Nominal video output signal on ULA pin 16 ZX81 rev 1?

[AMDeee]

Given that the /ROMCS signal is generated by the ULA, what conditions have to be met on the ULA for that pin to get stuck HIGH apart from the ULA being faulty itself?

[1024MAK]

Given in the topic I linked to earlier: "The ULA only uses address line A14 (and the Z80 CPU control line /MREQ) to decode the control signals for the ROM and RAM".

Both A14 and /MREQ come from the Z80.

Although the ULA may include slightly more complex circuitry for timing purposes and addition logic for getting display data from ROM and RAM, for the Z80, it's basically this:

When /MREQ is active (logic low) and A14 is low, the ULA should output a low on /ROMCS.

Similarly, when /MREQ is active (logic low) and A14 is high, the ULA should output a low on /RAMCS.

Generally, once the PCB has been inspected (use of a good light and a magnifying glass is recommended) for any damage to components, PCB tracks and pads (broken/cracked/lifted tracks, short circuits, including solder splashes some of which can be as fine as a human hair) on both sides of the board, the next step is to try a known good working RAM pack (if you have one).

An external RAM pack disables the internal SRAM chip(s).

Next, would be to try a different ROM chip (if the existing chip is in a socket and you have a spare, or if you have/can program a EPROM and make the wiring alterations).

After this, it gets a bit more tricky. If it's known or suspected that the board has been worked on before, or if other steps have not found a problem, then it's a good idea to continuity test each PCB track between the Z80, the ROM, the RAM and the ULA. Keep in mind that some lines have resistors in them. Also the address and data lines to/from the RAM chip(s) have some lines swapped around. Then test between tracks that run close to others for any short circuits.

Although a 'scope is an excellent tool for fault finding, with the ZX81, because a failure of any of the chips brings the entire system to a state of chaos, it makes fault finding rather tricky.

Mark

[AMDeee]

I appreciate your help, I'll try getting my hands on a ram pack or try making my own and I'll report back if anything changes. Thank you!

[1024MAK]

On your board, which chips are in sockets?

If you don't have a RAM pack, if your SRAM chip(s) are in sockets, you could try changing them.

I also have some other tricks that may help.

Mark

[AMDeee]

The only chip that isn't socketed on this zx is the Ram, it has a single 24pin ram chip. I don't have any other ram chips on hand but I might as well desolder it, socket it and buy a new chip and try that out

[AMDeee]

Given in the topic I linked to earlier: "The ULA only uses address line A14 (and the Z80 CPU control line /MREQ) to decode the control signals for the ROM and RAM".

Both A14 and /MREQ come from the Z80.

Although the ULA may include slightly more complex circuitry for timing purposes and addition logic for getting display data from ROM and RAM, for the Z80, it's basically this:

When /MREQ is active (logic low) and A14 is low, the ULA should output a low on /ROMCS.

Similarly, when /MREQ is active (logic low) and A14 is high, the ULA should output a low on /RAMCS.

Generally, once the PCB has been inspected (use of a good light and a magnifying glass is recommended) for any damage to components, PCB tracks and pads (broken/cracked/lifted tracks, short circuits, including solder splashes some of which can be as fine as a human hair) on both sides of the board, the next step is to try a known good working RAM pack (if you have one).

An external RAM pack disables the internal SRAM chip(s).

Next, would be to try a different ROM chip (if the existing chip is in a socket and you have a spare, or if you have/can program a EPROM and make the wiring alterations).

After this, it gets a bit more tricky. If it's known or suspected that the board has been worked on before, or if other steps have not found a problem, then it's a good idea to continuity test each PCB track between the Z80, the ROM, the RAM and the ULA. Keep in mind that some lines have resistors in them. Also the address and data lines to/from the RAM chip(s) have some lines swapped around. Then test between tracks that run close to others for any short circuits.

Although a 'scope is an excellent tool for fault finding, with the ZX81, because a failure of any of the chips brings the entire system to a state of chaos, it makes fault finding rather tricky.

Mark

I went ahead and probed /MREQ and A14 and there do appear instances where by the ULA should be pulling /ROMCS low by that logic, yet thats not happening. There is continuity between /MREQ and A14 on the z80 and the ULA

/MREQ -> blue
A14 -> yellow