23rd Feb 2021, 4:58 pm | #621 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,553
|
Re: Non-working Commodore PET 3016
With those three main devices removed the only components still left on the unbuffered data bus (apart from the data bus buffers) are the PROMs, so once again, with UC5,6,7 still out, can you remove combinations of UD6-UD9 to see if any of those allow the CPU to run when removed?
The EPROM which John programmed with the UD7 code for you, what sort of device was it in the end? Last edited by SiriusHardware; 23rd Feb 2021 at 5:04 pm. |
23rd Feb 2021, 4:59 pm | #622 |
Nonode
Join Date: Jan 2009
Location: Bristol, UK.
Posts: 2,061
|
Re: Non-working Commodore PET 3016
A little OT, but I managed to remove the top from a TMS25P32. It certainly looks like an EPROM inside.
Was it really cheaper to make with a lid than with a window or is it just an anti-recycling / anti re-flashing measure ? dc |
23rd Feb 2021, 5:03 pm | #623 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,553
|
Re: Non-working Commodore PET 3016
Maybe those windows were a bit more expensive than the plain lid (the material had to be UV-transparent, I guess, so possibly not ordinary glass).
|
23rd Feb 2021, 5:06 pm | #624 | |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,803
|
Re: Non-working Commodore PET 3016
On it. The EPROM is marked as TMS 2532JL-45
Colin. Quote:
|
|
23rd Feb 2021, 5:33 pm | #625 | |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,803
|
Re: Non-working Commodore PET 3016
With UC5, 6 and 7 OUT, I get a signal for a second or so with UD6 & UD7 OUT. It then flat-lines.
However... with UC5, 6 and 7 IN I only get continuous readings from Pin 9 when I take UD6 & 7 (and also with 8 and/or 9 out) out of their sockets. So that seems like some combination of UD6 and 7 stopping Pin 9. Colin. Quote:
|
|
23rd Feb 2021, 5:41 pm | #626 |
Nonode
Join Date: Jan 2009
Location: Bristol, UK.
Posts: 2,061
|
Re: Non-working Commodore PET 3016
Indeed, usually silica glass is used if you need UV transparent. I don't recall it being super expensive, unless you are comparing it to window glass!
Good to see Colin is making progress, even if there are still some remaining gremlins! dc |
23rd Feb 2021, 6:04 pm | #627 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,553
|
Re: Non-working Commodore PET 3016
We know that UD7 should be good and I'm confident that UD6 was too.
If we assume they are, then there are some possibilities - -Not all data lines of UD6 / UD7 making it back to the CPU due to a bad socket pin or some other lost connection between the two. -UD6 / UD7 not being selected when they should be due to an address decoding fault or an address line fault on the way to the address decoder -Missing address line on the way to the PROMs -UD6 / UD7 both being selected at the same time (clashing) due to similar problems to above. -Faulty data bus buffers loading the bus just enough to upset it when certain other devices are also connected. Time to go home. Later this evening we'll have a look at what the chip select lines on those two PROMs are doing, if anything. |
23rd Feb 2021, 6:24 pm | #628 |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,803
|
Re: Non-working Commodore PET 3016
Just as a keen amateur, I looked at the schematics and tested continuity of UD2 pins 14/15/16/17-> UD6/7/8/9 Pin 20 and they all check out OK.
Colin. |
23rd Feb 2021, 8:04 pm | #629 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,553
|
Re: Non-working Commodore PET 3016
(SEL C, D, E, F, the chip selects for UD6-UD9), that's fine
To check the data bus lines between all of the PROMs and the CPU, How about continuity between:- Code:
CPU UD6-UD9 pins 26 9 27 10 28 11 29 13 *note the jump 11-13 30 14 31 15 32 16 33 17 Code:
UC3 UD6-UD9 pins 18 8 3 7 16 6 5 5 14 4 7 3 12 2 9 1 UB3 UD6-UD9 pins 18 23 3 22 16 19 5 18 |
23rd Feb 2021, 9:01 pm | #630 | |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,803
|
Re: Non-working Commodore PET 3016
So my reading of the schematics is different to yours. i think (please tell me if I have it wrong) that the first set of pins should be:
Code:
CPU UD6-UD9 26 17 27 16 28 15 29 14 30 13 31 11 32 10 33 9 UC3 continuity as below all checks out fine with one slight outlier - the resistance on pin5 of UD8 seems slightly high at 2.1Ω UB3 continuity as below all checks out fine. As a check, I believe that Pins 5 on UD6-9 should all connect and I get continuity checks Ok for that with again a higher resistance on pin5 of UD8 (1.9Ω). Colin. Quote:
|
|
23rd Feb 2021, 10:10 pm | #631 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,553
|
Re: Non-working Commodore PET 3016
Yes, I managed to get the data lines in reverse order. . In spite of which you managed to measure them OK.
Next up then, do you have any activity on your PROM chip selects? Pin 20 of UD6-UD9. What normally happens is that the CPU outputs an address, the high part of which goes to the address decoder. The address decoder in turn selects the device which is mapped into the address range that the address falls within. The problem we have here is that in order to make the CPU run and put addresses on the address bus, we have to disconnect or partly disconnect it from the databus. This in turn means that the CPU is just running out of control, executing random data and jumping back and forth. Because of this, we can't be certain that every possible address is being placed on the bus and so we can't be sure that we can expect to see every device being selected. So have a look at the chip selects for UD6-UD9 and see if there is activity there. Apart from possibly replacing the databus buffers as a 'try it and see, we should probably do it anyway' the next thing may be to try a NOP test, unless anyone has any intermediate suggestions? |
23rd Feb 2021, 10:34 pm | #632 |
Octode
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,294
|
Re: Non-working Commodore PET 3016
I think setting up a NOP test would be a good next step. As Colin already has the databus disconnected from the 6502 it just needs resistors added to pull the data lines up or down to select the NOP instruction code. EA hex. I know some people tie the data lines to 5v or 0v but I think its safer to use resistors. Around 4.7k but not critical. D7, D6, D5, D3, D1 via resistors to 5v, D4, D2, D0 via resistors to 0v. Separate resistor for each data pin.
This should give nice regular pattern on the address lines and also repeating pattern on the address decoders chip select to the proms. It might also be possible to see valid logic levels on the data bus while the prom chip selects are active. |
23rd Feb 2021, 10:37 pm | #633 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,553
|
Re: Non-working Commodore PET 3016
I should have said: Look for activity on the PROM selects with the CPU databus disconnected from the system databus, so that the CPU runs continually.
|
23rd Feb 2021, 10:38 pm | #634 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,553
|
Re: Non-working Commodore PET 3016
|
23rd Feb 2021, 10:48 pm | #635 |
Octode
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,363
|
Re: Non-working Commodore PET 3016
If the pullups are the other side of the buffers and it works it is worth testing before more desoldering.
|
23rd Feb 2021, 10:52 pm | #636 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,553
|
Re: Non-working Commodore PET 3016
Pullups? You mean the resistors to impose the NOP? (Some are pull up, some pull down).
|
23rd Feb 2021, 11:36 pm | #637 |
Octode
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,803
|
Re: Non-working Commodore PET 3016
I get the attached from all of UD6-9 when the scope is set to 1X and pins 26-33 are removed from the 6502 socket.
https://drive.google.com/file/d/1_rR...ew?usp=sharing Colin. |
23rd Feb 2021, 11:45 pm | #638 |
Dekatron
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,553
|
Re: Non-working Commodore PET 3016
They certainly look like chip select pulses so it would seem that even in its semi disabled state the system is managing to sweep through the address ranges of the PROMs. Just one other thing to try while you are in the area: Scope any two of the PROM pin 20s at the same time. You should see that they are selected at the same rate / frequency but at different times, so the chip select pulses will be staggered, one before the other, showing that only one PROM is selected at a time. If any chip select pulses are directly one above the other that would show two devices being selected at the same time (which should not happen).
|
23rd Feb 2021, 11:58 pm | #639 | |
Octode
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,294
|
Re: Non-working Commodore PET 3016
Quote:
Running NOP should move the chip select pulses closer together as every cycle would be a read. Also there should be a burst of chip select on one prom, followed by a burst from the next. |
|
24th Feb 2021, 12:01 am | #640 |
Octode
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,294
|
Re: Non-working Commodore PET 3016
I’m also not sure if Tim means the resistors to put NOP on the 6502, but just want to note that those resistors would not be connected to the rest of the databus as the 6502 data bus should be disconnected from the socket on the pcb.
|