Commodore PET 4032/8050/3022 - Page 9 - UK Vintage Radio Repair and Restoration Discussion Forum

ScottishColin · 18th Jul 2023, 10:01 am

OK - done some cleaning and re-checking and 3/39 look better now.

Here's the 6502 anomolies:

pin 37 has a good wave form but is a little low on voltage at 4.27V

Pin 2 sits high at 4.77V

Pins 7 and 9 have good waveforms, but are at 3.5V and the frequency on both swicthes between 333 and 500 Mhz.

Others to follw.

Colin.

SiriusHardware · 18th Jul 2023, 10:33 am

If the CPU is running out of control or even under firmware control you would not necessarily expect to see steady waveforms or frequencies on those pins - that is why we use a NOP generator when required, to make the CPU execute one known instruction over and over again, which in turn generates steady, repetitive waveforms on the address lines and on the outputs from the address decoder.

We'll get to those shortly, but for now continue checking out the clock chain.

ScottishColin · 18th Jul 2023, 12:18 pm

OK here they are. I'm guessing that UE5 and UD2 don't look right?

Colin.

SiriusHardware · 18th Jul 2023, 12:44 pm

(Frown).

Very little of that looks right, but then I am depending on the diagram being largely correct. For example, UD3 pin 8 clock-in (pin 8) is shown connected back to UE5 pin 6, which is the buffered output from the 16MHz master oscillator and you say you do have 16MHz there.

As you might guess, you should have 16MHz on UD3 pin 8, and therefore 8MHz on UD3 pin 5, 4MHz on UD3 pin 9, 2MHz on UD3 pin 2, 1MHz on UD3 pin 12.

Check (power off) for continuity from UE5 pin 6 to UD3 pin 8.

ScottishColin · 18th Jul 2023, 12:56 pm

I'm reading page 30 of the service manual PDF which shows UE5 pin 6 connecting to UE2 pin 14 (see attached). Am I reading this wrong?

Colin.

SiriusHardware · 18th Jul 2023, 1:05 pm

Look at the section upper-left, which is the 16MHz oscillator comprised of three gates of UE4 followed by one gate of UE5. The output from UE5 pin 6 is connected to UD3 pin 8 and to UE3 pin 8 as well.

The IC to the right of the 16MHz oscillator section which has its pin 6 connected to UE2 pin 14 is, I think , UE3, although the '3' is a bit crumpled and could certainly be mistaken for a '5'.

ScottishColin · 18th Jul 2023, 1:21 pm

Can we tae a small diversion to help me?

Why is it that when I change the time/div setting for the same pin, the frequency reading changes (see attached file)?

Thanks.

Colin.

ScottishColin · 18th Jul 2023, 1:24 pm

Continuity confirmed.

Colin.

Quote:

Originally Posted by SiriusHardware

(Frown).

Very little of that looks right, but then I am depending on the diagram being largely correct. For example, UD3 pin 8 clock-in (pin 8) is shown connected back to UE5 pin 6, which is the buffered output from the 16MHz master oscillator and you say you do have 16MHz there.

As you might guess, you should have 16MHz on UD3 pin 8, and therefore 8MHz on UD3 pin 5, 4MHz on UD3 pin 9, 2MHz on UD3 pin 2, 1MHz on UD3 pin 12.

Check (power off) for continuity from UE5 pin 6 to UD3 pin 8.

ScottishColin · 18th Jul 2023, 1:28 pm

All of those frequencies on UD3 check out if (and only if) I leave the time/div setting at 1 us. If I change the setting to get fewer readings on the screen, the frequency changes.

I believe that's my confusion here.

Colin.

Quote:

Originally Posted by SiriusHardware

(Frown).

Very little of that looks right, but then I am depending on the diagram being largely correct. For example, UD3 pin 8 clock-in (pin 8) is shown connected back to UE5 pin 6, which is the buffered output from the 16MHz master oscillator and you say you do have 16MHz there.

As you might guess, you should have 16MHz on UD3 pin 8, and therefore 8MHz on UD3 pin 5, 4MHz on UD3 pin 9, 2MHz on UD3 pin 2, 1MHz on UD3 pin 12.

Check (power off) for continuity from UE5 pin 6 to UD3 pin 8.

SiriusHardware · 18th Jul 2023, 1:47 pm

Do you see the sample rate, in orange on the lower right hand corner of your capture images?

Only 1MHz in your first capture which is a mere 1/16th of the frequency you are trying to look at, so the scope isn't 'looking' at the signal often enough to 'see' every change.

In your second capture the sample rate is 48MHz, three times higher than that of the frequency you are trying to look at, so the scope doesn't miss any changes in the signal level. It has to be able to see every change in order to be able to calculate the number of cycles per second, otherwise known as the frequency.

The thing you need to be aware of is that the sample rate must be higher than the frequency you are expecting to see.

SiriusHardware · 18th Jul 2023, 2:24 pm

I don't remember you having problems like this the first time - at the time you were probably using the software which was bundled with the scope, are you using something else now by any chance?

ScottishColin · 18th Jul 2023, 3:38 pm

Aha - thank you.

So on my scope's software, there's no way I can see to directly change the sampling rate - it's an indirect consequence of changing the time/div timing.

Is this how all scopes work or just my cheap one?

I'll get on with re-measuring soon.

Colin.

Quote:

Originally Posted by SiriusHardware

Do you see the sample rate, in orange on the lower right hand corner of your capture images?

Only 1MHz in your first capture which is a mere 1/16th of the frequency you are trying to look at, so the scope isn't 'looking' at the signal often enough to 'see' every change.

In your second capture the sample rate is 48MHz, three times higher than that of the frequency you are trying to look at, so the scope doesn't miss any changes in the signal level. It has to be able to see every change in order to be able to calculate the number of cycles per second, otherwise known as the frequency.

The thing you need to be aware of is that the sample rate must be higher than the frequency you are expecting to see.

ScottishColin · 18th Jul 2023, 3:39 pm

Nope - it makes me wonder how I've got this far really....

Colin.

Quote:

Originally Posted by SiriusHardware

I don't remember you having problems like this the first time - at the time you were probably using the software which was bundled with the scope, are you using something else now by any chance?

SiriusHardware · 18th Jul 2023, 3:59 pm

All digital scopes are essentially sampling scopes and you have to watch out for this particular gotcha, yes, but posher scopes will tend to have higher maximum sampling rates. The good news is that the frequencies in the PET don't stay high for very long and are soon down to 1MHz or less.

With the above experience in mind, could you please try again to compile a table of pin states or frequencies for the following:

UE4 pin 10
UE4 pin 8

UE6 pin 8
UE6 pin 5
UE6 pin 9
UE6 pin 2
UE6 pin 12

UE7 pin 8
UE7 pin 5
UE7 pin 9
UE7 pin 2

UE5 pin 8
UE5 pin 3
UE5 pin 4
UE5 pin 5
UE5 pin 6
UE5 pin 11
UE5 pin 12

UD2 pin 6
UD2 pin 8

ScottishColin · 18th Jul 2023, 4:36 pm

Here they are. UE5 and UD2 aren't being helpful whatever settings I use.

Colin.

SiriusHardware · 18th Jul 2023, 5:07 pm

Everything up to and including the UE7 results looks a lot better now.

For the remainder, ugh, my bad. I've actually managed to make the same U3 / U5 identification error when reading the diagram earlier on in the thread, all because that '3' in 'UE3' is so indistinct.

So for these IC pins which I've asked you more than once to scope out, forget these ones....

Code:

UE5 pin 8
UE5 pin 3
UE5 pin 4
UE5 pin 5
UE5 pin 6
UE5 pin 11
UE5 pin 12

Please substitute these IC pin numbers below and try again. You don't actually need to put the results into a PDF - if you want to preserve the formatting of a chunk of plain text (including TAB spacing) when pasting it into a post, put <code> tags around it. (Like I've done in this post)

Code:

UE3 pin 8
UE3 pin 3
UE3 pin 4
UE3 pin 5
UE3 pin 6
UE3 pin 11
UE3 pin 12

Presenting the results in this way will make them more immediately readable and easier for anyone following or helping to take in. There will be times when we actually will need to see scope traces etc, but not just yet.

Based on what you've said about the UD2 pins (which were the correct pins) I'm hoping the results from the UE3 pins will be interesting. Hint: I think all of those pins should have 1MHz on them.

ScottishColin · 18th Jul 2023, 5:21 pm

I can save us all the trouble of a table. Only one reading from UE3/8 which was 16Mhz. No readings at all from the other pins.

Colin.

SiriusHardware · 18th Jul 2023, 5:34 pm

Lovely. (Again a mistake by me, it should indeed be 16MHz on pin 8, but 1MHz on the others).

We just need to check the frequency on UE3 pin 1 (Should be 1MHz) and the state of UE3 pin 2 (should be static 5V) and the state of UE3 pin 9 (should be logic '1')?

Let me know what you have.

SiriusHardware · 18th Jul 2023, 5:45 pm

I'm going to be out for the evening later: If you find that all is as stated in #178 but no 1MHz signals on the output pins of UE3 then that would suggest that UE3 is toast. If not, pause and let us know what isn't right.

But if so, remove UE3 in one piece and socket it - do you by any chance have the same type of IC socketed in your working PET, that you could remove and drop in for test / diagnostic purposes?

Oh: Let's not forget the fundamentals: Does UE3 have +5V power going to it? Pins 14 (+5V) and 7 (0V).

ScottishColin · 18th Jul 2023, 8:25 pm

Replaced and now getting the correct readings out of UE3, but no chirps or screen activity yet.

UD2/6 and UD2/8 now give 1Mhz readings too.

Onto the next one I guess.

Colin.

18th Jul 2023, 10:01 am	#161
ScottishColin Octode Join Date: May 2012 Location: Perth, Scotland Posts: 1,601	Re: Commodore PET 4032/8050/3022 OK - done some cleaning and re-checking and 3/39 look better now. Here's the 6502 anomolies: pin 37 has a good wave form but is a little low on voltage at 4.27V Pin 2 sits high at 4.77V Pins 7 and 9 have good waveforms, but are at 3.5V and the frequency on both swicthes between 333 and 500 Mhz. Others to follw. Colin.

18th Jul 2023, 10:33 am	#162
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 11,082	Re: Commodore PET 4032/8050/3022 If the CPU is running out of control or even under firmware control you would not necessarily expect to see steady waveforms or frequencies on those pins - that is why we use a NOP generator when required, to make the CPU execute one known instruction over and over again, which in turn generates steady, repetitive waveforms on the address lines and on the outputs from the address decoder. We'll get to those shortly, but for now continue checking out the clock chain.

18th Jul 2023, 12:44 pm	#164
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 11,082	Re: Commodore PET 4032/8050/3022 (Frown). Very little of that looks right, but then I am depending on the diagram being largely correct. For example, UD3 pin 8 clock-in (pin 8) is shown connected back to UE5 pin 6, which is the buffered output from the 16MHz master oscillator and you say you do have 16MHz there. As you might guess, you should have 16MHz on UD3 pin 8, and therefore 8MHz on UD3 pin 5, 4MHz on UD3 pin 9, 2MHz on UD3 pin 2, 1MHz on UD3 pin 12. Check (power off) for continuity from UE5 pin 6 to UD3 pin 8.

18th Jul 2023, 12:56 pm	#165
ScottishColin Octode Join Date: May 2012 Location: Perth, Scotland Posts: 1,601	Re: Commodore PET 4032/8050/3022 I'm reading page 30 of the service manual PDF which shows UE5 pin 6 connecting to UE2 pin 14 (see attached). Am I reading this wrong? Colin. Attached Thumbnails

18th Jul 2023, 1:05 pm	#166
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 11,082	Re: Commodore PET 4032/8050/3022 Look at the section upper-left, which is the 16MHz oscillator comprised of three gates of UE4 followed by one gate of UE5. The output from UE5 pin 6 is connected to UD3 pin 8 and to UE3 pin 8 as well. The IC to the right of the 16MHz oscillator section which has its pin 6 connected to UE2 pin 14 is, I think , UE3, although the '3' is a bit crumpled and could certainly be mistaken for a '5'.

18th Jul 2023, 1:47 pm	#170
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 11,082	Re: Commodore PET 4032/8050/3022 Do you see the sample rate, in orange on the lower right hand corner of your capture images? Only 1MHz in your first capture which is a mere 1/16th of the frequency you are trying to look at, so the scope isn't 'looking' at the signal often enough to 'see' every change. In your second capture the sample rate is 48MHz, three times higher than that of the frequency you are trying to look at, so the scope doesn't miss any changes in the signal level. It has to be able to see every change in order to be able to calculate the number of cycles per second, otherwise known as the frequency. The thing you need to be aware of is that the sample rate must be higher than the frequency you are expecting to see.

18th Jul 2023, 2:24 pm	#171
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 11,082	Re: Commodore PET 4032/8050/3022 I don't remember you having problems like this the first time - at the time you were probably using the software which was bundled with the scope, are you using something else now by any chance?

18th Jul 2023, 3:59 pm	#174
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 11,082	Re: Commodore PET 4032/8050/3022 All digital scopes are essentially sampling scopes and you have to watch out for this particular gotcha, yes, but posher scopes will tend to have higher maximum sampling rates. The good news is that the frequencies in the PET don't stay high for very long and are soon down to 1MHz or less. With the above experience in mind, could you please try again to compile a table of pin states or frequencies for the following: UE4 pin 10 UE4 pin 8 UE6 pin 8 UE6 pin 5 UE6 pin 9 UE6 pin 2 UE6 pin 12 UE7 pin 8 UE7 pin 5 UE7 pin 9 UE7 pin 2 UE5 pin 8 UE5 pin 3 UE5 pin 4 UE5 pin 5 UE5 pin 6 UE5 pin 11 UE5 pin 12 UD2 pin 6 UD2 pin 8

18th Jul 2023, 5:07 pm	#176
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 11,082	Re: Commodore PET 4032/8050/3022 Everything up to and including the UE7 results looks a lot better now. For the remainder, ugh, my bad. I've actually managed to make the same U3 / U5 identification error when reading the diagram earlier on in the thread, all because that '3' in 'UE3' is so indistinct. So for these IC pins which I've asked you more than once to scope out, forget these ones.... Code: UE5 pin 8 UE5 pin 3 UE5 pin 4 UE5 pin 5 UE5 pin 6 UE5 pin 11 UE5 pin 12 Please substitute these IC pin numbers below and try again. You don't actually need to put the results into a PDF - if you want to preserve the formatting of a chunk of plain text (including TAB spacing) when pasting it into a post, put <code> tags around it. (Like I've done in this post) Code: UE3 pin 8 UE3 pin 3 UE3 pin 4 UE3 pin 5 UE3 pin 6 UE3 pin 11 UE3 pin 12 Presenting the results in this way will make them more immediately readable and easier for anyone following or helping to take in. There will be times when we actually will need to see scope traces etc, but not just yet. Based on what you've said about the UD2 pins (which were the correct pins) I'm hoping the results from the UE3 pins will be interesting. Hint: I think all of those pins should have 1MHz on them. Last edited by SiriusHardware; 18th Jul 2023 at 5:16 pm.

18th Jul 2023, 5:21 pm	#177
ScottishColin Octode Join Date: May 2012 Location: Perth, Scotland Posts: 1,601	Re: Commodore PET 4032/8050/3022 I can save us all the trouble of a table. Only one reading from UE3/8 which was 16Mhz. No readings at all from the other pins. Colin.

18th Jul 2023, 5:34 pm	#178
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 11,082	Re: Commodore PET 4032/8050/3022 Lovely. (Again a mistake by me, it should indeed be 16MHz on pin 8, but 1MHz on the others). We just need to check the frequency on UE3 pin 1 (Should be 1MHz) and the state of UE3 pin 2 (should be static 5V) and the state of UE3 pin 9 (should be logic '1')? Let me know what you have.

18th Jul 2023, 5:45 pm	#179
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 11,082	Re: Commodore PET 4032/8050/3022 I'm going to be out for the evening later: If you find that all is as stated in #178 but no 1MHz signals on the output pins of UE3 then that would suggest that UE3 is toast. If not, pause and let us know what isn't right. But if so, remove UE3 in one piece and socket it - do you by any chance have the same type of IC socketed in your working PET, that you could remove and drop in for test / diagnostic purposes? Oh: Let's not forget the fundamentals: Does UE3 have +5V power going to it? Pins 14 (+5V) and 7 (0V).

18th Jul 2023, 8:25 pm	#180
ScottishColin Octode Join Date: May 2012 Location: Perth, Scotland Posts: 1,601	Re: Commodore PET 4032/8050/3022 Replaced and now getting the correct readings out of UE3, but no chirps or screen activity yet. UD2/6 and UD2/8 now give 1Mhz readings too. Onto the next one I guess. Colin.