite simple and uncomplicated systems, as computers go. But yes, the garbage screen on a pre-CRTC PET just means that it is not running its normal program, which could be caused by just about anything; however, it usually comes down to only a few common problems.

It sounds promising
Well, what I did until now was to extract every chip and substitute it with a new one or a working one. I'm presuming I did everything correct, but of course I could have done some mistakes, so maybe the right way to proceed is to think that we have a board with a problem, but every chip is on socket so we can easily replace them.

And I have some other spare ICs like cpu, pia, via, ram chip, most of the logic glue, etc. and of course every ICs that can be taken from the 8032 board.

One thing I didn't was to check the original ROMs integrity trying to read them with my eprom programmer. It needs a 23xx to 27xx adapter in order to make the eprom programmer to read them (assuming that it can be done). I think it would be fine to do this check.

About ROM: I tried to use the 2+1 basic ROMs from the 8032 (I found an article about a guy that's using basic 4.0 on his 3032 and I tried the same kind of upgrade, just to exclude a rom problem).

There was just a discussion in another thread here about using a C64 keyboard on an 8296; the 8032 might be useful for some testing if you can (temporarily) adapt the 3032 keyboard or even another one:
http://www.vintage-computer.com/vcf...265-Pet-8296-w-o-keyboard&highlight=color+pet

Again, do you have any tools (i.e. multimeter, oscilloscope, EPROM programmer, etc.?)

this will not be a problem, since:
1) I can simply swap the boards and use the 3032 keyboard, or
2) I can take the keyboard from one of my working 8032-sk and 8296-sk
...
as I told you belove, I haven't an oscilloscope, but I have the basic tools (soldering station, multimeter, eprom burner/eraser, etc.).

Simple things to check are the reset line (should be high) and looking at/measuring the address lines while running a NOP generator.

:-( sorry, can you explain better what should I do?

Yes, I read your post on CBM-hackers and Ruud's reply, but I don't see how it's relevant; a C64's video is quite different from a PET, and although the 3032 can be persuaded to display on a 'normal' monitor, the 8032 uses a higher non-compatible frequency. But like I said, that can be adjusted in software so if that's good enough then that's the simplest way to use the 9" display with an 8032. Find the discussion threads on CBM-hackers about Steve G's 'Color PET'.

Of course fix the 3032 is still my preferred choice: to find a way for make the 8032 board work with the 3032 crt was just an alternative way to make the 3032 to work...

I still don't get this; you want to remove the two flybacks from their respective boards and mount the 12" one on the 9" board?
Where are you? If you don't have any equipment maybe someone nearby can help you with that.

I'm italian but actually I'm living in Brazil. Believe me, nobody here can help me with this task :-/

thank you!

 

[MikeS]

The ROMs from an 8032 will not work in a 3032.

Since you have a working 8032 the first thing I would suggest is that you take the ROM chips from the 3032, carefully insert each one in one of the spare sockets in the 8032, and calculate the sum of the ROM contents; tell us the part number of the chips and their checksums and we can compare them to what they should be.

Note that 3 of the ROM chips will probably have similar numbers, e.g. 901465-01, 901465-02, 901465-03; those 3 are 4kB chips whereas the odd one (e.g. 901447-24) will only be 2kB, so for that one use 2047 instead of 4095 below:

Assuming the chips are in the 'A' socket (closest to the rest):
S=0:FOR I = 10*4096 TO I+4095:S=S+PEEK(I):NEXTRINT S

 

[giobbi]

The ROMs from an 8032 will not work in a 3032.

Since you have a working 8032 the first thing I would suggest is that you take the ROM chips from the 3032, carefully insert each one in one of the spare sockets in the 8032, and calculate the sum of the ROM contents; tell us the part number of the chips and their checksums and we can compare them to what they should be.

Note that 3 of the ROM chips will probably have similar numbers, e.g. 901465-01, 901465-02, 901465-03; those 3 are 4kB chips whereas the odd one (e.g. 901447-24) will only be 2kB, so for that one use 2047 instead of 4095 below:

Assuming the chips are in the 'A' socket (closest to the rest):
S=0:FOR I = 10*4096 TO I+4095:S=S+PEEK(I):NEXTRINT S

Of course I can do this, but I wish to know if I can make another test, instead: I have an EPROM programmer, so is there a way to make a pin adapter in order to read 23xx ROM as they were 27xx eprom? If I'm right, since my programmer only handle 27xx eprom, not 25xx ones that are pin-compatible with 23xx, I have to build an adapter; but I don't know if it will let me read the ROM content...

I wish to do this way because will be easier for me to check ROMs in future, simply comparing its content with the image file...


EDIT: I've done, here we have the results:
NO EPROM: 686080 >> Can I assume this will be the default value in case of totally burnt ROM ?
901465-01: 473144
901465-02: 479338
901465-03: 556184

since I wasn't sure, for the 2k ROM I run the program twice with 2 different parameters:
901447-24: 261567 >> using 2047 instead of 4095
901447-24: 348160 >> using 2047 instead of 4095 AND 2048 instead of 4096

901447-10: I didn't test it because there an equal ROM on the 8032 board; I already swapped them and they seem to work (but of course I can check it too if you think it would be better)

The ROMs from an 8032 will not work in a 3032.

I've seen it on a site (basic 4.0 on 3032 with eprom from the 8032 images) but actually I can't find the site again.

 

[MikeS]

Of course I can do this, but I wish to know if I can make another test, instead: I have an EPROM programmer, so is there a way to make a pin adapter in order to read 23xx ROM as they were 27xx eprom?

See:
http://www.vintage-computer.com/vcf...-for-EPROM-burning!/page8&p=173003#post173003

EDIT: I've done, here we have the results:

Unfortunately they all look OK; I assume you've double-checked that they're in the right sockets?
L-R, U9-U6: 03,24,02,01 IIRC

901447-10: I didn't test it because there an equal ROM on the 8032 board;

The character generator, obviously OK.

I've seen it on a site (basic 4.0 on 3032 with eprom from the 8032 images) but actually I can't find the site again.

I'll believe it when I see it. BASIC 4, no problem, but 8032 ROMs in a 3032??

Check the voltage on the CPU RESET pin. Other than that I'm afraid we might need at least a logic probe, preferably an oscilloscope. No local friends who might be able to help?

There is an image for a minimal diagnostic EPROM somewhere; that might tell us something.

Anybody else have any ideas?

 

[dave_m]

Anybody else have any ideas?

Mike,
I agree with you. If he has correctly checked all the chips, then he will need a scope as there may be a broken circuit trace or short circuit causing all the problems. Maybe we could start with the simple NOP Generator trick to check all the address lines and associated signal paths?
-Dave

 

[giobbi]

...yes, ICs are in the correct socket :-(

About the scope: do you think this cheap USB model can be useful for this kind of test (and future tests on this kind of boards)?

http://www.ebay.it/itm/NEW-2-Channe...15604?pt=BI_Oscilloscopes&hash=item1c282ce6d4

If not, what should be the lowest configuration I can accept (keeping in mind I will not use it so much, maybe never in the future)?

----------------------

Basic 4.0 on 3032: I took the idea from this article:
http://www.classic-computers.org.nz/blog/2011-07-04-pet-rom-upgrades.htm
In the picture #2 there are visible the two eprom with the 8032 code: 901465-22, -23, etc. so I assumed (maybe wrong) it could work.

Just a though: I changed every ICs on the board; what I didn't change yet are the old socket. The computer lied for years (maybe decades) in a garage. Do you think a stained socket can be the guilty? In other words, would you suggest to me to change those sockets with fresh ones?


-- Giovi

 

[dave_m]

About the scope: do you think this cheap USB model can be useful for this kind of test (and future tests on this kind of boards)?

http://www.ebay.it/itm/NEW-2-Channe...15604?pt=BI_Oscilloscopes&hash=item1c282ce6d4

If not, what should be the lowest configuration I can accept (keeping in mind I will not use it so much, maybe never in the future)?

Giovi,
No, this USB model is only good for 4 KHz signals. The master clock on the PET is 16 MHz. You would want a scope device with a bandwidth of at least 32 MHz and preferably 50 MHz or more. With the determination you have shown so far by changing all the ICs, you may be able to find the problem with a good multimeter and a logic probe. Are you familiar with these more affordable instruments?

Basic 4.0 on 3032: I took the idea from this article:
http://www.classic-computers.org.nz/blog/2011-07-04-pet-rom-upgrades.htm
In the picture #2 there are visible the two eprom with the 8032 code: 901465-22, -23, etc. so I assumed (maybe wrong) it could work.

No, the 8032 board is set up for an 80 column display which requires more bandwidth. You might change a lot of jumpers on the board and convert it to a 4032 40 column board. Then you could find an editor ROM which will work with your 9" screen. I think you should fix the 3032 with our help.

 

[dave_m]

I have a spare, working 8032sk without keyboad

Giovi,
I think you may be able to use the keyboard from a broken C64 and use it with a custom Editor EPROM designed for us by Steve Gray to work in your 8032sk.

Does anyone know if a C64 keyboard fits into an 8032sk chassis? I know it fits into a regular 8032.
-Dave

 

[MikeS]

...Basic 4.0 on 3032: I took the idea from this article:
http://www.classic-computers.org.nz/blog/2011-07-04-pet-rom-upgrades.htm
In the picture #2 there are visible the two eprom with the 8032 code: 901465-22, -23, etc. so I assumed (maybe wrong) it could work.

As I said, BASIC 4 will certainly work but the parts of the ROM firmware that deal with the different display size/type/frequency, keyboard layout, cassette ports etc. are different, e.g. 901474-02 instead of 91474-03 (I think).

Just a though: I changed every ICs on the board; what I didn't change yet are the old socket. The computer lied for years (maybe decades) in a garage. Do you think a stained socket can be the guilty? In other words, would you suggest to me to change those sockets with fresh ones?

Sockets and oxidation of the IC pins have certainly been a source of problems; most of the pins (address and data lines) of the normally socketed ICs are connected in parallel, so if you have a continuity tester you might want to check that all those IC pins are indeed connected together.

 

[giobbi]

you may be able to find the problem with a good multimeter and a logic probe. Are you familiar with these more affordable instruments?

multimeter, yes; logic probe, no (to be honest I don't know what is it) but I love to learn if somebody have time and patience to teach me

No, the 8032 board is set up for an 80 column display which requires more bandwidth. You might change a lot of jumpers on the board and convert it to a 4032 40 column board. Then you could find an editor ROM which will work with your 9" screen. I think you should fix the 3032 with our help.

Ok, I think this is the best way too, I really wish to fix it! So please forget the idea to mix the 3032 case/crt with the 8032 board...

I think you may be able to use the keyboard from a broken C64 and use it with a custom Editor EPROM designed for us by Steve Gray to work in your 8032sk.
Does anyone know if a C64 keyboard fits into an 8032sk chassis? I know it fits into a regular 8032.

Don't worry about the 8032-sk without keyboard, I have another, complete one (plus a 8296-sx), so I will keep it as "spare part", at least until I will find an original 8032-sk keyboard (or an empty 8032 case)

As I said, BASIC 4 will certainly work but the parts of the ROM firmware that deal with the different display size/type/frequency, keyboard layout, cassette ports etc. are different, e.g. 901474-02 instead of 91474-03 (I think).

ok, I understand.

Sockets and oxidation of the IC pins have certainly been a source of problems; most of the pins (address and data lines) of the normally socketed ICs are connected in parallel, so if you have a continuity tester you might want to check that all those IC pins are indeed connected together.

Fine, I didn't think at it! I will do a try and see if I will discover something...

Hey, people, thank you very much, you gave me some hope about this damned, not cooperating, old CBM

-- Giovi

 

[giobbi]

UPDATE:

I checked all of the existing sockets (I mean the original ones from Commodore that's actually hosting an IC); luckily a little part of the soldered trace was accessible by the multimeter probe, so I was able to check all of them. I found the 6502 socket was faulty. For I while I almost hoped in a miracle.... ...almost....

Well, changing the socket I did a little step foreward (or a giant leap backward...): now the garbage screen appears for about 1", then it turns to black (I have to switch the computer on, leave the crt warm a bit, then switch off and on again to see it). So this is the actual position:

- No CPU in the socket: garbage screen.
- CPU in the socket: garbage screen for about 1 sec. then black screen. I've checked two trusted, different 6502 IC to be sure.

It seems the CPU is working in some way now and the system is doing some kind of reset.

Do the news give you any idea?

 

[MikeS]

Definitely a step forward! We know that the CPU is now running, reading instructions from the ROMs and starting up normally, initializing and clearing the screen; we just have to find why it is not continuing to a complete startup. Try removing/exchanging the PIAs for a start.

If you didn't already, I'd still check continuity of the address and data lines across all the ROM chips (not the sockets).

 

[giobbi]

Thank you, good to hear that
I already tried without the two PIAs (you mean 6520, right?), it was the last test I did before to post the update here.
Tomorrow I will check continuity as you suggested, now I can't deal again with the board.... man, I need a *huge* beer urgently
Thank you for now, I will tell you a.s.a.p. the result of the continuity check.
cheers, have a good night!
Giovi

 

[giobbi]

ok I had my beer and a caipirinha (strawberry flavour) too, now I'm ready (?) to make some more tests :rockedov:

I checked continuity among roms, it seems to be ok, all pins are connected together (I mean: every pin 1 of every ROM is connected together with other pins 1, etc.): all but one.
This should be ok, I believe.

A clue: Not immediately at the power up, but after some minutes two dioes (CR1 and CR2) run quite hot, I measured 48°C. The other ones are cold. The same power connector is strangely hot...

The four ROMs: 901465-01/-02/-03 and 901447-24 (but not the char ROM - 901447-10) are hot too: you can't keep your finger on them more than few seconds.

I need a comfortable pillow and a warm bed now!
good night

 

[MikeS]

A clue: Not immediately at the power up, but after some minutes two dioes (CR1 and CR2) run quite hot, I measured 48°C. The other ones are cold. The same power connector is strangely hot...

That doesn't sound right; they should all be pretty well the same (CR1,2,4 & 5); check to make sure one of them isn't shorted. Might as well check the three voltages while you're at it, preferably at the IC pins.

The four ROMs: 901465-01/-02/-03 and 901447-24 (but not the char ROM - 901447-10) are hot too: you can't keep your finger on them more than few seconds.

Check the voltages.

I need a comfortable pillow and a warm bed now!
good night

Same here; good night.

Edit: Had a look at the schematic and it looks like CR1 and 2 do pass quite a bit more current than the rest of them, so they would be warmer. Check the voltages anyway, and also across the connector; they were a little underrated for the current they carried and were prone to oxidation and overheating.

 

[giobbi]

Good morning!

I checked the voltages. My multimeter has a bit of tollerance, but it should be basically ok for this kind of measurement.

Source supply voltages are +5.05v, -4.99v, +12.21v at output diodes (as shown in the Schematic 320349 page 9 of 9, voltage regulator 320349-9.gif from Zimmers).

- at ROMs pin 24 (Vcc): between 5.02v and 5.05v;
- At 6502 pin #8 and 6522 & 6520 pins #20: 5.05v;
- At RAM pin #1: -4.99v; at pin #9 (Vcc): 4.99v; at pin #8 (Vdd): 12.21v

I checked the two diodes extracting one pin, they're ok. I've re-soldered them, just to be sure.

ROM chips take some minutes to warm up, but I think they reach about 45-50°C, basing on my finger's inner thermometer

-- Giovi