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 7th Feb 2021, 6:03 pm   #521
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Non-working Commodore PET 3016

You seemed to get very close to successfully reading them, thwarted only by unreliable connections?

Let me know in PM when they are on the way and I will keep an eye out for them.
SiriusHardware is online now  
Old 7th Feb 2021, 8:28 pm   #522
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

I got frustratingly close, but even with new jumpers, the data was changing.

I'll try the pi tomorrow with the spare ROM. Thanks for your work on that.

Colin.
ScottishColin is offline  
Old 7th Feb 2021, 8:47 pm   #523
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Non-working Commodore PET 3016

If you feel your weak link was the connections then I fear you may have the same problems with that, but, nothing ventured, nothing gained. My level shifters have arrived as well, so I'll put some of those together tomorrow, work allowing.

I'll read the ICs using my device reader / programmer first but if I have time afterwards I'll also try recreating your efforts with the Mega to see what needs to be done to make that work. (Definitely should work).

Your having the 'Mega' offers the possibility of another technique which is to use it as a diagnostic replacement for the CPU (essentially, wire its pins to a header which can be plugged into the CPU socket, you can then use it to run all sorts of tests by generating fixed signal levels or doing repeated reads from or writes to single addresses or generating ascending addresses... (much like a NOP test)). Probably all a bit too advanced just now, but it is all there as a future line of attack.
SiriusHardware is online now  
Old 7th Feb 2021, 9:07 pm   #524
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

Quote:
Originally Posted by SiriusHardware View Post
If you feel your weak link was the connections then I fear you may have the same problems with that, but, nothing ventured, nothing gained. My level shifters have arrived as well, so I'll put some of those together tomorrow, work allowing.

I'll read the ICs using my device reader / programmer first but if I have time afterwards I'll also try recreating your efforts with the Mega to see what needs to be done to make that work. (Definitely should work).

Your having the 'Mega' offers the possibility of another technique which is to use it as a diagnostic replacement for the CPU (essentially, wire its pins to a header which can be plugged into the CPU socket, you can then use it to run all sorts of tests by generating fixed signal levels or doing repeated reads from or writes to single addresses or generating ascending addresses... (much like a NOP test)). Probably all a bit too advanced just now, but it is all there as a future line of attack.
That sounds interesting. I'm learning so much with your help - thanks very much.

Colin.
ScottishColin is offline  
Old 7th Feb 2021, 9:16 pm   #525
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

Quote:
Originally Posted by SiriusHardware View Post
If you feel your weak link was the connections then I fear you may have the same problems with that, but, nothing ventured, nothing gained. My level shifters have arrived as well, so I'll put some of those together tomorrow, work allowing.

I'll read the ICs using my device reader / programmer first but if I have time afterwards I'll also try recreating your efforts with the Mega to see what needs to be done to make that work. (Definitely should work).

Your having the 'Mega' offers the possibility of another technique which is to use it as a diagnostic replacement for the CPU (essentially, wire its pins to a header which can be plugged into the CPU socket, you can then use it to run all sorts of tests by generating fixed signal levels or doing repeated reads from or writes to single addresses or generating ascending addresses... (much like a NOP test)). Probably all a bit too advanced just now, but it is all there as a future line of attack.
I've ordered some of these to see if they will help - full length pins each side of the connector:

https://www.amazon.co.uk/Gikfun-2-54...tronics&sr=1-3

Colin.
ScottishColin is offline  
Old 8th Feb 2021, 2:02 pm   #526
NottsIan
Tetrode
 
Join Date: Mar 2018
Location: Nottingham, Nottinghamshire, UK.
Posts: 88
Default Re: Non-working Commodore PET 3016

You might also find the pins and housings on this page https://www.hobbytronics.co.uk/cable...imp-connectors useful. The crimp tool is a bit pricey but comes in useful for so many things.
NottsIan is offline  
Old 8th Feb 2021, 2:30 pm   #527
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

Quote:
Originally Posted by NottsIan View Post
You might also find the pins and housings on this page https://www.hobbytronics.co.uk/cable...imp-connectors useful. The crimp tool is a bit pricey but comes in useful for so many things.
Thanks - I'll take a look.

Colin.
ScottishColin is offline  
Old 8th Feb 2021, 2:35 pm   #528
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

I still have UF10 here (the character ROM). Am I right in assuming I can apply the same principles to that ROM chip for address lines and data lines?

If so, do I need to earth pins 12, 18 and 20 or just one of them?

Colin.
ScottishColin is offline  
Old 8th Feb 2021, 4:22 pm   #529
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Non-working Commodore PET 3016

On that one, chip select pins 18, 20, go to 0V and pin 21 to +5V. Power supply pins are pin 12=0V and pin 24=+5V.

As pin 18 is a chip select rather than A11 on that IC, that suggests it is a 2K ROM, so reduce the max read address to 2048 and don't connect the A11 output from the Mega / Pi to pin 18 on the PROM.
SiriusHardware is online now  
Old 8th Feb 2021, 5:09 pm   #530
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

Quote:
Originally Posted by SiriusHardware View Post
On that one, chip select pins 18, 20, go to 0V and pin 21 to +5V. Power supply pins are pin 12=0V and pin 24=+5V.

As pin 18 is a chip select rather than A11 on that IC, that suggests it is a 2K ROM, so reduce the max read address to 2048 and don't connect the A11 output from the Mega / Pi to pin 18 on the PROM.
Apologies for the daft question. How can I tell from the schematic that pin 21 is +5V?

And how can I tell from the schematic that 12,18 and 20 are different? They look connected to my untutored eyes and look like they all go to Ground...

Colin.
Attached Thumbnails
Click image for larger version

Name:	F10.PNG
Views:	42
Size:	11.7 KB
ID:	226286  
ScottishColin is offline  
Old 8th Feb 2021, 5:17 pm   #531
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Non-working Commodore PET 3016

As noted before, there are no daft questions.

The datasheet for that series of PROMs says that the power pins are 24 and 12 and also that pins 18, 20 and 21 are CS (chip select) pins. On the 4K chips only 20 and 21 are CS pins, pin 18 is A11.

The diagram you've posted shows 18 and 20 connected to 0V so that's obviously how they have to be connected in order for the chip to be in read mode. Pin 21 is connected to a line (_INIT) which is normally high ~+5V, therefore pin 21 needs to be held high (tied to +5V or pulled up to +5V with a resistor) when you are trying to read the chip.

So 18, 20 = low and 21 = high in order to put the chip into read mode.
SiriusHardware is online now  
Old 8th Feb 2021, 5:28 pm   #532
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

Thank you. I'll Keep that for future reference for other chips.

Colin.
ScottishColin is offline  
Old 8th Feb 2021, 7:50 pm   #533
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Non-working Commodore PET 3016

To be more specific about pin 21, it isn't totally obvious from the circuit in the immediate vicinity of the PROM what state the _INIT line (pin 21) will be in, but there are clues.

If you follow that _INIT line around the rest of the circuit sheets you'll see that it goes to a lot of chip pins which are active-low clear or reset inputs on the chips in question - (that is, when the _INIT line is in the low (0V) state all those ICs are cleared, or held, or reset). It wouldn't make any sense for all those ICs to be held or reset while the system is running normally, therefore we can deduce that the usual state of the _INIT line is high, or somewhere near +5V. The _INIT line most likely goes low for a short time after power-up or reset, then returns high and stays high.

The way the _INIT line is labelled, with a bar drawn above the letters, means that the line performs its named function when it is taken low, rather than when it is taken high.

You could verify that by just measuring the voltage on pin 21 of that PROM when the system is powered - you should find that it is high / logic 1. If so, your reader also needs to hold that PROM pin high when it is trying to read the PROM.
SiriusHardware is online now  
Old 8th Feb 2021, 7:54 pm   #534
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

Quote:
Originally Posted by SiriusHardware View Post
To be more specific about pin 21, it isn't totally obvious from the circuit in the immediate vicinity of the PROM what state the _INIT line (pin 21) will be in, but there are clues.

If you follow that _INIT line around the rest of the circuit sheets you'll see that it goes to a lot of chip pins which are active-low clear or reset inputs on the chips in question - (that is, when the _INIT line is in the low (0V) state all those ICs are cleared, or held, or reset). It wouldn't make any sense for all those ICs to be held or reset while the system is running normally, therefore we can deduce that the usual state of the _INIT line is high, or somewhere near +5V. The _INIT line most likely goes low for a short time after power-up or reset, then returns high and stays high.

The way the _INIT line is labelled, with a bar drawn above the letters, means that the line performs its named function when it is taken low, rather than when it is taken high.

You could verify that by just measuring the voltage on pin 21 of that PROM when the system is powered - you should find that it is high / logic 1. If so, your reader also needs to hold that PROM pin high when it is trying to read the PROM.
Thanks.

So before I plug this all into the Arduino, I am making (perhaps erroneously) the following assumptions:

Address lines and data lines will be plugged into the appropriate pins of the Arduino that the code refers to.

I plug ground from the Arduino to the following:
12 18 and 20

I plug 5V from the Arduino to the following:
21 and 24

Have I got that right?

If I don't need to use pins 24 and 12, could you explain why not? My assumptions are based on the pins in the schematics which show 12 and 24 connected.

Thanks.

Colin.
ScottishColin is offline  
Old 8th Feb 2021, 8:04 pm   #535
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Non-working Commodore PET 3016

from post #531

Quote:
The datasheet for that series of PROMs says that the power pins are 24 and 12
You MUST connect pins 24 and 12, these are the power pins of the PROM. Arduino +5V to pin 24, Arduino GND / 0V to pin 12. The PROM needs to be correctly powered before it can do anything sensible.

The required states of the chip select pins to enable the chip is deduced partly by looking at how they are connected in the circuit diagram, and partly by looking at the datasheet for the PROMs.

In the case of the character PROM which appears to be a 2K device, don't connect the Arduino's A11 address output to anything, that pin is only connected when reading 4K PROMs.
SiriusHardware is online now  
Old 8th Feb 2021, 8:46 pm   #536
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Non-working Commodore PET 3016

The content of your character PROM should probably coincide with the

characters-2.901447-10.bin2009-08-18

file on this page under 'Files':

http://www.zimmers.net/anonftp/pub/c...pet/index.html

The first few lines should match the attached image.
Attached Thumbnails
Click image for larger version

Name:	CharacterProm_Start.jpg
Views:	49
Size:	139.4 KB
ID:	226312  
SiriusHardware is online now  
Old 8th Feb 2021, 10:40 pm   #537
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

And it so nearly does, and yet it doesn't....

Colin.
Attached Thumbnails
Click image for larger version

Name:	UF10 ASCII.PNG
Views:	36
Size:	27.1 KB
ID:	226329  
ScottishColin is offline  
Old 8th Feb 2021, 10:44 pm   #538
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Non-working Commodore PET 3016

Does your UF10 PROM have a xxxxxx-yy number written on it?

Edit - Ah, found it, your post# 48

Quote:
UF10 - 901447-10
So yes, that should be an exact match for the online file. Your image in post #537 is obviously a screenshot from HxD. How did you get the code out from the reader into a file for Hxd to read?

Last edited by SiriusHardware; 8th Feb 2021 at 10:52 pm.
SiriusHardware is online now  
Old 8th Feb 2021, 10:50 pm   #539
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

901447-10

Would you be able to run your eye over the code I'm using? I'm not convinced by it; it wanted to ignore anything below ASCII 32 and above 127 so I have changed that, bit it still doesn't convince me....

Colin.
Attached Files
File Type: txt PETUF10 - Copy.txt (3.9 KB, 32 views)
ScottishColin is offline  
Old 8th Feb 2021, 11:08 pm   #540
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

Quote:
Originally Posted by SiriusHardware View Post
Does your UF10 PROM have a xxxxxx-yy number written on it?

Edit - Ah, found it, your post# 48

Quote:
UF10 - 901447-10
So yes, that should be an exact match for the online file. Your image in post #537 is obviously a screenshot from HxD. How did you get the code out from the reader into a file for Hxd to read?
I open the serial monitor and watch it run, then cut and paste.
ScottishColin is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 10:56 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.