Hello,

I have a faulty CBM 3032 (garbage screen) I'm trying to fix since 2001 with no result... I've changed *every* chip on the board: ROM/RAM/CPU/VIA/other damn'd without luck. The problem persists so I think maybe I added another problem instead to fix the old one...

Since I have a spare 8032 without keyboard, I was wondering if there is a way to change the 8032 board in some way to make it work on the 40 columns default OR at least to make some changes on the 3032 monitor to accept an 80 cols driver (I'm suspecting this is quite impossible).

The 8032 board is dated 1980 and marked as ASSY 8032030 - 8032 80 columns

I know that some boards suited both 40xx and 80xx series so maybe it's possible to change some jumpers, ROM chip, etc. to make it to work with the old 9" CRT from the CBM 3032...

any idea?

thank you!
cheers,

P.S. of course if you have any experiences about the garbage screen problem on the 3032 model, please tell me! I heard some people had this problem and the suggest was always "check the ram" but all the ram chips I have were tested on a 8032 are ok...

IOW, take the 8032 board and dumb it down to produce 40-column video signals that the CRT assembly from the 3032 will accept? IOW take the CRT assembly from the 3032 and tweak its circuitry so it'll accept 80-column signals (which it wasn't designed for) from the 8032, right?

I hope one of our CBM fans speaks up, since I don't have that background myself. Do you know whether the horizontal scan frequencies are different between 80- and 40-column operation? They just might be the same, meaning that 2nd option may involve few or even no modifications.

As for repairing the 3032 board as-is, is there someone who could offer you some in-person, hands-on assistance?

cheers,
Jeff

_________________
In 1988 my 65C02 got six new registers and 44 new full-speed instructions!
https://laughtonelectronics.com/Arcana/ ... mmary.html

Hi Jeff, thank you for your answer!



exactly. I know that CBM did the 4032 and 8032 series using same or very similar boards. So, maybe, there's a way to change something on the 8032 board (a jumper, a chip, a logic) and downgrade it from 8032 to 4032...



...or to take the 8032 CRT assembly adapting it in some way to fit the smaller 9" CRT



honestly, I don't know. What can I tell you is that I tried to directly connect the 8032 board to the 9" monitor but the result is a messy screen, something like when you to connect a 1280x1024 VGA card to an old CRT monitor with a max resolution of 1024x768... you can still see something, but messy.



Not here in Brazil, where actually I live. Very few people know Commodore Computers (it seems that just some unofficially imported C=64 and Amiga reached the Brazil in the '80s and '90s) and honestly I don't think I can find here somebody with the right mix of skill and interest about it...

Ok, no surprise. The horizontal (?) scan oscillator in the CRT assembly isn't falling into sync with the signal it's being sent. It might be expecting a slightly different frequency or it might be expecting a drastically different frequency.

2 brief suggestions:
- if you haven't already, on the CRT assembly, locate the pot or variable inductor marked H or HSYNC or HOR -- whatever. Twisting the control all the way yields a slight change in frequency. See if altering it will drop the picture into place. Check for Vertical as well.
- on the 'net, see if you can find scan frequencies for the two CRT assemblies / video modes in question.

bye for now
Jeff

_________________
In 1988 my 65C02 got six new registers and 44 new full-speed instructions!
https://laughtonelectronics.com/Arcana/ ... mmary.html

Hi Jeff,

I took a look inside the 9" CRT assembly; it's a quite simple board with two setting: brigthess (pot), height (pot) and width (inductor). I can't see any HOR or HSYNC settings (while there is plenty of settings on the 12" assembly...) - Maybe you can check it looking at the 9" schematics and part layout (see below).

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

Here we have schematics for the 8032, 12" CRT:
http://www.zimmers.net/anonftp/pub/cbm/ ... index.html

Schematics: http://www.zimmers.net/anonftp/pub/cbm/ ... 321448.gif
Part layout: http://www.zimmers.net/anonftp/pub/cbm/ ... 321449.gif

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

and here, the schematics for the 3032, 9" CRT (with oscilloscope wave forms, etc.):
http://www.zimmers.net/anonftp/pub/cbm/ ... index.html

This is the right schematics: http://www.zimmers.net/anonftp/pub/cbm/ ... 321445.gif
and the right part layout: http://www.zimmers.net/anonftp/pub/cbm/ ... 321446.gif

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

I somebody could confirm that I can try to connect the 12" crt assembly on the 9" tube without to blow anything, I could at least do this shot-in-the-dark and see if it works... but I'm afraid to destroy the assembly or the 9" tube...

thank you again

Swapping CRT's is kinda iffy, but might work. If you take the daring, risky approach you can find out quite quickly (!). But the slow and cautious approach takes a lot longer, since there are a lot of details to check in advance.

Is there a plug that allows you to disconnect the "yoke" -- the electromagnetic coil assembly on the neck -- from the circuit board? The CRT and the yoke usually cannot be separated, so the yoke & CRT both need to disconnect. To get an acceptable image you'd probably have to monkey with the pots etc that control the vertical & horizontal size of the image. Another liability is the focus control (probably located on the "the original crt transformer or whatever it is"). If any of these controls has insufficient range to match the different CRT then you are going to be slightly or perhaps entirely disappointed. I'll have a look at those schematics, but I don't expect we'll get the level of reassurance you're hoping for.

Do you have access to an oscilloscope? A logic probe? What if we were to have another look at repairing the 3032?

-- Jeff

_________________
In 1988 my 65C02 got six new registers and 44 new full-speed instructions!
https://laughtonelectronics.com/Arcana/ ... mmary.html

...of course to repair the 3032 is the best (and my preferred) choice.

Swap the boards and the CRTs was just an alternative way to make the 3032 to work using what I have at home.
Honestly, my desire is to fix the 3032 board!

That's the situation: I have a 3032 board where every IC is socketed. I don't mean just CPU, ROM, etc.: I mean *every* chip, also the 74xxxx ICs and the 555 IC

Since every other approach failed, I decided to check every IC on the board.... to discover that it didn't work.


aaah, I almost forgot this (maybe not helpful) detail: I can extract almost all main ICs (ram, rom, cpu, via, pia) from the board without to see any difference in the garbage screen (except for the char ROM and the two 2114 video ram).
BUT with one of the spare 6502 I have here, I got a strange behaviour: the garbage appears for a couple of seconds, then disappear and the screen become black. With the other 6502 the garbage remains. I found it quite strange, because if you extract the CPU the garbage still remains. So there is something with this CPU that changes the rules...

This 6502 IC come from the spare 8032 and it seems to work, but I can't be 100% sure because I didn't use this spare 8032 enough to affirm it's 100% ok; it could have some intermittent behaviour, I don't know.

Giovi

That might very well be the correct initial behavior. Incomplete, perhaps (due to some additional problem elsewhere), but correct. As I said, I don't have any CBM background. Anyway, you have two CPUs that behave differently, so it's possible (although not proven) that one of them is defective. Are they from the same manufacturer? If I were in your shoes I'd stick with the one that clears the screen.
Can you do the same test in reverse, using the 8032 board to prove the alternate CPU is bad?

-- Jeff

_________________
In 1988 my 65C02 got six new registers and 44 new full-speed instructions!
https://laughtonelectronics.com/Arcana/ ... mmary.html

I have 5 or 6 cpu from CMOS, but just one of them has this strange behaviour; and this cpu is the one originally mounted on the 8032 working board... board that works fine with all of the CPUs I have. Booooohhhh?

All 6 chips are presently known to work in the 8032? And, in the 3032, 5 chips are presently known to leave garbage on-screen, while the reamaining chip clears the screen to black -- is that right?

Are the chips superficially similar, identical, or what? And "CMOS"? I don't recognize that as the name of a manufacturer.

-- Jeff

_________________
In 1988 my 65C02 got six new registers and 44 new full-speed instructions!
https://laughtonelectronics.com/Arcana/ ... mmary.html

Yes, it's correct. There are some little differences (printed text, plastic housing, etc.), showing they come from different lots/ages, but they are all from the same manufacturer: that is MOS and not CMOS, sorry. Since MOS was owned by Commodore, we can expect there aren't compatibility issues among them.

I'm not even sure we need to focus on the cpu behaviour, since the garbage screen issue is present even if you leave the board without CPU at start...

And about fixing the board, I think you're totally right: only now I looked at both keyboards (the 8032-sk is in another place so I didn't look at it before) and I discovered they are quite different; 3032 has a "vic-20" style keyboard (with graphic signs below every key), while 8032 is a "pure" keyboard; and the keys layout is different too. So what I have here isn't just a "simple" resolution issue: it seems the only way will be the logical way: try to repair the 3032 board.

Lots & age probably don't matter, but are they the same part number?

Even in a 100% functional machine, garbage is the normal powerup condition. All being well, in the moments after powerup there will be a sequence of events that replaces the garbage with something meaningful. That would involve several steps, the first probably being to wipe the screen to all black. Later some text would get written there. That's why I think the chip with the unique behavior (clearing the screen) is actually working better than the others.

I have a strong hunch that you need to start over again and re-verify your conclusions about which chips work in which boards. You said you've been working on that repair for years. Honestly, the best way I can help right now is to doubt the (years-old?) information you're giving me. Please take the unique CPU chip and all (or at least 2 or 3) of the others. Label them so they can be identified. Test them all in the 3032, writing down the results. Then, similarly, test them all in the 8032 and report back.

-- Jeff

_________________
In 1988 my 65C02 got six new registers and 44 new full-speed instructions!
https://laughtonelectronics.com/Arcana/ ... mmary.html

Yes, same part number: 6502

I did the test you asked me:
- all the 6502 work in the 8032
- all, except one, produce a garbage screen in the 3032. The different one produces a garbage screen for about 1-2 secs., then the screen become suddenly black. If I switch off and then on again rapidly, sometimes only few chars on a black background appear.

Since this is the only difference I've seen in this faulty board, I think tomorrow I will change the 6502 socket: that's one of the few original sockets and it could be rusted. Probably it won't do the difference, but I think it worth a try...

Giovi

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

Ditto on the cap's, although back then they were probably for a linear power supply, so they may be OK? If it was an efficient switching type of today's standards (and cheap Cap's), I would highly suspect the capacitors in that case...
Nonetheless, I would concentrate on quality control of the power supply output first, especially for the digital electronics.

_________________
65Org16:https://github.com/ElEctric-EyE/verilog-6502