View Full Version : CBM 3032 Screen Problem!

November 30th, 2014, 12:31 PM

I have just got hold of a Commodore CBM 3032, but there is a problem.
The screen is completely green, I can see very weak text through the green screen and everything else seems to work. It's just the green screen that is the problem.

I have tried to remove the chip from the sockets on the inside to see if it helps in any way, but it did not.

Does anyone know what may have caused this problem with the screen?

December 1st, 2014, 09:36 AM
First of all (as this is your first post) welcome to the forum.

Secondly - your problem could be due to either the logic on the main board OR within the monitor circuitry.

Can I ask for some clarification as follows:

What do you mean by "everything else seems to work"? How have you tested it? When you power the CBM on - does it make a little 'chirp' or 'beep' noise? Can you read any of the text on the screen at all. It looks to me as though the video and the synchronisation has gone - plus the fact that the picture seems shrunk both vertically and horizontally (indicating a possible problem within the monitor) - although this is not 100% guaranteed as these signals are sourced from the main logic board).

Have you located the schematics for the machine on the internet and do you know how to read electrical/electronic schematics?

Do you have any test equipment (e.g. a multimeter, logic probe and/or an oscilloscope)?

Do you have any experience with high-voltage electronics? If not, please do NOT open the monitor box in an attempt to fault find. There are lethal high-voltages powering the Cathode Ray Tube (CRT) that can kill.


December 1st, 2014, 09:45 AM
The second photograph indicates a non uniform raster (those lines on the screen) is not stretching to the corners of the screen. The monitor's analog board will need attention if the power supply us outputting all the correct voltages. It doesn't hurt to replace any large lytic capacitors. They're all dried out by now.

If the text you are seeing is the same thing I am squinting and seeing is the same thing, your screen is full of garbage characters. A number of things can cause this to a bad ROM chip to a character generator but we can deal with that in better detail once the monitor is tended to.

It sounds like a lot of problems but don't worry, PET's these days never work right after sitting in storage for decades. They always need some sort of repair.

December 1st, 2014, 11:34 AM
I can see the BASIC text and the computer can run simple text programs.
It does not make a beep when I power it on and I always have to restart the computer once before I can can see anything on the screen.

I don't own a multimeter, logic probe or a oscilloscope but I can loan a multimeter if I need one,
and I don't have any experience with high-voltage electronics.

December 1st, 2014, 11:37 AM
The computer only have garbage-screen for a few seconds after startup, then it disappears and the computer runs BASIC like normal.

December 2nd, 2014, 10:58 AM
The garbage screen right after startup is normal behaviour. Its usually not seen because the screen takes a while to brighen after startup. If you turn it off and on again quickly, you will see the garbage screen appear and soon switch to the basic greeting when the boot is finished.


December 2nd, 2014, 11:34 AM
+1 for garbage characters after a power on.

I would ignore the lack of beep or anything like that for now.

If you can enter and RUN a simple BASIC program and 'sort of see' the correct characters on the screen that correspond to what you are doing - then I would start by assuming the monitor is at fault and have a look at solving that problem first. If you can enter and RUN a BASIC program then most of the main logic board is probably working OK.

I agree with NeXT that the first thing to do is to measure the power supply voltages within the monitor to ensure they are within specifications and then look at all of the electrolytic capacitors to see if any have started to leak their electrolyte or are starting to bulge - (or rather the other way round - visually inspect the components and replace any that look suspect first).

My previous warning still stands though - there are dangerous voltages here that can kill you - so if you are unsure about what you are doing I would consider seeking some expert help (or try to find an exchange monitor?)


December 2nd, 2014, 01:56 PM
I can see the BASIC text and the computer can run simple text programs.
It does not make a beep when I power it on

The 3032 PET does not have piezo transducer for sound; so it will not give a beep. Turn down the brightness in back until you can not see the retrace lines. Does the garbage screen on turn-on have both inverse and regular characters as is normal, or only inverse?

December 2nd, 2014, 05:02 PM
When my 2001N was new to me, it only had inverse characters on the garbage screen. There were a multitude of problems on the logic board which I fixed. Now, it never has inverse characters on the garbage screen. Yet, inverse/normal works just fine the rest of the time. Figure that one out.

December 2nd, 2014, 06:19 PM
When my 2001N was new to me, it only had inverse characters on the garbage screen. There were a multitude of problems on the logic board which I fixed. Now, it never has inverse characters on the garbage screen. .

On power-up, the video RAM comes up scrambled such that bit 7 is set in some of the 1000 locations. They will be the inverse characters. If all characters were in inverse, I would suspect a possible problem with Latched Screen bit 7 (LSD7).

December 2nd, 2014, 06:32 PM
On power-up, the video RAM comes up scrambled such that bit 7 is set in some of the 1000 locations. They will be the inverse characters. If all characters were in inverse, I would suspect a possible problem with Latched Screen bit 7 (LSD7).

On mine, bit 7 is never set in any video RAM location on startup. That's what's weird. I can set and reset bit 7 anywhere and it works fine.

Prior to my working on it though, it had bad video RAM, a bad gate for inverse video, and bad glue logic for the character generator. Now, it works the way it's supposed to in every way, except that there's never any inverse characters in the 'garbage screen', not that it matters, it's just weird.

December 30th, 2014, 12:25 PM
Now I've got a multimeter so I can start measuring the voltage from the power-supply to begin with. But i don't know between which pins I should measure.

Here is a picture on the board and one on the power-supply.

December 31st, 2014, 03:25 AM
First of all - have you obtained the schematics?

Have a look on http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/pet/ for something that fits the bill.

I would certainly start by measuring the unloaded voltage from the separate transformer windings (after disconnecting from the PET main board and monitor) using the AC voltage setting on the meter - then plug the transformer into the main board and check the voltages on the output of the voltage regulators on the main board (DC voltage setting on the meter). But personally - I think that if the PET seems to work 'normally' (i.e. it actually powers up into BASIC) then you will probably find these to be OK.

A meter is not the correct tool to look at the video and synchronisation signals from the main board to the monitor - you really need an oscilloscope for these.

I suspect the problem to be in the monitor - and I would not go poking around there unless you knew what you were doing. There are high voltages here that can (and will) kill!


December 31st, 2014, 04:15 AM
I have the schematics but I can't understand much of it. I don't want to give up but everything seems too complicated for me, and I don't know anyone that have any experience with CRT monitors. Maybe I can buy a composite video adapter and use a small TV instead of using the monitor.

January 1st, 2015, 11:45 AM
First of all - what is the part number on the transformer - I can't quite read it from the photograph as the wires are in the way.

Most of the wires from the transformer go to the main logic board. They should enter the main logic board on a connector marked J8. Is this correct? My guess would be the two whites, the two blues and the black wire go to J8 (in addition to some red and black wires from the large capacitor).

You will also see two 'skinny brown' wires going from the transformer to the monitor.

Pins 1, 3 and 5 of J8 contain one set of voltages and pins 8 and 9 of J8 contain the other set of voltages.

Follow the wires back from pins 1, 3, 5, 8 and 9 of J8 (noting the colours) to the transformer. The coloured wires should terminate in numbered tags on the transformer. For example, the 'skinny' brown wires to the monitor are terminated on tags 7 and 8 of the transformer (as seen in your photograph).

Disconnect the monitor and J8 from the logic board. Set your multimeter to AC VOLTS (20 ish) and measure the voltages of the following:

Between transformer tags 7 and 8 (brown wires to the monitor) - this should be about 15V to 16V A.C. for a 9" monitor.

The voltage between pins 1 and 3 of J8 should be about 8V A.C.
The voltage between pins 3 and 5 of J8 should be about 8V A.C.
The voltage between pins 1 and 5 of J8 should be about 16V A.C.
The voltage between pins 8 and 9 of J8 should be about 15V A.C.

(Note that it will be much easier to measure the voltages on the tags of the transformer - but for this you will have to identify which pins of J8 go to which transformer tags).

I have an 8032 - but the transformer tag numbers and wire colours appear to be different to your photograph.

You should be able to find the J8 pin-outs on the power supply schematic for your PET and(if you are lucky) the corresponding transformer tag numbers and wire colours.

Reconnect the PET logic board and monitor.

Things should work the same as they did before.

If it was me - I would measure the voltage coming from the 12V regulator of the monitor next. The monitor has a 7812 D.C. voltage regulator IC. You should be able to find this on the schematic and on the monitor PCB layout. I would attach a couple of wires to the PCB with a soldering iron and bring out the 0V (GND) and the +12V regulated supply line (possibly across capacitor C3 (a 47 microfarad 16V capacitor). Your meter should measure 12V. If it is low - then we need to look at why.

You need to be VERY CAREFUL around the monitor. There is 10kV (that is 10,000V) here and it will kill you if you are not careful. After you have measured the +12V voltage rail - I would leave things overnight to discharge before touching the monitor circuit again just to be on the safe side.

If the 12V rail appears to be OK - I would look next to measure the -30V (or -45V) D.C. voltage across the brightness potentiometer (depending upon which version of the monitor you have). Which reminds me - can you point me at the schematics on the internet that match your specific model of PET (just the power supply and monitor should do) - so that we both have a consistent set to refer to.

However - before trying ANYTHING - make sure that you understand thoroughly what I am saying and try to follow it on the schematics. I can't take any responsibility for any additional damage that may occur to your PET or injury to yourself. Also - before attempting any sort of measurements - just have a visual inspection of the components and soldering - trying to look for components that seem 'burnt' or 'stressed' or soldering that does not look right.

Let's take things slowly and make sure you understand the individual steps.

Just a thought - do you know any TV repair men/women in your area? I know they are a dying breed - but if it is your monitor that is at fault; they should be able to fix it for you.


January 2nd, 2015, 04:20 AM
Thanks for helping me!

The part number on the transformer is 3201 89-30

The voltage between transformer tags 7 and 8 is 17,1V A.C.
pins 1 and 3 of J8 is 8,5V A.C.
pins 3 and 5 of J8 is 8,5V A.C.
pins 1 and 5 of J8 is 17,7V A.C.
pins 8 and 9 of J8 is 17,4V A.C.

Here is all the the schematics I use and have printed out ftp://ftp.zimmers.net/pub/cbm/schematics/computers/pet/2001N/

I found the VR5 voltage regulator and the C3 capacitor but I don't understand the things you wrote about attaching wires to the PCB and bringing out the ground and +12V.

I don't like to open the monitor but heres some picures from the inside.

I'm going to contact a old TV repair man, one of my friends knew a man who has been working with TVs. And it takes some time for me to reply because everything I write has to be approved before it becomes visible.

January 2nd, 2015, 04:26 AM
I have forgotten to say that the computer was working fine when I started it up the first time, It had been in the school basement since the 80s and when I started it up it was working, but about three hours after that when I tried again the screen was like it is now.

January 2nd, 2015, 09:40 AM
Either the computers don't work when you first turn them on after 25-30 years of hibernation - or they do come on and last for a short while before failing. Yours appears to be the latter.

Be warned - even after you get it fixed - it is likely that something else will go duff and so on until all of the marginal components have been replaced. If I were going to keep it for myself - I would be looking to replace all of the tantalum and electrolytic capacitors and give the boards a good clean to try and minimise issues in the future. Tantalum and electrolytic capacitors don't age too well...

The unloaded transformer voltages look about right.

However, the 7812 regulator you found (VR5) is actually the +12V regulator on the main logic board not on the monitor. I was referring to the 7812 regulator on the monitor (which I think is identified as IC901 in the schematics you pointed me to). Also, the capacitor I was referring to should have been C902 according to the schematics (I was looking at a different schematic - doooo).

As this regulator/capacitor is on the MONITOR board - I wouldn't be 'poking' around there with a multimeter when the thing was switched on (hence my recommendation to solder a couple of wires onto the PCB as a means of extending the measuring points outside of the monitor case to make the measurement in relative safety).

I would say (however) that the option of contacting a TV repair man is very sensible given that monitors contain dangerous voltages. If he can get the monitor working for you - then fault finding on the logic board is much safer!

A few more posts and you will be 'off probation' so keep going... It is good to see that you are determined to get this PET going again though!

I see you are in Sweden. I used to visit Nykoping frequently some years ago with my job. A very enjoyable time.