Page 1 of 2

Yet another repair

Posted: Wed Jun 27, 2018 4:19 pm
by Floopy
Hello, my name is Floopy.

I'm brand new here, but I've been playing with old hardware for quit some time. I was recently given a TS1000. I hate the keyboard, but the computer is restoration worthy. Anyway, when I power it up all I get is a black screen and occasionally a brief white line. My first step is checking the clock to make sure it's running, that way I can know where to install the logic analyzer clips. It should be 6.5Mhz?
When I use the oscilloscope to check pin 35 on the ULA all I see is +5V. No oscillation, when I check pin 14 there is no output on it either. Once I can I will post some numbers on current consumption and heat. It seems to have gotten very hot inside. Is that normal?

Is this something that fails? Where would I even get a replacement crystal.

Re: Yet another repair

Posted: Wed Jun 27, 2018 4:38 pm
by Paul
The Oscillator is 6.5 MHz and can be bought at Reichelt.de or on SellMyRetro.
The resulting Clock going to the processor is half of it, 3.25MHz.
These resonators (3 Pin) normally dont break. If its dead its normally the ULA that fails.
Kind regards
Paul

Re: Yet another repair

Posted: Wed Jun 27, 2018 4:44 pm
by Floopy
Tha's what I was thinking at first. I'll try to measure again tonight and see what I come up with.

Re: Yet another repair

Posted: Wed Jun 27, 2018 8:07 pm
by zx81jens
Hey Floopy.

Welcome :-)

Greetings
Jens

Re: Yet another repair

Posted: Thu Jun 28, 2018 5:46 am
by Floopy
zx81jens wrote: Wed Jun 27, 2018 8:07 pm Hey Floopy.

Welcome :-)

Greetings
Jens
Thank you,
I went back to measure and I can confirm that the pin 35 on the ULA is always high. 4.2V to be exact, no oscillation. I'm not very familiar with this kind of setup, does the ULA have to work to be able to create a clock? The current consumption is about 400Ma. I also tested the transistor(TR2) and it tested fine.

I'm not very familiar with this family of computers and that's why I'm doing such random testing procedures.

Re: Yet another repair

Posted: Thu Jun 28, 2018 7:24 am
by Paul
Yes, the ULA needs to be working to create a clock signal.
Is your Mainboard unmodified? Or could there be a short circuit some where in your Mainboard? Which ICs are socketed?
Can you please also measure your 5V supply?
In which continent are you?
Kind regards Paul

Re: Yet another repair

Posted: Thu Jun 28, 2018 11:33 am
by 1024MAK
The ULA in a ZX81 is essentially all of the glue logic in one chip. This includes the master oscillator (master clock). On the ULA, pin 35 is the connection to the ceramic resonator. The oscillator runs at 6.5MHz appropriately. I don't recommend testing here, as loading by connecting test gear may stall the oscillator.

Internally the ULA divides this master clock by two, and then outputs the 3.25MHz signal on pin 14. On an oscilloscope (that is sufficiently fast) you should see a square wave, but with a poor shape.

TR2 then inverts the clock and feeds it to the Z80 CPU (pin 6 on the CPU).

If the CPU is running, you should see activity on various control output pins, such as /MREQ (pin 19).

Without a clock, the CPU will just sit there with no activity and will likely have invalid states on the control output pins.

A link to the chip pin-outs is in my sig.

It's normal for the heatsink for the 7805 to run hot. The ULA also runs hot. The Z80 CPU when running, runs warm. The RAM and ROM should run cool or just slightly warm (depends on the actual type of chips fitted).

If the ULA is in a socket, the quickest way to fault find, is to swap it with a known good working ULA. Of course, this only helps if you have another one...

Mark

Re: Yet another repair

Posted: Thu Jun 28, 2018 6:16 pm
by Floopy
1024MAK wrote: Thu Jun 28, 2018 11:33 am If the ULA is in a socket, the quickest way to fault find, is to swap it with a known good working ULA. Of course, this only helps if you have another one...
Mark
That part where you suggest to swap another ULA. About that, where do you even find another ULA, it's a custom chip. The CPU is not doing anything, that's why I'm so obsessed with finding the clock signal. So the ULA is the one generating the signal. I think that it's probably toasted.

Anyone have made a custom replacement? I don't see the point of salvaging a working unit to make another work.

Re: Yet another repair

Posted: Thu Jun 28, 2018 6:44 pm
by mrtinb
If you had a working unit and a broken unit, you would be able to swap parts around to find which part is broken.

Re: Yet another repair

Posted: Thu Jun 28, 2018 8:48 pm
by 1024MAK
Original 'new' Sinclair ULA chips are hard to find now. But a member on here has produced a modern replacement module. See this thread ;)

Mark