Interactions with 8085 CPU board

[Alan Cox]

I'm seeing very weird interactions between the ancient computing 8085 board and the Z80 CTC card.The mere presence of the CTC on the bus appears sufficient to make the boot time RAM test sometimes fail, and everything to die shortly after.

I've ruled out a bad CTC card (it works fine on my Z80 system).

I did wonder about the interrupt handling but firstly I am not generating any and secondly M1 is wired high so it should never see M1 low and IORQ to respond.

Right now the CTC has the usual 7.3MHz crystal but that is not being provided to CLK or CLK2 - just some counter inputs (CLK2 isn't in use CLK is coming off the 80C85 so will be 3.072MHz)

Any ideas ?

Alan

[Steve Cousins]

I don't know the 8085, so these are just wild ideas.

Does it do the same with the CTC set to a different address.

Does the system work okay with, say, a PIO instead of the CTC, perhaps set to the CTC's address.

Does it make any difference removing the CTC modules oscillator signal from the CTC inputs.

Actually, I can't think of any scenario that accounts for the problem so these really are just complete guesses.

Steve

[Mark T]

Are you performing any reads from the CTC? The CTC uses the lack of /RD in combination with /IORQ to generate an internal write signal, possibly using the CLK signal to drive a state machine for when the /RD signal is expected. If the phase of the 8085 clock relative to /IORQ is different from Z80 then maybe its generating writes to the ctc and setting up some unintended interupt mode.

Did you check the state of the interupt pin?

Mark

[Bill Shen]

Alan,
Have you tried different slots on the backplane? Try with a small backplane. If you can control the 5V voltage, try raise or lower the voltage by +/- 10%.
Bill

[Alan Cox]

I tried playing with the backplane and power - no difference.

Address doesn't seem to matter. CTC inputs don't seem to matter. Swapping the CTC for a PIO produces a similar but not identical crash but putting an 82C55 (PPIDE) card in that slot is fine.

/IORQ is generated by a 74HCT14 inverting the IO/M line from the CPU (the M line going directly to MREQ) so presumably there is 6ns where you get both or neither.

The cycles are in https://www.jameco.com/jameco/products/prodds/51705oki.pdf page 10. For the 8085 board it demuxes AD0-7 using a '573.

Alan

[Alan Cox]

Not found an answer yet, but did find a great book for nailing everything else except Z80 stuff together 8)

http://bitsavers.trailing-edge.com/components/amd/Peripheral_Processor_Interface_Guide_1985.pdf

[Phillip Stevens]

Alan Cox wrote:

Not found an answer yet, but did find a great book for nailing everything else except Z80 stuff together 8)

http://bitsavers.trailing-edge.com/components/amd/Peripheral_Processor_Interface_Guide_1985.pdf

Ah. My daily reference guide when I was nailing the Am9511A-1 to the Z8S180.

Gave lots of ideas and help. Pity it didn't go into the Z180 quirks which made things so difficult.

What did someone say about "unknown unknowns"? Yes, that.

[jopil]

Allan hello.
From my dropbox, for 2 days only, the following will be available:
https://www.dropbox.com/s/hxj51yee6y153wz/MICROPROCESSOR_INTERFACING_TECHNIQUES_3rd_1979_RODNAY-ZAKS.pdf?dl=0
I hope it'll answer all of your queries, (Wait about an hour from now, 16:41, before clicking to allow for the upload)
Regards,
John

[jopil]

This,  http://www-clips.imag.fr/projet-systeme/Z85230/chapter6_2.pdf  may also help on some Z180 interfacing issues along with the ZILOG's free on line AppNote AN971800200.
John

[jopil]

Also, on 8085 specifically,
By HEATHKIT, 8085 Microprocessor Interfacing and Applications:
https://archive.org/details/8085MicroprocessorInterfacingApplications/page/n17
and
https://www.dropbox.com/s/5i9ad4qsmy9phk8/8085_MICROPROCESSOR_INTERFACING.pdf?dl=0
https://www.dropbox.com/s/8iwy4pm9hwqpjyu/HARDWARE_INTERFACING_WITH_INTEL_8085.pdf?dl=0
and
intel free online databooks of μPs & Peripherals Interfacing of 1980/1985
John