Just got myself a commodore pet setup but not working

[ScottishColin]

My 3016 also takes power directly to the monitor from the transformer through the same brown wiring setup.

Something I noticed - the photo of the 4032 from the web site has an earth wire running down from the monitor to the earth screw at the back left hand corner of the motherboard. Neither mine nor the OPs have that in place.

Colin.

[SiriusHardware]

I think that might just be a safety thing because when you have two heavily painted metal parts, ie, - monitor housing and main housing - screwed or bolted together there is no absolute guarantee that there will be a zero ohm connection between them, which there ideally must be for all of the metal cabinet parts to be equally well earthed - so there is often an earth bond wire running between the two to ensure that they are properly electrically connected together, even though you would intuitively expect that two metal objects pressed together would soon find a way to connect to each other even with a bit of paint in the way.

We have PSU units and other items in white painted steel casings at work, the lids / doors are all bonded to the main cabinets with thick earth wires even though the hinges are metal and theoretically conductive.

If the wire is present in this PET and not connected, it would do no harm to connect it.

[ajgriff]

Originally Posted by SiriusHardware

The photo of the one you referred to in the link only appears to show a single loom going from the mainboard to the monitor (From J7). If you look at the diagram for the monitor (not included in your dossier) it shows that power to the monitor is AC, going in on pins 1 and 2 of the connector at the monitor end.

I’ve attached another PDF with the VDU schematic and PSU layout diagram. The pair of twisted brown wires supply AC to pins 1 & 2 on the VDU board which seems to be a standard PET arrangement.

Originally Posted by ajgriff

I note that J10 isn't connected in the photo'.

Interestingly in the 4032 photo’ (see below) on Dave Dunfield’s site there’s no sign of a connector going to J10 and there isn’t one shown on the PSU layout diagram either.

Originally Posted by ScottishColin

Something I noticed - the photo of the 4032 from the web site has an earth wire running down from the monitor to the earth screw at the back left hand corner of the motherboard. Neither mine nor the OPs have that in place.

Looks like the earth wire is a user modification. Not relevant to anything in particular but the jeweller’s screwdriver lying on the main board of the Dunfield machine amuses me.

Alan

[pudwink]

Hi all
Many thanks for your quick reply I am at work at the moment but will look for tube glow later I did come across a YouTube video for the same computer and he discovered that the screen will not show anything unless it receives a signal to start it up.
https://youtu.be/sR6CnaRWPoI

[SiriusHardware]

I'm at work and can't look at the video just now but it is true that the monitors on these need to see sync signals in order for them to run. We just need to know if the monitor has any power at all before going down the road of looking for those sync signals, and the easiest telltale is whether you have a heater glow in the tube neck.

[SiriusHardware]

While you're in the monitor looking for the glow look for any other potential physical problems (input connector on monitor PCB hanging half way off, etcetera). If all seems OK there have a look with your scope for signals on pins 3 and 6 of 7486 IC UC2 on the mainboard. These are the Horizontal and Vertical drive output signals, without which the monitor will not start up.

If you see nothing there look on pins 39 and 40 of UB13, the CRTC Controller IC.

At the moment I'm assuming that because you get that startup jingle, a good portion of the machine is actually working.

[ScottishColin]

If you type ? CHR$(7) on a PET with an internal speaker, it should make a noise. This can be done even if the monitor has failed as long as the PET is booted OK. You could try this on the PET even if you can't see what you're typing. That should tell us whether it has booted or not.

Colin.

[ajgriff]

Originally Posted by SiriusHardware

At the moment I'm assuming that because you get that startup jingle, a good portion of the machine is actually working.

I think that’s a reasonable assumption and I don’t disagree with your suggestions regarding checking out the signals to the monitor. Nonetheless it would still be wise to verify all the voltage supply rails at an early stage, in line with good fault finding practice, and in the light of the previous repairs.

Alan

[SiriusHardware]

Oh, Yes .

This is just a quick diversion, to ascertain whether the monitor is at fault or whether the mainboard is failing to generate the sync signals. To cut the problem in half and see which half is good, and which bit bad, in effect.

Once we establish that it is the mainboard (if it is) then we'll take the usual sensible path, although that may depend to a certain extent upon how much time Pud is able to spare for sometimes exhausting and monotonous tests.

Certainly a check on the supplies would be sensible given the obvious rework which has been done in the regulator area.

[SiriusHardware]

If you type

on a PET with an internal speaker, it should make a noise.

Brilliant, that's exactly the sort of user side dodge you can only pick up from people who actually know the machines. So yes, please try that.

I had already gone looking for something like a BEEP or PLAY command in PET BASIC 4 which Pud could try entering and found nothing so far.

[SiriusHardware]

I note from a parallel 4032 thread over on VCFED - a machine with the other type of board as it happens - that it is possible for the startup chime to be present even when there are possible problems in zero page RAM, so the firmware which generates the chime must be careful not to use any RAM resources itself. If so that was a good decision by Commodore since it provides an indication that the CPU is able to execute firmware from the PROMs even when zero page RAM is not working. It's a useful diagnostic indicator.

[pudwink]

Hi
I have checked pins 3 and 6 of uc2 and there is activity on both pins and there is neck glow.
First time I have had a closer look at the board and noticed two wires have been cut off the power to board connection and soldered them underneath and there are corroded legs on some of the chips.

[SiriusHardware]

Oh, now that's unexpected.

Could you please post pictures of what you are getting on UC2 pins 3 and 6 and also state the Volts/Div and Time/Div settings at which the readings have been taken? Also throw in a picture of what you see on UC2 pin 8, again with the Volts/Div and Time/Div settings used.

For the moment, don't mess with the mainboard too much as there may be nothing wrong with it.

[ajgriff]

Also check for activity on pins 39 & 40 of UB13 (CRTC chip) as advised by Sirius earlier.

With regard to the PSU connector (J9) someone has soldered the wires directly because of poor contacts between the plug and socket (not an unheard of problem!). At some point you’ll want to come up with a better solution because the wires will have to be disconnected every time you need to remove the main board. For the moment just check that the correct voltages are appearing on the main board near to J9.

I wouldn’t be too worried about the slight corrosion on the IC pins although sometimes this sort of deterioration can cause chips to fail. In the end it’s a question of identifying faulty ICs if indeed there are any.

Alan

[SiriusHardware]

This is what threw me:-

there is activity on both pins

Just need to verify that those signals are what they should be. If they are we could actually be looking at a monitor fault.

[SiriusHardware]

I'll be out of circulation for about 2-3 hours from now. Will look in later to see if there are any developments.

[pudwink]

Hi
Please find the pictures requested just checking I have used the pets metal bottom case as the earth for the osciloscope is this ok and the settings are shown on the screen.
pic 1 ub13 pin 39
pic 2 ub13 pin 40
pic 3 uc2 pin 3
pic 4 uc 2 pin 6
pic 5 uc2 pin 8

[Mark1960]

The signals are noisy due to using the case as an earth but they look like they have the correct activity. It would be better to use 0 v reference from the ICs where you are measuring outputs. Usually pin 7 on 14 pin ICs or pin 8 on 16 pin, but check schematic or IC pinout in datasheet to confirm.

Were you able to see an orange glow on the back of the CRT?

[pudwink]

Hi
Yes glow from the back of the tube.

[ajgriff]

Adjusting the timebase would make the waveforms easier to see. I've attached a copy of the 'scope waveforms extracted from the manual. The numbers relate to points on the schematic I prevously posted. We are interested in 3 and 13 (vertical and horizontal drive). Note the timebase settings.

Alan