Thanks ttl, a friend has a "PIC Microchip programmer" which I'm not sure will work, but I will borrow it and see what I can see.

I'm not sure at this point what the fault is - two of the ROM chips are pretty cool to the touch, while the others are roasting hot. I'm taking it for granted that those two cool ones aren't working. I'll keep taking advice though. And given the number of dead RAM chips, if the ROMs are faulty then I'm thinking it's probably best to buy the board from Tynemouth Software that goes into the CPU slot and just pop the remaining good chips to the side.

Rather than a parallel programming interface like EPROMs use, PICs use synchronous-serial programming interfaces that work a little bit like I²C, or like the PS/2 keyboard interface.

When I design PIC microcontrollers into our products, I always put a row of six holes on the board that a programming fixture with Pogo pins lines up with, so that if necessary, the software can be updated long after the board is assembled, without requiring unsoldering the surface-mount microcontroller. It's kind of neat how they made it so there are only two primary pins involved in programming (a data line and a clock line), and the end product can usually use these for something like pushbutton switches which leave the pin disconnected (except for a weak pull-up resistor) when the button is not being pressed. So why six holes for the programming? Data, clock, Vcc, master clear / Vpp, ground, and a sixth one to short the PIC's OSC1 pin to ground during programming so the onboard oscillator doesn't try to oscillate. I initially used only five, but ran into some kind of problem 20+ years ago which was solved by grounding the OSC1 pin. I don't remember details anymore.

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?

It might be simpler to make a EEPROM and have all the ROMs on one but we are seeing a broken reset signal now. My guess it the 555 but it could be the LS04.
We've had a number of bad chips what a pain.
The vertical is 60Hz. It might show up as a little fuzzing of the top or bottom line and a wavy horizontal. It still should work.
Dwight

Latest is that the 555 timer was keeping the CPU in reset. I replaced the 555, but the fault remained. Eventually discovered that the large ceramic capacitor on pin 2 is short circuit and not charging. I snipped a leg off, connected it temporarily at powerup and then disconnected it (forcing pin 2 high), and the CPU started working. Tape machine stopped, display went blank. Currently waiting for a replacement cap.

Quick update, it seems that ROMs H5 and H7 aren't working so I got a Tynemouth board in place of the CPU. Managed to get a BASIC prompt and then discovered that the 6522 was holding the CPU's IRQ high, so I removed that and now I have a cursor and can type. And 9 bad RAM chips by the look of things. So the computer is now running on a replacement ROM and RAM set (on the Tynemouth board).

Next test is to use the board to confirm that those ROMs are bad before I consider replacing them with EPROMs. And I also have to find out why the 6522 isn't working correctly, could be a faulty chip or it could be something else. But the PET is now almost working.