Re: Commodore PET 2001 voltage regulators
 

Thanks for checking those connections - when stating the results of resistance measurements could you also please state the units of measurement as well as the figures (so Ohms, K-Ohms M-Ohms, etcetera). I guess with your stated figures of 620 and 580 you mean Ohms, but maybe not.

If your meter has a 'diode test' range on it could you please do them again but this time measure all five with the probes placed red to tapered end, black to flat end, and then measure them all again but this time with black probe to tapered end, red probe to flat end. In this mode your meter indicates the voltage drop across the diode rather than its resistance.

Incidentally these measurements need to be made with JP8 plug unplugged, otherwise you may be measuring the mains transformer as well as the diodes.

Re: Commodore PET 2001 voltage regulators
 

For ease of reference I'm attaching a copy of the 320351 board layout. This might help in identifying test points in relation to the schematic as component designations are clearly marked.

Alan

Re: Commodore PET 2001 voltage regulators
 

I can't help feeling that it would be a good idea to lift one end of each of the suspect rectifier diodes (CR1, CR2, CR4, CR5 & CR3) in order to carry out proper open circuit diode tests. If nothing else this would at least eliminate them from our enquiries. In circuit testing can produce such confusing results. The solder pads are quite large and with a little deft application of the soldering iron it should be possible to avoid troublesome board damage.

It does concern me a little that the board has been worked on before which can make things tricky in diagnostic terms especially if the previous repairer gave up and consigned the PET to the loft in frustration.

I've just switched on my PET 2001 for the first time in a while and am relieved to say that it's still working well.

Alan

Re: Commodore PET 2001 voltage regulators
 

I'm not sure (because we don't know Retromit well yet) how confident he (or possibly she) is at desoldering components cleanly especially in antique double sided PCBs, so I was hoping to see an obviously bad diode without desoldering any, but if the results are inconclusive then you are right, it may be necessary to measure them with one end disconnected from the circuit. If we do identify a bad one in-circuit then it would be easier and less stressful to the PCB to snip out the diode body and remove the legs separately.

Going all the way back to #1 we can see that the machine has worked for a period, however brief, so we can say that the various mods are harmless and probably original, except for the socketing of the RAM which was obviously done when someone had a RAM problem at some point in the past. I have some concerns that the machine running without +5V but with the other supplies still present may have caused some device damage that was not there originally.

Re: Commodore PET 2001 voltage regulators
 

Let's see how the OP feels about a bit of soldering work. With regard to possible damage due to having no +5V lines the fact that the PET briefly showed signs of life again @ post #11 is a hopeful sign. However this again indicates a possible poor connection/joint rather than complete component failure. I must admit that I'd be doing a bit of wiggling and prodding (plastic chopstick being the instrument of choice!).

Alan

Re: Commodore PET 2001 voltage regulators
 

The case history as I understand it is that when acquired it had an apparent display fault, was fixed by replacement of the screen RAM, was turned off apparently working but on next start, would not turn on and, on investigation, had lost its 5V supplies.

The machine has not been seen working since, so we'll just have to see what happens once the supplies are up again.

Re: Commodore PET 2001 voltage regulators
 

My understanding of post #11 is that the PET did briefly produce a random character screen after the OP had carried out some of the suggested checks before dying again.

Alan

Re: Commodore PET 2001 voltage regulators
 

Indeed so, I should pay more attention.

Re: Commodore PET 2001 voltage regulators
 

Hi,

Thanks for all the replies/thoughts, and apologies for my late update on these.

Just from some of the replies, I've included the original screen shot of when I first aquired the PET. It was from this that I replaced the video RAM which (briefly) showed the Commodore BASIC screen. After switching the PET off, then back on, this is where I am now.

It did briefly kick into life (with the garbled screen picture I posted), but since them the screen has been blank.

I'll start to go through the other replies, and some more tests that have been carried out.

Regarding soldering, I'm happy to do this. I have worked previously on other 8-bit machines, and have soldered/desoldered. This is the first time I've ever had the chance to look over a PET.

Right, I'll update with the rest of the findings!

Thanks,
Tim.

Re: Commodore PET 2001 voltage regulators
 

Yes, that pattern is more reminiscent of a problem with the display RAM which you obviously did fix when you replaced that. Just have to get to the bottom of this ******* power problem and then we'll see where we really are.

Good to know that you are confident with soldering, we never really know where to pitch advice initially - if we aim too low we run the risk of patronising someone who has been working with electronics for years but if we aim high we could end up frightening off a newcomer who doesn't understand a word of what we are saying. We usually find the right level in the end.

Re: Commodore PET 2001 voltage regulators
 

For the diode test, I carried out the following. I believe the output is in ohms, but with this being a basic multimeter, there is no value for the unit.

Multimeter on diode test and red end tapered, blank flat.

All the bottom four diodes (X and Y, plus the two others near the silver capacitor) registered a value of '1'; the diode (W) near the capacitor registered 1689.

When black was on the tapered end, and red on flat:

The bottom four diodes (left to right) measured:

1 = 570
2 = 591
3 = 578
4 = 583
W (near the capacitor) = 532

Re: Commodore PET 2001 voltage regulators
 

I looked over the resistance readings again, and apologies for missing the units that should have been applied. When setting the multimeter to 2000 ohms range, the following was carried out:

Black on flat end, red tapered:

All bottom four diodes registerd '1'. The diode near the capacitor (W) measured a 4 figure value.

Red on flat end, black tapered:

All showed a four figure value in the 2000 ohms range.

Re: Commodore PET 2001 voltage regulators
 

If your meter was on 'Diode test' (was it?) then those readings in #31 look broadly OK. If you try doing one of the later group again you may see a decimal point ahead of the three figures, ie, '.570'. If so that means a forward voltage drop of 0.570 - generally speaking a forward voltage reading of around 0.5V to 0.6V is healthy for a Silicon diode, which these are.

Your resistance readings also look OK to me.

So it doesn't look like that is it either. :(

Re: Commodore PET 2001 voltage regulators
 

Please don't worry about patronising me! I am a beginner who enjoys working on these old machines. I'm just self-taught, but happy to keep learning (and am learning!) each day. I just like to work on these machines and try and get them working again. Again, thanks for all the help you've already provided.

Re: Commodore PET 2001 voltage regulators
 

Yes, the meter was on diode test (it looks to be the same function selection as the continuity test). It's only a very basic multimeter (Astroai am33d).

Re: Commodore PET 2001 voltage regulators
 

I am just wondering why there was a difference in the diode test for the diode near the capacitor (marked W on the picture you gave). When the multimeter was on diode test with the red end on the tapered side and the black on the flat side, then all the other diodes registered a value of '1', whereas the 'W' diode registered 1689?

I thought they would all register '1'?

Re: Commodore PET 2001 voltage regulators
 

Going back to some measurements you made earlier -

That 7.03 volts on the output of the positive voltage rectifier is not actually far short of what it should be but it's low, so something may be loading it. We'll get to that in a bit. Let's just first find out how things stand now. Sorry to take you round in circles but when things keep changing we have little choice but to go back and re-check. Can we ask you to make the following measurements again:-

Referring to the layout diagram kindly attached by AJ in post #22, please measure the DC voltage (with respect to 0V) on

-CR1 / CR2 tapered ends (which I previously marked as 'Z')

-JP1 pin 4

Electrically speaking, these are the same place / same circuit node so if they are not at exactly the same voltage there must be an open circuit or high resistance between them.

If the voltage on 'Z' has changed again and is the same as the voltage on JP1 pin 4, then it looks as though we have an intermittent fault to contend with. I suppose this is to be anticipated since we have already seen the supply present, then absent, then briefly present, then absent again.

By the way - what / where are you using for your 0V point to put the black
meter lead on when measuring voltages?

Re: Commodore PET 2001 voltage regulators
 

You are measuring the diodes in-circuit, so the reading you see is the reading from not only the diode but other components around it. In the case of the 'W' diode, that one has a large electrolytic capacitor and the inputs of the two 5V regulators, to name just a few things, connected across it. That is why AJ suggested desoldering one end of each diode so that when you made your measurements you would ONLY be measuring the diodes.

If one of the diodes had been short-circuit or very low resistance (which is the most usual way for them to fail) then that would have been obvious even when measuring them in-circuit. There are times when it's reasonable to try to measure things in-circuit and times when you definitely should separate the suspect component from everything else it is connected to before trying to measure it.

At the moment my feeling is the diodes are OK, but we may come back to them.

Re: Commodore PET 2001 voltage regulators
 

Grr, I really shouldn't try to do this late at night. I meant JP8 pin 4, sorry.

Re: Commodore PET 2001 voltage regulators
 

If it's ever required, I still have the two test EPROMs with the daver2 and slothir's test code on them if that'll help. Happy to post them on to the OP of this thread.