• Please review our updated Terms and Rules here

Commodore Pet 3032 no power

@Hugo Holden
I found a 5W 39ohm. But I guess thats to high value. I'll have another 7805 on the way. So I have to wait a couple of day before it arrives.

I have been checking all the small 0.1uf caps from the 5v line and all seems fine.

Whould a shorted 74xxx cause the regulator to shut down?

I was looking at the partlist "the dynamic pet"
This list suggest that C15 is a 10uf cap. But I can see on schematics its a 1uf.
 
Yes, a shorted 74 series device would cause the same symptoms.

But is your machine a "dynamic PET" though?

It is unlikely to be a 0.1 uF decoupling capacitor. It is more likely to be an electrolytic or a tantalum bead capacitor in my opinion.

5V / 39 Ohms = 0.1 A (I have rounded the resistor to 50 Ohms to make my mental arithmetic easier). This is about one tenth of what is required.

Dave
 
@daver2
Ok ,thats good to know. This PET gonna live!
But its quite annoying that the regulator shuts down (if its not broken) becouse its makes it alot harder to find a faulty IC.

My bad Dave. I dident get it that it was a mix of diffrent Pets on the 2000N page.
 
It may be quite annoying, but it saves doing more damage!

If you remove the regulator from the PCB and temporarily wire up the input and ground pins, you can insert a series resistor inline with the regulator output pin and the PCB. This will allow the regulator to operate, and the resistor prevents excessive current flow to shutdown the regulator. You may stand a chance then of going around the board with your multimeter checking where the lowest voltage is. Thus will help home in on the faulty device.

You will have to do a bit of Ohm's law (V=I*R and P=I*I*R) to work out a suitable resistor for 0.1A and then what the power dissipation will be.

Don't forget that you can make a higher powered resistor network by paralleling up multiple resistors (of the same value).

Dave
 
@Hugo Holden
I'll have another 7805 on the way. So I have to wait a couple of day before it arrives.
One trick, when you are in the field of servicing for a while, and you need spare parts and you think you need one for a job, buy 2 or 3.

The same applies if you are buying spare IC's etc. It saves a fortune on postage and then if you don't use the part for that job, you can put it in a spare parts inventory and they are there to dig you out of a hole, later on another job and there is no waiting around to get parts.

I have been doing this for decades, so most cases now, if I need a part like any 74LS IC or a voltage regulator I tend to have it on hand. However, this sort of thing really only applies to vintage electronics now, like the PET and many other computers from that era (Altair, SOL, IBM,Apple etc etc), which all used pretty much the same IC's & regulators. I would not take this approach very often with modern throw away appliances or modern computers unless it was something I really valued.

Daver2 mentioned some of the equations for calculating power in a resistor. One of the very handy ones, because a Voltmeter is usually at hand is to measure the voltage across the resistor terminals. The power (heat) generated in the resistor is:

V squared divided by the resistor value R. or: V*V/R, where V is the DC voltage, or if it is AC the rms AC voltage. This is handy because you don't need to place an Amp meter in series to examine the current.

This way you can quickly make sure that the max power rating of your resistor is not being exceeded.
 
Last edited:
Hello guys!
Ok I finaly got the socond 7805 regulator. And you know what? It boots! Im so happy. A big step forward. 2 regulators was actually bad!

But now there is another problem. The keyboard dosent seems to work. The only key that work is "inst del".
The cursor is blinking and there is some garbage on the first line.
 

Attachments

  • 20230105_132427.jpg
    20230105_132427.jpg
    1.7 MB · Views: 8
Excellent work.

So you likely have one of two faults (or in your case both...):

1. The keyboard PIA.
2. The keyboard itself.

You can remove the keyboard cable from the logic board and then have one of two options:

1. Simulate the keyboard keys with a piece of wire. This will test out the keyboard PIA and logic.

2. Test the keyboard keys with a multimeter.

Dave
 
@daver2
I swapped place with the two PIA. The garbage is now gone. But there is no cursor.
I have no replacment PIA but I have a 6522.
 
Unfortunately, the 6522 is a VIA, not a PIA. They both have different pinouts - so are not interchangeable I am afraid...

It sounds like the 6520 keyboard PIA (C7) is dead and you require a replacement.

You may also have some dead keys as well of course once you replace the keyboard PIA.

Dave
 
I know from personal experience that the Western Design Centre W65C21 and the Motorola MC6821 are suitable replacements for 6520 PET PIAs although I'm sure there are other equivalents.

Alan
 
6520 = 6521 is a reasonable assumption (but not 6522 even though there is the same arithmetic progression between them!)

Alan's suggestion of the MC6821 (Motorola source) is also a good idea.

Dave
 
Ok guys! The Motorola MC6821 has arrived and the PET is fully alive! The keyboard works perfect! This is so cool! I finaly have a nice looking PET. And its working.

BUT its still not perfect. I have notice that the screen has some flickering. Its crystal clear but the picture shakes a bit. Maybe I should recap the CRT? Or do you guys think I need to demagnetize it?

Thank you guys for all help!
 
Good work on getting the PET back alive.

I would suggest posting a video for us first so we can see what the screen looks like first.

A picture paints a thousand words...

Dave
 
You should have credit for bringing it alive man.

Hmmmm I coulden upload the video. But here is a youtube link.

Its very hard to see on camera. But its fast flickering.
 
The only (obvious) thing to check is the amount of ripple on the monitor power supply.

There is an on-monitor voltage regulator (12V or 18V) and some smoothing etc. If there is ripple here (or at other DC voltage points on the monitor) that could explain your issue (as a first guess).

Dave
 
It is hard to see the fast flickering you say is there.

With this VDU, if there is ripple in the power supply, it tends to cause the raster scan to bounce up and down vertically at a few Hz, because there is some positive feedback in the V scan stage, just on the verge of damped oscillation. I cannot see any of that so probably, whatever is causing the problem, is not the power supply. Can you try taking another video to demonstrate the problem ?
 
I can just see a cyclic change in the brightness over approximately 8 seconds. I suspect this to be a 'beat' frequency between what is happening on the monitor and the camera (possibly).

Are you viewing the screen under daylight or artificial light? If artificial light, what do you observe if you view the screen in the dark?

Just trying to get some data points on the problem...

Can you also identify exactly which monitor PCB you have please. Take a photograph. There were a number of models that Commodore potentially used. It is possible that one of the intermediate DC voltage smoothing capacitors has gone faulty. There is one feeding the cathode output stage and one feeding the brightness control. All relatively low voltage still...

Dave
 
Last edited:
the light in the room is artificial.
It seems impossible to get the shakes on film. It's so small and fast. I could describe it as vibrations (tremors) that would occur if I were drilling with a drill.
I will open up the CRT and check as you suggest.
 
Back
Top