Commodore PET 8032 Monitor problem(s)?

[DistantStar001]

So I've been on a bit of a repair kick lately, and am now turning my attention to one of my more stubborn machines. It's a later Commodore PET, model CBM 8032. When I got it, its power cord had been chewed through near the base of the machine, and the stem on the CRT had been broken off, just behind the connector to the neck.

Since then, I've shortened the cord past the area where it had been chewed. And have been able to determine that the computer is working. It gives a boot chime, and an error chime if I max out the characters in BASIC. The old CRT had the analog board spitting sparks, so I've tried to replace it with one salvaged from an old Apple Monitor II that had some broken inside plastics. Nonetheless, the CRT does work, as I tested it with an Apple II prior to salvage. Unfortunately, I'm still not getting a picture. I've tried adjusting the controls on the analog board and got nothing. However, I am getting high voltage! and the heater is warming up. I checked the pins on the motherboard with an oscilloscope, and I am getting both vertical and horizontal sync signals. The screen also physically fits into the case, and the neck connector fits as well.

At this point, I'm guessing that the pinouts between the screen I had and the replacement are different in some way, but I haven't been able to locate a schematic of either the CRTs or the boards to confirm this.

The monitor I had was an Amperex M31-334GH. The replacement I have is a Samsung 12GBY31N. My question here is: Can I adapt the Samsung to work with my PET?
Also, could there be another culprit keeping me from getting a picture?

 

[Hugo Holden]

So I've been on a bit of a repair kick lately, and am now turning my attention to one of my more stubborn machines. It's a later Commodore PET, model CBM 8032. When I got it, its power cord had been chewed through near the base of the machine, and the stem on the CRT had been broken off, just behind the connector to the neck.

Since then, I've shortened the cord past the area where it had been chewed. And have been able to determine that the computer is working. It gives a boot chime, and an error chime if I max out the characters in BASIC. The old CRT had the analog board spitting sparks, so I've tried to replace it with one salvaged from an old Apple Monitor II that had some broken inside plastics. Nonetheless, the CRT does work, as I tested it with an Apple II prior to salvage. Unfortunately, I'm still not getting a picture. I've tried adjusting the controls on the analog board and got nothing. However, I am getting high voltage! and the heater is warming up. I checked the pins on the motherboard with an oscilloscope, and I am getting both vertical and horizontal sync signals. The screen also physically fits into the case, and the neck connector fits as well.

At this point, I'm guessing that the pinouts between the screen I had and the replacement are different in some way, but I haven't been able to locate a schematic of either the CRTs or the boards to confirm this.

The monitor I had was an Amperex M31-334GH. The replacement I have is a Samsung 12GBY31N. My question here is: Can I adapt the Samsung to work with my PET?
Also, could there be another culprit keeping me from getting a picture?

It almost certain the 12GBY31N will replace the M31-334GH and work with the PET VDU board, but the pin out at the base maybe different for the cathode, g1 A1 and the focus electrode. Obviously the heater is correct.

A while back on a thread I wrote up how to identify these by looking into the neck but I can't remember the thread title.

Otherwise we could easily do it in a few seconds, to check and adapt it, if we had the schematics of both VDU's.

Possibly there is another problem but we would not know until we sorted out the CRT connections.

 

[DistantStar001]

I went looking for the schematics and found these.

I found two schematics for the 12-inch PET:

http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/pet/8032/8032034.gif

http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/pet/8032/321448.gif

And this one for Apple Monitor II:

Apple Schematics: Apple Monitor II Main Schematic 050-5020-A : Free Download, Borrow, and Streaming : Internet Archive

Apple Schematics: Apple Monitor II Main Schematic 050-5020-A

archive.org

Unfortunately, I'm not sure what I'm looking at, as I'm not that good at reading schematics. Also, the printing on the Apple schematic is a little hard to read. The Commodore one is very clear (text-wise) but lacks the wire color info present in the Apple.

Near as I can tell, Pins 2, 3, and 4 are identical. Pin 1 either does not (or might not) exist. And I'm still trying to make out pins 5, 6, and 7 on the Apple. If I recall correctly though, both screens have/had six pins and one blank spot for orientation. I've been reluctant to disconnect the neck as it's on there pretty tight and I really don't want to mess with it until I have to.

 

[Hugo Holden]

As far as I can tell from those two schematics, the tube bases look identical, though it is hard to tell because of the blurred image of the Apple CRT which is labelled as pin 6, or not, see attached.

But I think they are the same. So there is another issue, most likely.

What happens when you turn the brightness control to full, can you see a scanning raster ?

 

[DistantStar001]

As far as I can tell from those two schematics, the tube bases look identical, though it is hard to tell because of the blurred image of the Apple CRT which is labelled as pin 6, or not, see attached.

But I think they are the same. So there is another issue, most likely.

What happens when you turn the brightness control to full, can you see a scanning raster ?

Thanks! However, now I have a new problem. I've removed the connector from the neck hoping that I could just redirect pin 5 to join with 1, only to discover that there is no connection to pin 5. I really didn't want to salvage any other parts from the Apple monitor, as I was hoping to find another CRT for it. The plastics are the only thing broken in it. And a little bit of acitone will likely fix that.

The connector appears to be made up of two parts, so maybe I can get them appart and add pin 5? The only other options are to salvage the connector from the Apple, or find/fashon a replacement...

In any case, thanks again! For soon my PET shall live!

The folloing is for my own memory shuld I need to replace it (maybe it will help someone else down the line). Anyway, here are the wire colors for my PET:

1=Green
2=Yellow
3=Brown
4=Black
5=NO CONTACT (but for this monitor should be Green)
6=Red
7=Blue

 

[Hugo Holden]

It looks to me from the diagram, albeit blurry, that pin 1 and pin 5 of your replacement CRT are connected inside the CRT (check with the meter on the CRT's pins) So it does not matter then that the socket only has the wire connected to pin 1, it does not need 5 connected in this case.

 

[DistantStar001]

It looks to me from the diagram, albeit blurry, that pin 1 and pin 5 of your replacement CRT are connected inside the CRT (check with the meter on the CRT's pins) So it does not matter then that the socket only has the wire connected to pin 1, it does not need 5 connected in this case.

You're right... And I confess, I wish you weren't. That means that the CRT I have is compatible, and now I have no idea what the problem is, beyond the analog board. I guess now I get to venture into high-voltage waters, for which I have no experience...

 

[Hugo Holden]

You're right... And I confess, I wish you weren't. That means that the CRT I have is compatible, and now I have no idea what the problem is, beyond the analog board. I guess now I get to venture into high-voltage waters, for which I have no experience...

High voltage in itself is not a problem. You could walk across the carpet and charge your body up to a few thousand volts. But the current is low if you discharge it.

You might ask yourself how it was possible that Nikola Tesla sat inside a cage with hundreds of thousands of volts causing massive plasma arcs around him, yet he was not electrocuted.

It is the current that is more of a worry. Generally it takes about > 30mA via a Human to be any significant risk, this is why the ELCB's (earth leakage contact breakers) on your domestic power board trip off that that current. It really takes > 30 to 100mA via the two arms and the chest area to have any significant chance of upsetting the cardiac conduction cycle.

Due to the resistance of skin, it is hard to get > 30mA via the body, unless the applied voltage is over 70V, but it helps if you are wet and rub your skin down with a salty solution.

Therefore, in the CRT VDU, like 110V or 220V power, you need to be cautious if the voltage anywhere is above 70v. But, there is no need to be scared of voltages above this value, if you understand them. The EHT supply in a VDU has very low current sourcing capability.

There is an exception though, be very cautious of line voltages over a few kV, for example an 11kV or 66kV transmission line power, for line power distribution, that Linesmen have to deal with. It has a very low internal resistance. If you get anywhere near that and you get caught in that pathway, it will coagulate your blood vessels and kill you in short order, and blow pieces off your body, much as an electric bug zapper does to bugs, because, not only is the voltage high, but so is the current. So all respect to Linesmen who have to work with this, it is a risky job.

So what I am saying here, if you are dealing with any "high voltage supply" you need to consider its internal resistance and how much current it can supply. I would happily connect myself across a 2000v supply, if it could only source a few mA, because that is harmless and my body's resistance would cause the supply output to collapse.

I really like The Glen Campbell song: " I am a lineman for the County and I drive the main road, searching in the sun for another overload".

 

[Hugo Holden]

Further to the above, in a typical VDU the voltages on the secondary side of the horizontal output transformer for the crt electrodes range up to 400V and they are exposed on the pcb and often there is a neck pcb too in many sets. And some can source over 30mA. So it pays to be very cautious when the VDU is powered. These voltages drop to zero fairly quickly after turn off. However with the VDU running, you can test these voltages with the usual insulated meter probes and x10 scope probes . One thing that helps, is to buy a x100 1500 to 2kV rated scope probe, for working on VDU's there is less chance of over voltaging the input to a scope or the probe being damaged if the voltage goes over 400v. Also, obviously, one of the more dangerous areas in the VDU, like any line powered appliance, is where the line power enters and the line side of the power supply. In SMPS supplies, which some VDU's have, that gets rectified to a high DC voltage and that is quite hazardous, it pays to wait at least half an hour for the electrolytic caps there to discharge after you turn off the VDU.

Fortunately, the EHT voltage is in a very well insulated system with good cable and a rubber cap over the EHT electrode on the CRT, and in most cases for fixing VDU's you don't have to go under that anode cap. It is fairly easy to deduce if the EHT is present, often at turn on or off you can hear electrostatic like micro discharges. Also, when there is a raster scan visible on the CRT face, and it is the correct size, then by definition the EHT is about right because there is an inverse relation between the EHT value and the size of the scan. Specifically the scan is inversely proportional to the square root of the EHT voltage. Also, in most cases, if the H output transformer is generating the correct auxiliary voltages on its secondary, and the collector voltage flyback peak on the H output transistor (HOT) is about right in amplitude, then it is almost certain the EHT is also ok. For most monochrome transistor VDU's in the 9 to 12" size range, that flyback peak voltage on the HOT's collector is in the order of 100 to 250v. Watch out in color sets though as it can be as high as 1kV, so they must be tested with the x100 probe.

In other words it is not only yourself you have to be careful with working on the powered VDU, it is your test instruments, especially the scope.

 

[DistantStar001]

Okay. I've started to dig into my analog card. I began by rechecking the capacitors and ended up replacing six. Two had some mild leaking, including the large non-polarized electrolytic. four others were just wildly out of speck. Unfortunately, this didn't help as I still don't get a picture. I then checked the 'neon' bulbs as they were exhibiting some interesting behavior when this thing was spitting sparks but now don't seem to do anything. However, a Hivoltage test shows that they are fine, and when I looked into it, it appears that they only light up when there's a problem. So impasse.

On the off chance I was wrong about the motherboard, I managed to find an original 8565 chip from eBay of all places. I'm relatively certain it's both real and working as the scope readings on the video signals are downright perfect. The original part in the PET was a substitute component that had always been a bit off, but still within spec.

I'm Linking to the AtariAge post I did last year where I have the video of what this thing was doing when I first got it.

Got a sick PET (8032) that needs help!

I recently acquired a Commodore PET that appears to have been in an industrial environment, then left to rot in storage. Over all, it's pretty beat up, and there was something living in it. The power cord was chewed to the copper. But no ruin or droppings. So I lucked out on that. Just a bunch of...

forums.atariage.com

Maybe the video will help locate the problem.

 

[daver2]

The "spitzen und sparken" in that video look hideous!

Where are you now with this repair (i.e. what are the symptoms now)?

Dave

 

[Hugo Holden]

Don't forget the "large non-polarized electrolytic capacitor" cannot be replaced with any type of electrolytic capacitor, normal or bipolar, that you can buy. It was a special part with an ESR of typically less than 0.15 Ohms. It has to be replaced with a film capacitor. It is the horizontal yoke coil's coupling capacitor and also the "S" correction capacitor. If it does not have the electrical properties of a low loss (welded plate style) film or oil filled capacitor, it will 1) heat up and 2) the horizontal scan linearity will be degraded. This special part only very rarely fails.

If the VDU is not running, one of the most useful tests is a recording, via a x10 scope probe, scope on DC coupling, scope on 10v/cm (giving a 100v/cm scale), of the collector voltage of the horizontal output transistor. This gives a lot of insight into what is going on with the H scanning and EHT generation.

 

[DistantStar001]

The "spitzen und sparken" in that video look hideous!

Where are you now with this repair (i.e. what are the symptoms now)?

Dave

The sparks turned out to be a symptom of a cracked CRT. The stem was broken off just past the neck connector. I've since replaced the CRT. So now the only symptom is no picture. I'm still getting high voltage, the signals from the motherboard are good.

Don't forget the "large non-polarized electrolytic capacitor" cannot be replaced with any type of electrolytic capacitor, normal or bipolar, that you can buy. It was a special part with an ESR of typically less than 0.15 Ohms. It has to be replaced with a film capacitor. It is the horizontal yoke coil's coupling capacitor and also the "S" correction capacitor. If it does not have the electrical properties of a low loss (welded plate style) film or oil filled capacitor, it will 1) heat up and 2) the horizontal scan linearity will be degraded. This special part only very rarely fails.

Done.

If the VDU is not running, one of the most useful tests is a recording, via a x10 scope probe, scope on DC coupling, scope on 10v/cm (giving a 100v/cm scale), of the collector voltage of the horizontal output transistor. This gives a lot of insight into what is going on with the H scanning and EHT generation.

Thanks! Will do! As soon as I figure out how...

 

[daver2]

>>> Thanks! Will do! As soon as I figure out how...

I would solder a wire onto the desired measuring point and 0V and bring them out to a "connector block". You can then reassemble the monitor and operate it safely.

You can then measure the signal on the connector block with your x10 oscilloscope probe safely.

Dave

 

[Hugo Holden]

In the CRT gun, the final anode connection on the gun, which is connected to the internal conductive coating and EHT connection, is only some millimeters from the focus electrode an A1 anode. Under vacuum, there is no worries and the final anode voltage in the region of 10kV to 20kV is no problem.

Generally in air, an arc will form between two conductors at around 1000v/mm of gap. So it is inevitable that the gun will arc over if there is air in the CRT, generally the EHT is in the 10 to 15kV region for these sorts of VDU's.

When the arc (spark plasma) is present, it acts fairly similarly to a Zener diode with a very low internal resistance which acts as a negative resistance, and a fixed voltage drop of around 600v for a spark in air. So it effectively shorts out the CRT's bulb capacitance, discharging it and the EHT output from the transformer. Also, it is not a continuous spark, because when the capacitance discharges, the spark extinguishes and it takes a while for the transformer and EHT rectifier current to recharge it, because this system has a fairly high internal resistance. So the spark generating system forms a "relaxation oscillator" with a characteristic sound of the spark, coming and going at some frequency.

One thing we have seen recently on another VDU repair thread is that if a set with a cracked CRT (let down to air) is run for a while, it does put quite a lot of stress on the horizontal output transformer. There was a recent case where it developed shorted turns as a result, but that is normally quite rare otherwise. If the auxiliary voltage produced by the horizontal output transformer's secondaries are normal, likely the output transformer and horizontal output transistors are fine.

For a CRT to have beam current and light up the face, it requires 1) the heater is running 2) the relative grid to cathode voltage is not too negative, generally -30 to -40v grid to cathode extinguishes the beam. At zero relative volts the beam current is at max. 3) the A1 anode has close to the correct positive voltage and the EHT is present. The focus electrode is not too critical but it can cause issues if it is floating/open circuit.

 

[DistantStar001]

Still not sure if I'm doing this right (actually I'm almost certain I'm doing it wrong as I'm still trying to figure out how to use an oscilloscope), but on my Scope, G1 is being consistently held low, G2 is consistently being held high, and G4 gives a signal that looks like a sign wave with a ripple on the upper curve.

Assuming that G1 is the cathode voltage, then it is way too low. Somewhere in the -120v range according to my multimeter. G2 (assuming it's the A1 anode) is more than 200v positive. Both these measurements were taken with my meter set at 600V for DC voltage, as I wasn't getting a reading at lower ranges.

Somehow, I'm thinking this is very wrong. Either in my readings or my measurements. On the multimeter, I'm a bit out of my depth as I've rarely needed to measure beyond 14v. As for my scope... I'm still getting the hang of it. I bought it used years ago, but it wasn't working at the time. A friend of mine managed to get it going a few months ago, but I'm still trying to figure out how to use it.

Even so, faulty or not, I think I can form a working hypothesis around these readings. Whatever's regulating the voltages for G1 and G2 aren't doing their jobs. If anyone wants to tell me what bits those are, it would be appreciated. Otherwise I'm going to have to follow the traces and start removing components to test them.

 

[daver2]

>>> Assuming that G1 is the cathode voltage.

This is the first incorrect assumption I am afraid. G1 is the first Grid electrode.

Can we stop please until we sort out exactly what we are doing...

First of all, which schematic relates to your actual monitor electronics PCB?

Let's get a baseline on what you have first...

Dave

 

[daver2]

On the assumption that this is your schematic: http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/pet/2001N/321445.gif...

Black and brown are the CRT heater connections.

Yellow is the cathode (k). This voltage should be between 0V (full white) and +40V (full black). I think!

Green is Grid 1. The DC voltage here should range from 0V to -45V (dependent upon the brightness control setting).

Red is Grid 2. This DC voltage should be +93V.

Blue is Grid 4 (focus). This DC voltage should either be 0V or +93V (i.e. the same as grid 2). The focus control is 'binary'...

Grid 3 does not exist as an external CRT connection.

But, before we go poking about, let us make sure that this IS the correct schematic. The easiest way to do this is to check the PCB parts layout http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/pet/2001N/321446.gif with your actual PCB.

Dave

 

[DistantStar001]

>>> Assuming that G1 is the cathode voltage.

This is the first incorrect assumption I am afraid. G1 is the first Grid electrode.

Can we stop please until we sort out exactly what we are doing...

First of all, which schematic relates to your actual monitor electronics PCB?

Let's get a baseline on what you have first...

Dave

Thank you for this! If you can't tell... I don't actually know what I'm doing. I mean, I can figure out an old computer. I'm good with general low-voltage stuff. If this was only an issue with the motherboard, it'd be working already. But CRTs are new territory for me.

But, before we go poking about, let us make sure that this IS the correct schematic. The easiest way to do this is to check the PCB parts layout http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/pet/2001N/321446.gif with your actual PCB.

No. This was not my board. This is my board:

Sorry, probably should have posted this sooner. Also, there is a number that can't be seen in the picture: 2817791Z117M

 

[powerlot]

I've got the same analog board in my pet I believe... Issues I've had with it where dirty brightness pot and cracked solder joints on the main connector (upper right in your picture)
This should be the correct schematic http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/pet/8032/321448.gif