Bought a PET - Not working ^^

[Commodore_PET]

So could it be that im looking at the wrong 74LS10 Chip? Instead of in socket UG7, theres a 74LS10 Chip in socket UA5? And to that 74LS10 chip, i do have continuity to 7417. I found it by tracing any path from 7417 to where ever it goes.
If the one in UG7 is the one that has to be connected, i can now pretty confidently say that there is no original path on the board that would go from 7417 to the 74LS10 in UG7.

Just a general question, are you guys confident that one of those PET diagnostic boards would help me now? Not at all that im giving up, but if there are purpose built tools for the job, i might as well use them.

Emmanuel

[Commodore_PET]

So i think something some where went terribly wrong. Im now gonna go through every step of this thread and note down my results. I just have the feeling ive done some things wrong, and maybe reported some wrong information.

1: I have measured Voltages on the board to see if I have the necessary voltages present.
I measured for example the RAM chips to find Pin 1 with -5v, Pin 8 with 12v, Pin 9 with 5v

2: I have then measured the Voltages of Phi0, Phi1, and Ph2 on the 6502. Which I reported wrong, because I have not set the multimeter to AC. here are the correct readings: On the 6502 and on AC range 200 i have following numbers:
Phi 0 has 2.7V (instead of reported 1.6V)
Phi 1 has 4.2V (instead of reported 2.35V)
Phi 2 has 4.3V (instead of reported 2.3V)

3: Next, we have done DC readings in the 6502
Pin 40: 5.14 V
Pin 4: 0.12V
Pin 6: 5.13 V
Pin 7: 0.93 V
These numbers still hold up now after checking them a second time.
Here we have established that IRQ is low and should be high.

4: I have tested the resistance of resistor in R17 and it measures 3.3 kOhm

5: I have measured the resistance between Pin 1 and Pin 4 of the 6502 and measured 3.47kOhm

6: Turning the machine on, Measuring Voltage of resistor in R17 i have 5.14V on one end, 0.12 V

7: I removed all 65XX chips from their socket and and powered up the board.
The IRQ Pin in the socket of the 6502 is now at 5.12V. Inserting all but UC7 gives 5.12V on IRQ of 6502.

8: on the affected socket UC7, i have monitored RESET and have observed a solid second of it being low, before going high

9: Here again something i reported wrong: On the affected socket Phi2 is showing 3.9V with the multimeter set to AC and 200 range. I first reposted 4.2 and later 0.4. I cannot remember why.

10: Now on AC setting i can find at least 3.8V on everything that should be connected to B-Phi2 that i could have gathered. The 74LS10, the 74LS00 and the 7417.

So i feel like i led everyone on a wild goose chase because i failed to read my multimeter right and we have to look somewhere else now. Still socket UC7 but not B-Phi2.

Emmanuel

[BigEd]

Does your multimeter have a frequency mode? That can be handy.

[Commodore_PET]

No it does not, but my logic probe should finally arrive on monday.

Emmanuel

[Dr Jefyll]

Commodore_PET wrote:

7: I removed all 65XX chips from their socket and and powered up the board.
The IRQ Pin in the socket of the 6502 is now at 5.12V. Inserting all but UC7 gives 5.12V on IRQ of 6502.

Details like this represent real progress, so don't worry too much about minor booboos in other areas.

Quote:

Just a general question, are you guys confident that one of those PET diagnostic boards would help me now? Not at all that im giving up, but if there are purpose built tools for the job, i might as well use them.

What sort of capabilities do these boards offer? Can you provide a link, please?

-- Jeff

[Commodore_PET]

https://www.tindie.com/products/tynemou ... cs-module/
This is the diagnostic board i was shown earlier.

when im ordering something from the guy, i was wondering if this is a "Good to have" because many people on youtube who have these machines, have one of these too:
https://www.tindie.com/products/tynemou ... odore-pet/

Its just minimum of 30$ for a shipment into Switzerland, i could save 30$ by ordering both at once.

Emmanuel

[BigEd]

That looks to me like a relatively small outlay which could save a lot of time. (It's from Dave Curran, mentioned upthread.) So, if you value the time more than the money, go for it!

[Commodore_PET]

Alright, the guy who sells it is on vacation till June 7th and is not taking orders, but i signed up for the mailing list to get reminded when he returns. I will then order it.

Emmanuel

[1024MAK]

I have ordered items from Dave Curran before. His customer service is very good. Once, in a kit I received from him, an incorrect connector was supplied, he sent the correct part immediately, it arrived a day later.

I have since ordered other items, so although I don't have a PET and hence don't have a need for the PET ROM/RAM board or the PET Diagnostics Module, I would fully expect one of these to help. However, like any tool, it can only do so much, so there is no guaranty that these will quickly point you to the fault.

When fault finding, there are a multitude of different techniques using various different tools. One slightly cheaper way would be to get a logic analyser.

His products are not cheap, but you can see from his blog and the quality of his products how much work and effort he puts into them.

I do recommend you read all his posts on his blog where he investigates faulty PETs.

His blog is here: http://blog.tynemouthsoftware.co.uk.

Due to the lockdown and social distancing in the U.K., postal and courier services are being affected.

I also suggest that before jumping ahead, you may wish to wait and see how far using your logic probe takes you.

Mark

[Commodore_PET]

Yeah i have been reading his blog quite a bit looking for a post with a similar issue. Its an interesting read for sure.

Also public holiday here so the Postal service has not delivered my logic probe.

I a bit lost at what i should be doing right now.. Wait for the Logic porbe and then check B-Phi 2 (Pin 37) on the affected socket with the chip removed and report the results?

Emmanuel

[Commodore_PET]

Hello Guys, The saga continues!

My logic probe has arrived. After watching a few youtube videos on how to use it, i set out to check if the system has a clock.

i go check the clock on the affected socket UC7, B-Phi2 on pin 25. it should be "oscillating". I get both LED's lighting up, indicating a pulse higher than 50Hz.
I have checked B-Phi 2 all 65XX chips and get the same results.

one interesting thing, when i remove the 6502 chip and start the machine with fully empty slots, it now shows a screen full of random characters. It has never done this before and its doing it consistently now. So i investigated a bit further. Removing the chip in UC7 and UC4 will result in the screen being on with random characters. Placing the 6502 back into UC4 and starting the machine creates a blank screen but leaves IRQ high on the 6502 high. We knew that but i thought i repeat it now with this new screen phenomenon. So there is still something about the UC7 socket that creates a low IRQ once the chip is in the socket.

I have seen a couple of posts about this random character issue so this gives me further reading material.


Emmanuel

(picture of the screen)

[Risc]

The reason you’re seeing garbled characters on startup without the CPU is most likely because nothing (a.k.a the CPU) have been able to clear screen RAM yet, so the memory just holds trash. But you can cross the fly back transformer off the list now

You mentioned hot ROM. If you got a programmer available, try to dump the contents to check if the chip is still good. That’s an easy test if it’s socketed.