Looking for help with a Pet 2001 dead video problem

[rglenn]

I recently acquired a 2001 PET with chicklet keyboard. When I turn in on I hear something happening briefly but the monitor never shows anything. When I turn it off I see nothing on the monitor as well - not even the dot in the center of the screen.
I removed the cover from the back of the monitor and turned it on ..... Nothing glowing.

Not sure where to start and could use some suggestions - thank you.

 

[daver2]

The first thing is to check the power supply to the monitor with a multimeter.

You should see two, thin wires (they may be brown in colour) going from the transformer to the monitor. These two wires carry the raw AC supply voltage to the monitor. Set your multimeter to read AC Volts (20 to 30 Volts at maximum) and measure the voltage between the two wires.

Post the measured voltage.

Next, set your multimeter to measure DC Volts (similar range as before). There is a voltage regulator on the monitor. The next thing to do is to read the DC voltage between the output of the regulator and the ground pin of the regulator.

Post the measured voltage.

Have you found the schematics and PCB layout for your monitor over at zimmers.net?

Dave

 

[rglenn]

Thank you very much for the help. This is the first CRT I've worked on.

The 2 brown wire connections measured from 16.0 - 16.4 volts
The voltage regulator measured from 3.75 - 3.76 volts

I attached a pic of the VR I measured

 

[daver2]

Excellent.

So your mains incoming supply and transformer are OK to start with if you have a good healthy 16 Volt AC supply to the monitor.

With the voltage regulator being a 7812 - you should have +12V DC on the output, not the low voltage you are reading.

The 7812 pin-out is as follows:



The problem could be the four rectifier diodes (CR19 - CR22) used to generate a raw DC supply from the AC transformer input, or the initial smoothing capacitor C1.

Use your multimeter (set to 20 V DC) to measure the voltage either across capacitor C1 or between pins 1 and 2 of the 7812 voltage regulator.

Post the reading.

If this is OK (higher than 14V DC) then the problem is either the voltage regulator itself, or there is a short circuit on the monitor somewhere causing the voltage regulator to shutdown and protect itself (and everything else) from destruction.

Let's cross that bridge when we are happy that we have a healthy DC input to the voltage regulator.

I am assuming your 3 (ish) Volt reading was taken between pins 2 and 3 of the voltage regulator?

Dave

 

[rglenn]

The voltage across capacitor C1 and between pins 1 and 2 of the 7812 voltage regulator both = 21v
The voltage across pins 2 and 3 of the 7812 voltage regulator = 3.7v

 

[daver2]

Excellent.

So things look good up to the voltage regulator but everything is sad afterwards...

So, we can go in one of three ways now.

1. We could remove the voltage regulator and replace it regardless.
2. We could remove the voltage regulator and test it out of circuit.
3. We could attempt to find a problem on the monitor board.

For option 3:

With the PET power switched OFF (and disconnected from the mains) - and left for a while to allow various capacitors to decay - use your multimeter (set to a fairly low resistance range (say 1kOhm max)) and measure the resistance between pins 2 and 3 of the voltage regulator. It is possible that the resistance starts out fairly low and then increases over time due to the action of the smoothing capacitors. After a while, the resistance should 'max out'.

What does it read when you get a stable reading?

If you just get a very low value - this would indicate a 'hard' fault on the monitor board that we need to locate.

I would also suggest taking a reading with the multimeter probes one way around, and then reverse the multimeter probes and take another reading. Sometimes you get different readings depending upon the semiconductor devices within the system. Reversing the probes can help.

Dave

 

[rglenn]

I went ahead and ordered a replacement voltage regulator. When I pull the current one, what measurements would I look for when I test out of circuit?

This is what I ordered -

10 X ST L7812CV L7812 7812 Positive Voltage Regulator 12V 1A TO-220 IC - USA​

Thanks a million.

 

[daver2]

You should have asked before buying...

There are 'modern' parts and 'not modern' parts. The modern parts have thin metal tabs that are expected to be soldered to a PCB heatsink. What you want is a 7812 with a thick metal tab to make good contact with the heat sink. Don't forget to order some heatsink compound...

To test the voltage regulator offline you just need a 16V (or so) power supply. Do you have a bench power supply?

If not, you can use a couple of 9V batteries. Connect the + (positive) lead to pin 1 of the voltage regulator and the - (negative) lead to pin 2 of the voltage regulator. Measure the DC output voltage using your multimeter. It should now be +12V.

This tests the voltage regulator off load - but it does indicate that it is basically working.

If you have a bench power supply - and a 12V lamp - you could apply a bit of load to see if everything is still good. Obviously, you will need some form of external heat sink if you start to push the current flowing through the voltage regulator. Don't exceed the rated current of 1A though. 12V at 1A = 12W lamp maximum (give or take). If you try to exceed the current rating of the voltage regulator, or it gets too warm, it should shut down.

These vintage linear voltage regulators are fairly good at protecting themselves, but (like everything) they get old...

Dave

 

[rglenn]

Using the (2) 9v battery method, the output from the VR is 3.6 - 3.7v.

 

[daver2]

It's dead...

Dave

 

[rglenn]

I appreciate your help yet again. I have ordered heatsink compound. I'm looking for a VR with a thick tab.

 

[daver2]

No problem...

I tested out the voltage regulators on my Cromemco disk supplies before actually connecting them up to the disk drives. The regulators are on a separate PCB with heatsinks and a few other bits and pieces.

Interestingly, some of the voltage regulators were faulty, but short-circuit from the IN to the OUT terminal. It is very, very unusual for a linear voltage regulator to fail in this manner - but I had one. It is a good job I tested them all before connecting them up to anything and blowing up the floppy disk drive unit!

Whilst you are waiting, you can perform similar tests on the main logic board of your PET. What is the part number of the PCB (as there are multiple configurations of this PET)? It should be something like 32xxxx. What is the xxxx?

I think there are four (4) off +5V regulators on this board - all the inputs of which are derived from a common input voltage (from the same transformer and via a diode rectifier and smoothing capacitor. You need to measure the outputs of all four (4) voltage regulators.

By the way, what other test equipment do you have? Do you have (for example) a logic probe or oscilloscope?

Dave

 

[rglenn]

My PCB number is 320008
Yes, I have a logic probe and a small oscilloscope.
I'll check the 4 VRs on the main board.

 

[daver2]

An original machine with 6540 ROMs and 6550 RAMs...

Have a look for four (4) test points identified as TP1-1, TP1-2, TP1-3 and TP1-4. Measure these voltage test points with reference to 0V/GND.

I think capacitors C7, C8, C9 and C10 are directly on the outputs of the four (4) voltage regulators if these are more easy to measure.

Dave

 

[rglenn]

I replaced the VR and put it back together. Now when I turn it on, the screen looks good for about 5 seconds and then goes wonky until it is just a thick horizontal block.

 

[daver2]

We need to divide and conquer...

If you look at the VDRIVE and HDRIVE signals on the connector between the logic board and monitor (using your oscilloscope) to see if they are the same at cold as they are when warm/hot. I would disconnect the monitor for this test.

If the traces are the same, the problem is in the monitor. If the traces are different, the problem is on the logic board.

Dave

 

[dave_m]

Vertical Drive - skinny pluse with a frequency of 60 Hz and a period of 16.6 mS at J7 pin 3

Horizontal Drive - wide pulse with a frequency of 15.6 KHz and a period of 64 uS at J7 pin 5

As daver2 says, disconnect P7 cable from PET main board first. Cable goes to monitor board.

 

[rglenn]

Fair warning - my oscilloscope experience would be measured in minutes rather than hours so I may have the settings totally wrong. That said, I uploaded pics of the results.

 

[daver2]

They look fine. A bit of a weird X and Y configuration, but it indicates what I would expect to see.

So the most likely failure is within the monitor circuitry.

So do you know whether we still have problems with the monitor voltage regulator?

Dave

 

[rglenn]

Output measurement fluctuates from 12.26 --> 12.27