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 2020, 7:21 pm   #21
John Earland
Hexode
 
Join Date: May 2013
Location: Middlesbrough, North Yorkshire, UK.
Posts: 442
Default Re: Commodore PET 3016

Yes I will do that. I followed what the 8 but guy did too. I got some characters displayed from doing that-not the same ones as he did but I did get some, not all different though.
John Earland is offline  
Old 7th Feb 2020, 7:23 pm   #22
John Earland
Hexode
 
Join Date: May 2013
Location: Middlesbrough, North Yorkshire, UK.
Posts: 442
Default Re: Commodore PET 3016

Looking at your post again, I think I need to attempt it again in a more methodical way to be sure. However, if I get a random load of characters then it could be the decoder IC?
John Earland is offline  
Old 7th Feb 2020, 7:57 pm   #23
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Commodore PET 3016

Let's just recap on how the keyboard works.

Each keyboard key, when pressed, joins together a specific and unique combination of one row line and one column line. When the hardware detects that a particular row line and a particular column line are joined together, it looks up that row / column line combination in a table to determine what character it should display.

The problem appears to be that one or more of the column lines is being activated at the wrong time, most likely due to a hardware fault which could be a bad 74LS145 or just a bad or unwanted connection somewhere in that area.

The first thing to do is to rule in or rule out the keyboard itself as the cause.

What you need to do is pick just one row key pin on the main board connector, keep the end of your link wire on that pin and then momentarily touch the other end of your link wire to each of the column pins in order, in turn, recording which character appears on the screen each time.

When you've done that, move the 'fixed' end of the link wire to another row pin and touch the other end to each column pin again in order, again recording the characters which appear.

Repeat this for several of the row pins, and then post the results here.
SiriusHardware is offline  
Old 7th Feb 2020, 8:57 pm   #24
John Earland
Hexode
 
Join Date: May 2013
Location: Middlesbrough, North Yorkshire, UK.
Posts: 442
Default Re: Commodore PET 3016

Hi folks,
I have tried the test methodically now and I am getting characters being generated but not the range of characters seen in the video - certainly not a nice stream anyway - essentially many of the characters I get from using the keyboard so I am guessing it could be the IC as mentioned. I just don't fancy unsoldering it.

Any tips for doing so??

Best wishes,
John
John Earland is offline  
Old 7th Feb 2020, 9:25 pm   #25
John Earland
Hexode
 
Join Date: May 2013
Location: Middlesbrough, North Yorkshire, UK.
Posts: 442
Default Re: Commodore PET 3016

There are 10 rows: 1-10 and 8 columns ABCDEFHJ
Here are the results in a spreadsheet (attached)
Attached Files
File Type: xlsx Keyboard Results.xlsx (9.9 KB, 59 views)
John Earland is offline  
Old 7th Feb 2020, 9:26 pm   #26
John Earland
Hexode
 
Join Date: May 2013
Location: Middlesbrough, North Yorkshire, UK.
Posts: 442
Default Re: Commodore PET 3016

Quote:
Originally Posted by SiriusHardware View Post
Let's just recap on how the keyboard works.

Each keyboard key, when pressed, joins together a specific and unique combination of one row line and one column line. When the hardware detects that a particular row line and a particular column line are joined together, it looks up that row / column line combination in a table to determine what character it should display.

The problem appears to be that one or more of the column lines is being activated at the wrong time, most likely due to a hardware fault which could be a bad 74LS145 or just a bad or unwanted connection somewhere in that area.

The first thing to do is to rule in or rule out the keyboard itself as the cause.

What you need to do is pick just one row key pin on the main board connector, keep the end of your link wire on that pin and then momentarily touch the other end of your link wire to each of the column pins in order, in turn, recording which character appears on the screen each time.

When you've done that, move the 'fixed' end of the link wire to another row pin and touch the other end to each column pin again in order, again recording the characters which appear.

Repeat this for several of the row pins, and then post the results here.
Thanks - results posted in the message below in a spreadsheet
John Earland is offline  
Old 8th Feb 2020, 6:11 am   #27
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,188
Default Re: Commodore PET 3016

To remove a common IC like the 74LS154, I'd assume it was dead and cut the pins off close to the IC. Then remove the body, melt the solder on each pin in turn and pull it out. Clean out the holes by using the soldering iron on the bottom side of the board and the solder sucker on the component side.

Then fit a 24 pin turned pin socket and the new IC.
TonyDuell is offline  
Old 8th Feb 2020, 9:45 am   #28
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Commodore PET 3016

That's exactly the procedure I would advise as well once we are reasonably sure the part in question is dud AND there is a replacement present and ready to fit. Note that the IC in question is a 145 rather than a 154, though. We are just trying to establish whether the fault is definitely on the main board first.

John, sorry but I don't see your table of results, maybe you meant you were going to attach it in a separate post?
SiriusHardware is offline  
Old 8th Feb 2020, 11:02 am   #29
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,188
Default Re: Commodore PET 3016

Ooops! Th '145 is a 4 bit to 10 line open-collector output decoder, the '154 is a 4 bit to 16 line totem pole output decoder. It gets confusing..

So also replace my comment about a '24 pin turned-pin socket' with '16 pin turned-pin socket'
TonyDuell is offline  
Old 8th Feb 2020, 11:35 am   #30
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Commodore PET 3016

Ah, John's posts somehow arrived out of order, I see the post with the results now. About to go out for the day so I'll take a look tonight. Anyone else with ideas, please pitch in.
SiriusHardware is offline  
Old 8th Feb 2020, 11:53 am   #31
Refugee
Dekatron
 
Refugee's Avatar
 
Join Date: Apr 2012
Location: Worksop, Nottinghamshire, UK.
Posts: 5,549
Default Re: Commodore PET 3016

I did see the results and agree that posts #25 and #26 have indeed swapped places some how.
To me it looks like two lines are crossed either within a faulty chip or due to a tin whisker.
I would be tempted to take the board outside and give it a good scrub with a dry paint brush on both sides getting the brush under the chips as far as you can before changing the decoder chip.
Cutting the old chip is the best way to go about replacement.
Refugee is offline  
Old 8th Feb 2020, 9:26 pm   #32
John Earland
Hexode
 
Join Date: May 2013
Location: Middlesbrough, North Yorkshire, UK.
Posts: 442
Default Re: Commodore PET 3016

Thank you everyone for posting - you are wonderful people and a great help! I have ordered some rubbing alcohol and so will strip the board down as far as possible getting as much dirt out as possible - it is quite dirty.

I'm using the ROM/RAM replacement board from Tynemouth Software - it allows you to bypass the onboard ROM and RAM which means that the keyboard issue remains of course.

I have ordered a 145 - should be arriving sometime next week. I'll give the board a good clean first and then move onto the IC when it arrives! I may see if I can find someone who is good at soldering!

Thanks for all your help so far - I am so very grateful!
Best wishes,
John
John Earland is offline  
Old 9th Feb 2020, 12:12 am   #33
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Commodore PET 3016

John, thanks for recording the results so meticulously and clearly. What stands out is that in many instances, joining two different rows and columns generates the same character. That should not be happening.

What this appears to prove is that the keyboard itself is blameless because you get the same 'character twice' problem even when the keyboard is disconnected and you simulate the operation of every keyboard key using a wire link instead.

It therefore looks like a problem on the key scanning hardware on the main board, be it the 74LS145 or some problem with the associated tracks.

The following suggestions for measurements assume that the keyboard scanning circuit is as shown in the diagram linked to earlier in the thread. That may not actually be the case, but it's the best information we have at the moment. I'm also assuming you know how IC pins are numbered, but if not, just ask.

With your meter on ohms/resistance and with the PET powered OFF, try measuring between the following pairs of IC pins:

6520 pin 2 and 74LS145 pin 15
6520 pin 3 and 74LS145 pin 14
6520 pin 4 and 74LS145 pin 13
6520 pin 5 and 74LS145 pin 12

You should see very a low resistance (near zero ohms) between each of these pairs of pins. There's no need to record the exact resistances as long as each pair of pins is connected together with little or no resistance between the two pins.

If you do see those low resistances, that's great as it not only shows that the connections between the 6520 and the 74LS145 are good, it means that we probably have the correct circuit diagram as well.

If you do find the connections to be as above, then try some further measurements, this time keeping the black meter probe on 0V / Ground and measuring between

74LS145 pin 15 and 0V
74LS145 pin 14 and 0V
74LS145 pin 13 and 0V
74LS145 pin 12 and 0V

This time you will (or should) see a relatively high resistance from each of those four pins down to 0V. The actual value you see is not too important, what is important is that all four measurements from those four pins to 0V should be very similar. If you see one which is very noticeably different, that will suggest that the input in question is faulty or that there is an unwanted connection from that pin to something else.

Finally, try checking the resistance between

74LS145 pin 15 and 74LS145 pin 14
74LS145 pin 14 and 74LS145 pin 13
74LS145 pin 13 and 74LS145 pin 12

The resistances values you see in the above three checks are not critical but they must not show a very low resistance or short circuit.

Let's see what you get from the above checks first and then we'll take a look at the column drive lines on the output side of the 74LS145.

Last edited by SiriusHardware; 9th Feb 2020 at 12:18 am.
SiriusHardware is offline  
Old 9th Feb 2020, 11:28 am   #34
John Earland
Hexode
 
Join Date: May 2013
Location: Middlesbrough, North Yorkshire, UK.
Posts: 442
Default Re: Commodore PET 3016

This great! Thank you. I shall have a go this morning!


This is a photo of the board
Attached Thumbnails
Click image for larger version

Name:	2ABC0D84-F8AE-43D4-B6BF-C2AB530DD05F.jpg
Views:	92
Size:	135.6 KB
ID:	198649  
John Earland is offline  
Old 9th Feb 2020, 1:21 pm   #35
John Earland
Hexode
 
Join Date: May 2013
Location: Middlesbrough, North Yorkshire, UK.
Posts: 442
Default Re: Commodore PET 3016

Hi, I'm about to check the settings as you suggested but there are two 6520s - I shall try them both.
John Earland is offline  
Old 9th Feb 2020, 1:38 pm   #36
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Commodore PET 3016

Sorry, I was thinking only in terms of the 6520 which is associated with the keyboard. The one you are looking for is likely to be the one which is physically closest to the keyboard connector.

The overall PCB picture you posted has been crunched down to a smaller size and resolution by the forum engine so it isn't possible to read individual chip part numbers or component identification numbers from that.

The PET keyboard circuit I linked to earlier shows the 6520 as C7 (IC7?) and the 74LS145 as C9 (IC9?). Hopefully those component numbers also apply on your 3000 series main board.
SiriusHardware is offline  
Old 9th Feb 2020, 1:40 pm   #37
John Earland
Hexode
 
Join Date: May 2013
Location: Middlesbrough, North Yorkshire, UK.
Posts: 442
Default Re: Commodore PET 3016

Quote:
Originally Posted by John Earland View Post
This great! Thank you. I shall have a go this morning!
Hi again, I don't know where the ground is - can you help?

Here are the results from the first tests:


6520 (dot) pin 2 to 74LS145N pin 15 0 Ohms
6520 (dot) pin 3 to 74LS145N pin 14 0 Ohms
6520 (dot) pin 4 to 74LS145N pin 13 0 Ohms
6520 (dot) pin 5 to 74LS145N pin 12 0 Ohms


6520 (No dot) pin 2 to 74LS145N pin 15 007 Ohms
6520 (No dot) pin 3 to 74LS145N pin 14 007 Ohms
6520 (No dot) pin 4 to 74LS145N pin 13 007 Ohms
6520 (No dot) pin 5 to 74LS145N pin 12 007 Ohms


74LS145N pin 15 to 74LS145N pin 14 008 Ohms
74LS145N pin 14 to 74LS145N pin 13 008 Ohms
74LS145N pin 13 to 74LS145N pin 12 008 Ohms

Hope you can read those okay!
J

Last edited by John Earland; 9th Feb 2020 at 1:45 pm.
John Earland is offline  
Old 9th Feb 2020, 1:52 pm   #38
John Earland
Hexode
 
Join Date: May 2013
Location: Middlesbrough, North Yorkshire, UK.
Posts: 442
Default Re: Commodore PET 3016

These are the results from the ground test (I think I found the ground!!) Not sure though to be honest so any help would be gratefully received. I've attached what I think the 3016 schematic is too. I have used the GROUND which is labelled GND on the top(ish) left hand corner labelled VR6 but that might not be the one I should be using!

74LS145N 15 to 0V 004
74LS145N 14 to 0V 004
74LS145N 13 to 0V 004
74LS145N 12 to 0V 004
Attached Thumbnails
Click image for larger version

Name:	3016.jpg
Views:	55
Size:	6.6 KB
ID:	198662  

Last edited by John Earland; 9th Feb 2020 at 1:57 pm.
John Earland is offline  
Old 9th Feb 2020, 1:54 pm   #39
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Commodore PET 3016

It looks like the 6520 (dot) is the one associated with keyboard scanning.

For 0V, referring to your overall PCB picture, do you see the thick track running down the left hand side of the PCB, along the bottom edge and up the right hand edge? That will be 0V, so you can hold the black probe anywhere on that area when making the second set of measurements. The -ve end of the large capacitor with the cable tie around it is also likely to be a 0V point.
SiriusHardware is offline  
Old 9th Feb 2020, 1:56 pm   #40
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Commodore PET 3016

Unfortunately the actual picture of the layout which you attached is very small, probably a thumbnail rather than the original image.

Perhaps you could provide a link to where you found it instead?
SiriusHardware is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 10:28 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.