Sinclair ZX80 / ZX81 / Z88 Forums

I have replaced C10. No change. R33 and R34 were measuring their correct resistances. All the continuity back to ULA pin 20 is fine. I used the test program linked above and it just detects zeros whatever is playing via EAR. I have replaced the 4K7 resistor at R33 for a 3K3 one and it made no difference.

One observation: on the working machine the bottom end of the ULA is the hottest part. On this non-functioning machine, the bottom end of the ULA (which is where pin 20 is located) is the least hot part. I think I have to conclude that this ULA is shot, and it's soldered to the board so it's not like I can easily swap in another to try it out.

Clutching at straws here, but I didn't yet change the electrolytic caps C3 & C5. Is there any chance these could affect the EAR input? I've looked at the schematic. I know very little about electronics, but C5 looks to be involved with X1 (I presume that's the oscillator?), and C3 looks like it's across the 9V supply.

C3 (22µF) is the bulk capacitor for the 9V rail, and helps to decouple the input supply to the 7805 voltage regulator.
C5 (1µF) is the power on reset capacitor for the Z80A microprocessor.
They don’t affect the ear / tape input system, if the ZX81 is otherwise okay.

I have to conclude that the input circuitry for the ULA pin 20, the Tape In pin is faulty.

Just one more test, assuming you have not done it already. With the power off, measure the resistance between pin 20 and 0V/GND. Also between pin 20 and the +5V rail.

Mark

My multimeter reads "1." for both when I do that, which is the same reading it gives when neither probe is touching anything, so I guess that means there is no short of pin 20.

I'm considering removing the ULA and socketing it, and then trying to buy a battered Zeddy on eBay to be a donor. Is this madness, or can it reasonably be removed - if for example I've previously managed to successfully remove the RAM chips without damage?

Yes, of course it’s possible.

It’s just a case of being careful if you don’t want to damage either the chip or the PCB.

Always double check that each and every pin is free to move before trying to lift it out.

It is possible to replace the lost functionality by using two 74LS series logic chips.

Mark

Hang on, looks like I measured my resistances wrong. I didn't realise it would be directional, and it only works with the probes one way around. Is that because the connectivity to +5V and GND is going via a diode inside or upstream of the ULA?

If I put the negative probe on the ULA pin 20, and with the selector knob on 2K resistance scale, with R33 value still substituted to 3K3 (instead of 4K7) I get:
ULA pin 20 to +5V = 1.807 (is that K?)
ULA pin 20 to GND = 0.765

Does that allow any conclusions to be drawn? Is there a short somewhere?

Opened up my working Issue 1 to test:
ULA pin 20 to +5V = 1.737
ULA pin 20 to GND = 0.768

So the values are pretty consistent. Which is odd because it kind of suggests that the ULA circuitry on the defective Issue 3 should be ok.

I tried changing R34 from 220R to 470R and this also had no impact. So I guess it's ULA trouble.

If I buy a donor machine, can I use an older ULA in a system that originally had the newer one with back porch?

Yes. Unlike the ZX Spectrum (where newer ULA versions were backwards compatible and would work in older issue boards, but older ULA versions were not recommended for older issue boards), it makes no difference (as far as we know) which ZX81/TS1000 board a ZX81 ULA is fitted in.

In actual fact, although there are some minor differences between the ZX81 issue 1 board, the ZX81 issue 3 board and the TS1000 board, all the important circuitry is the same of very similar (and some TS1000 cases have ZX81 issue boards in them).

Mark