Oric 1 and Atmos - repair help needed - EDIT: Oric 1 works

[YuT666]

I piggy backed the RAMs and there was no change.

So i'll go back to the 6502:



As i mentioned, RES & NMI are ok (both Orics) so what i should measure now? The A lines? Out?

[Symoon]

Sorry in advance for the newbie guess I'm offering, but who knows...
In which order do you plug the Oric power supply?
I remember my Oric-1 didn't boot if I plugged it first to the Oric itself, and then in the wall plug. It kept stuck at the RAM pattern, like yours.
I had then to unplug on the Oric side and plug it again.

[YuT666]

Thanks for your answer and don't worry ... i appreciate your help.

I tried different methods of plugging power supply, but the failures are still the same.

[YuT666]

Measured ULAs CLK and got this, so this should be ok.

[iss]

Great! Can you record now address lines A0..A15 just after powerup?
Maybe your LA has only 8 inputs, but you can do two recordings, for instance: F2 + A0..A6, then the rest.
After this we should know for sure if the 6502 is OK!

[YuT666]

Well, i don't know how the LA settings should be, so i can measure them. I can't get any results.

[iss]

ok, can you check (exactly like ULA's CLK) 6502's pin 39 (F2) and if there are no pulses then pin 37.

[YuT666]

I get pulses on both pins ...

[iss]

So, obviously the cpu generates F2 which is good .
Now you can attach chan0 to RESET - pin 40 and set it as trigger on rising edge,
chan1 to F2 - pin 39, and chan1..chan7 to address lines A0..A6.
Important is what happens after RESET goes to '1' i.e. just after you power-on.

[YuT666]

So, obviously the cpu generates F2 which is good .
Now you can attach chan0 to RESET - pin 40 and set it as trigger on rising edge,
chan1 to F2 - pin 39, and chan1..chan7 to address lines A0..A6.
Important is what happens after RESET goes to '1' i.e. just after you power-on.

It's bad, that i never worked with an LA before ... i have to get into it. So I'll do two measures ... 40 & 39 and A0-A6. Don't know what you mean ... "trigger on rising edge" ... i know what it is in german, but it is nearly my first time with an LA software.

@you and all other guys who help me: Thanks so much for your patience and endurance.

[Chema]

So I'll do two measures ... 40 & 39 and A0-A6. Don't know what you mean ... "trigger on rising edge" ... i know what it is in german, but it is nearly my first time with an LA software.

There must be some menu or config item called "Trigger" with different options. What iss means is that the capture starts when the signal goes from low to high (in the "edge").

The thing is, I think, to test if the expected addresses are being accessed by the CPU. At boot time the 6502 would fetch the contents of $FFFC/$FFFD, so I guess, so you should see the ines A0-A7 getting the values 11111100 and 11111101. The content is the address where the code for booting the computer is stored. In an ATMOS ROM (v1.1) that address is $F88F so the CPU should start getting the instructions at that address (lines a0-a7 should start 1000 1111 and increment from there on). The first few instructions are:

Code: Select all

F88F A2 FF      LDX #$FF 
F891 9A          TXS              ; Set stack pointer to #FF.
F892 58          CLI               ; Disable interrupts
F893 D8          CLD              ; Clear decimal mode.
F894 A2 12      LDX #$12      ; Copy the above jump table into
F896 BD 7C F8 LDA $F87C,X  ; page 2 of memory. 

so until $f894 no other memory is being accessed, but subsequent addresses from $f880 onwards (up to $F8998, where a read to $F87C is performed).

Basically if the first access to FFFC and FFFD is OK, the processor is trying to boot properly. If the subsequent address is wrong (higher lines up to A15 should be monitored to check) then there is a problem reading the ROM contents for some reason. If the processor then does not address from $f880 several consecutive addresses, it has hanged for some reason, probably due to a wrong value read from RAM, resulting in an invalid OP-code or a wrong instruction.

That is, of course, if I understood everything properly. If you had some experience with the soldering iron, it would be much easier to replace the CPU and test

BTW, I think there could be problems trying to read the memory even if the RAM chips were okay. There should be other chips involved and a cut track could be enough to have one wrong bit and a computer hang. This could be testing by also monitoring the data bus D0-D7 during boot.

With all the information of all the address and data lines, it would be easier to check what is going on.

[YuT666]

I bought some sockets and 6502 MOS, so i'll desolder the old cpus. After that we could ascertain, if the 6502 caused the errors. Thanks Chema.

[carlsson]

Good luck! I thought the 6502 was one of the more robust chips in old computers. It tends to be RAM, ROM, VIA and all sorts of other chips that break down before the CPU does (in particular in two different computers) but one never knows before you have tested and tried to replace it.

[Godzil]

Good luck! I thought the 6502 was one of the more robust chips in old computers. It tends to be RAM, ROM, VIA and all sorts of other chips that break down before the CPU does (in particular in two different computers) but one never knows before you have tested and tried to replace it.

The 6502 can die, one of my Oric have a dead one

[YuT666]

Ok, bought some UM6502 in Ireland ... i desoldered the old cpu of the Oric 1 and soldered a socket (my first 40 pin socket so far) ... put the first cpu in ...

... AND ...

Tried some keys and they works so far. I'll test the Oric 1 with a test ROM, so i hope, that the cpu was the only bad component ...