UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Computers

Notices

Vintage Computers Any vintage computer systems, calculators, video games etc., but with an emphasis on 1980s and earlier equipment.

Closed Thread
 
Thread Tools
Old 2nd Mar 2021, 8:49 pm   #761
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,265
Default Re: Non-working Commodore PET 3016

Quote:
Originally Posted by ScottishColin View Post
Sorry Mark - could you point me at the post where I did this before? I'm learning fast but I'm not quite getting the meaning of your post.

Colin.


Quote:
Originally Posted by Mark1960 View Post
I think it would be good to check the data lines again with the duff prom replaced, see if DA0 is still showing low logic levels from one or more proms.

You could try first with the 6502 direct connected, but its not running NOP now so the pattern will be different and you might not be getting chip select on every prom now.

Either with or without the NOP set up.

Yellow probe to pin 20 chip select of each prom and green probe to pin 9 DA0 of any prom. Ideally check pin 9, 10, 11 etc for all data lines on the prom, but I’d like to see if the voltage level on DA0 is better now that the bad prom is removed.

With the NOP removed and 6502 data lines connected, you might not have a chip enable for every prom as it depends on what the software is running. It won’t be a regular pattern on chip select, but so long as it goes low the prom is being enabled and we can see the level on DA0.
Mark1960 is offline  
Old 2nd Mar 2021, 9:36 pm   #762
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

So UD9, 8 and 6 are consistently the same voltage level (c 4V) when I check their own pin 20 against their own pin 10 (for example). UD7 is higher (slightly).

Interestingly, the height of the wave seems to follow the Pin 20 I use. So if I test Pin 20 of UD7 (new IC) against pin 10 of UD8, I get higher voltages than if I select pin 20 of UD8 of pin 10 of UD8.

Does that make sense?

Colin.
ScottishColin is offline  
Old 2nd Mar 2021, 9:49 pm   #763
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Non-working Commodore PET 3016

As explained earlier, D0, D1, D2....D6, D7 of all of the PROMs are connected together, so when UD7 is selected its data output from (for example, D0) travels out along the databus D0 line and goes not only to the CPU but to the D0 pin of every other PROM (and every other device on the databus) as well.

The increased height signals you see on the data pins of UD6, UD8, UD9 when UD7 is selected are coming out of UD7 and travelling along the databus to the pins of UD6 / UD8 / UD9, they are not coming out of UD6 / UD8 / UD9.

Try the same observation in reverse. With scope channel 1 on the chip select for UD6, look at the data pins of UD7 using scope channel 2. You will find the signals there are 0V-4V, because they are coming out of UD6 and travelling to the UD7 pins via the databus.

Last edited by SiriusHardware; 2nd Mar 2021 at 9:54 pm.
SiriusHardware is online now  
Old 2nd Mar 2021, 10:00 pm   #764
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

You are (as you knew) right. Thanks though - another thing to learn for me.

Is there anything to be worried about regarding the higher voltage that I get on UD7? Or is it OK as it's a new chip?

Colin.


Quote:
Originally Posted by SiriusHardware View Post
As explained earlier, D0, D1, D2....D6, D7 of all of the PROMs are connected together, so when UD7 is selected its data output from (for example, D0) travels out along the databus D0 line and goes not only to the CPU but to the D0 pin of every other PROM (and every other device on the databus) as well.

The increased height signals you see on the data pins of UD6, UD8, UD9 when UD7 is selected are coming out of UD7 and travelling along the databus to the pins of UD6 / UD8 / UD9, they are not coming out of UD6 / UD8 / UD9.

Try the same observation in reverse. With scope channel 1 on the chip select for UD6, look at the data pins of UD7 using scope channel 2. You will find the signals there are 0V-4V, because they are coming out of UD6 and travelling to the UD7 pins via the databus.
ScottishColin is offline  
Old 2nd Mar 2021, 10:07 pm   #765
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Non-working Commodore PET 3016

I am honestly sure it is just because it is a different chip, all the electronics inside that EPROM are completely different to those in the original PROMs, so it would actually be more surprising if the EPROM did have exactly the same output levels as the original PROMs.
SiriusHardware is online now  
Old 2nd Mar 2021, 10:13 pm   #766
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Non-working Commodore PET 3016

I take it from what you say, anyway, that we no longer have suspiciously low levels on any data lines, such as the D0 data level which Mark was concerned about?
SiriusHardware is online now  
Old 2nd Mar 2021, 10:18 pm   #767
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

They look nicely consistent to me.

Colin.

Quote:
Originally Posted by SiriusHardware View Post
I take it from what you say, anyway, that we no longer have suspiciously low levels on any data lines, such as the D0 data level which Mark was concerned about?
ScottishColin is offline  
Old 2nd Mar 2021, 10:25 pm   #768
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Non-working Commodore PET 3016

OK. Would you mind just doing a position check on UD8 - UD9 for me?

UD8 should be the one marked
901447-24 (....Edit)

UD9 should be the one marked
901465-03 (...Kernel)

I remember that it was still easily possible to read the number off UD9 - the other one, if you really try hard and under the right light you can see that the number does end in '24'.
SiriusHardware is online now  
Old 2nd Mar 2021, 10:34 pm   #769
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

UD9 confirmed.

UD8 I can read 901447 but for the life of me I can't see the 24.

Colin.
ScottishColin is offline  
Old 2nd Mar 2021, 11:05 pm   #770
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Non-working Commodore PET 3016

That's OK. I believe the 'EDIT' PROM is the only one of the group to carry the 901447, so that looks like the right PROM in the right place. If you were in any doubt you could read the 2K of code out of it and compare the first few bytes with the 'edit-2-n.901447-24.bin2009-08-18' file on the website.

So where are we now - the CPU can run, at long last, with all of the ICs fitted and we have normal looking data levels on all data lines, but we still don't have video signals being generated. We could still replace the data bus buffers of course.

I'll throw this out to those with specific PET experience, are there any hardware states or zones of activity which are maintained by the system under software control? For example, keyboard scanning is one likely area to look at but are there any other simpler hardware activities or cues which happen under the control of the kernel code?

One of the first jobs the Kernel code probably does is to initialise the various I/O chips, so if there are any port pins which are known to be set to specific states during the initial startup process, that would be the kind of thing we could be looking for. Things like the cassette motor drive control outputs - the I/O pins on the 40-pin ICs which drive those will be inputs until they are changed to outputs under software control with the cassette motor drive set to 'off'. That kind of thing.

There also remains the future possibility of using your Mega as a kind of substitute 6502, not as a direct drop in CPU replacement but as a way of exerting specific states on the address bus and control lines and reading the data bus to see what comes back.

As an example, if you were to use the Mega to apply address 0xF000 to the address bus, that would select PROM UD9. If you also use the Mega to assert the read / write line (and possibly in this case the outgoing clock signals) into the correct state, then the data in the first location of UD9 should appear on the D0-D7 outputs of the PROM and travel down the data bus to the CPU socket where it can be read by the Arduino.

For the moment that is still probably a little bit too advanced so we'll stick with more conventional means for now.
SiriusHardware is online now  
Old 3rd Mar 2021, 12:24 am   #771
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Non-working Commodore PET 3016

Let's try following the timing chain for a while.

With all ICs inserted as normal, power on, please scope the following IC pins:

UH2 pin 2
UG9 pin 3
UH6 pin 12

You should have the same signal on all three of the above pins. If you don't even have a signal on UH2 pin 2, stop and report back.

Assuming you do, could you then look for further signals on

UH6 pin 3
UH9 pins 12, 9, 8, 11

UG9 pin 6
UH7 pin 9
(Both of the above two should be the same)

UH7 pin 12
SiriusHardware is online now  
Old 3rd Mar 2021, 2:25 am   #772
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,265
Default Re: Non-working Commodore PET 3016

Quote:
Originally Posted by SiriusHardware View Post
So where are we now - the CPU can run, at long last, with all of the ICs fitted and we have normal looking data levels on all data lines, but we still don't have video signals being generated. We could still replace the data bus buffers of course.
I think there is no indication that the data buffers need to be changed now that the faulty prom seems to have been the cause of the low voltage levels on DA0.

Your suggestion to check the divider chain is the right way to proceed. It should be possible to get a garbage display even if the 6502 is not running code correctly.
Mark1960 is offline  
Old 3rd Mar 2021, 10:31 am   #773
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Non-working Commodore PET 3016

It might still be that the data buffers are not transmitting correct data back and forth between the unbuffered and buffered side of the data bus.

Let's see how Colin gets on with the divider test. At the moment I'm mainly looking for why there is no 'Horizontal Drive' signal since that is what is fundamentally needed for the monitor to spring to life.

The last time we were in this area we ground to a halt because Colin didn't have much in the way of test gear and the frequency meter was rather choosy about what it liked to show, but he has better gear now.
SiriusHardware is online now  
Old 3rd Mar 2021, 11:56 am   #774
ajgriff
Nonode
 
ajgriff's Avatar
 
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,583
Default Re: Non-working Commodore PET 3016

Has Colin scoped the vertical, horizontal and video drive signals at the motherboard socket (J7?) connection to the monitor? The expected signals are indicated on the schematic for the monitor. Worth giving the pins and contacts at J7 a good clean as well.

Alan

Last edited by ajgriff; 3rd Mar 2021 at 12:11 pm. Reason: Additional comment.
ajgriff is online now  
Old 3rd Mar 2021, 12:33 pm   #775
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

Nothing on H2/2 (unless I don't earth in which case I get a nice sine wave which I assume is just background?).

I tried this with and without the NOP device in place.

Colin.


Quote:
Originally Posted by SiriusHardware View Post
Let's try following the timing chain for a while.

With all ICs inserted as normal, power on, please scope the following IC pins:

UH2 pin 2
UG9 pin 3
UH6 pin 12

You should have the same signal on all three of the above pins. If you don't even have a signal on UH2 pin 2, stop and report back.

Assuming you do, could you then look for further signals on

UH6 pin 3
UH9 pins 12, 9, 8, 11

UG9 pin 6
UH7 pin 9
(Both of the above two should be the same)

UH7 pin 12
ScottishColin is offline  
Old 3rd Mar 2021, 1:25 pm   #776
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Non-working Commodore PET 3016

Quote:
Originally Posted by ajgriff View Post
Has Colin scoped the vertical, horizontal and video drive signals at the motherboard socket (J7?)

Alan
Posts #754-755 (at the IC pins rather than the connector)
SiriusHardware is online now  
Old 3rd Mar 2021, 1:35 pm   #777
ajgriff
Nonode
 
ajgriff's Avatar
 
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,583
Default Re: Non-working Commodore PET 3016

Understood. I think I'll stick to an observational role with this one!

Alan
ajgriff is online now  
Old 3rd Mar 2021, 1:55 pm   #778
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Non-working Commodore PET 3016

Was a perfectly good suggestion. As I referred to IC pin numbers rather than signals it wouldn't have been immediately obvious what was being looked for there.

Today's initial checks were aimed at trying to find out why there is no Horizontal_Drive signal because that's what the monitor needs in order to get going (it does not have an independent line oscillator like a TV or most other monitors).

The trail seems to have run dry sooner than expected, we'll have to try to work out why that is - unfortunately I am really busy at work these days so I can't really think about it properly until the evening - maybe Mark can take the late afternoon (his morning) shift.
SiriusHardware is online now  
Old 3rd Mar 2021, 6:20 pm   #779
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,265
Default Re: Non-working Commodore PET 3016

Working back from UH2-2.

Scope UH3-3, this is the source for the input to UH2-1.

UH3-8 should be the same 16MHz clock thats feeding the counter thats generating the 1MHz for the processor. You’ll need to reduce the time per division on your scope to have any chance to measure this, probably to the fastest your scope can manage.

UH3-1 should be 1MHz square wave.

UH3-9 should be mostly high, but I haven’t been able to find the source of this signal in the schematics yet.
Mark1960 is offline  
Old 3rd Mar 2021, 6:54 pm   #780
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: Non-working Commodore PET 3016

Quote:
Originally Posted by SiriusHardware View Post
It might still be that the data buffers are not transmitting correct data back and forth between the unbuffered and buffered side of the data bus.
.
Maybe worth another comparative look at D0-D7 either side of the buffers one on each trace just to make sure it is clean now?
Timbucus is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 11:43 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.