Hello, new member, trouble building ZX80 :)

[MaFli]

Hi Zeddex,
Had a quick look at the schematics and pin 9 IC 18 drives part of the video circuits. There's an C-R after this pin that drives the input of IC11-2 and IC16-11. Do you have the correct values installed? 47pF and 470 ohm? I suspect that probing this point adds a bit of capacitance and causes the pulse to change a bit and therefore fixes your signal.
I've just checked on mine and the signal after C11 barely reaches 1.5V (with my probe's capacitance added). So that is officially to low for the Vih iput spec (2V) of 74LS logic. I assume you have 74LSxx logic and not 74HCxx devices. Correct?

SCRN0001.png (20.89 KiB) Viewed 3364 times

[zeddex]

Hi,

If I understand you correctly, that measurement comes from the "upper" part, of C11 the side closer to the rear of the case?

I am a total scope novice, but took a measurement there:

pic_32_3.jpg (105.7 KiB) Viewed 3346 times

(When I do this, the text on screen disappears, if I probe the other side, I get a 3.2Mhz square wave and the flickering calms down.)

Yes all parts are LS, the resistor is 470, I believe the capacitor is 47pf; my cheap VC97 multimeter seems unable to measure capacitors of that size, even on the Real ZX80 it thinks the cap is 1nf! They were the cheapest ones I could find so I will find something better and drop that in...

[1024MAK]

Are you trying to test a capacitor on it's own, or in circuit?

On it's own, a small value like 47pF will often read as a value that is greater, due to the stray capacitance of the meter leads, you, and maybe even the circuitry in the meter.

For values this small, forget in-circuit testing, there are just too many leakage or stray capacitive effects caused by wiring/PCB tracks and other components.

A cheap 47pF capacitor will be a ceramic type. I'm not convinced that this is the problem. The tolerance with this type of capacitor is fairly wide anyway. So even if the value is a bit off, it should not affect the circuit that much.

The capacitance of the 'scope probe will however be significant compared to a 47pF capacitor.

I suspect the problem may be elsewhere and not this capacitor.

Mark

[MaFli]

Maybe the C11 is not the problem, but looking at the schematic this part is where normal/inverted characters are generated. And looking at the TV snapshots on the previous page, this seems to be the problem. Just to make sure, only inverted characters come out wrong? Normal characters remain the same?

Do you see any signals on IC11-3 or IC16-3? If so, it is probably something else. I see short 50ns pulses on both. Otherwise have a look at IC12. That is the discrete Set/Reset Latch that switches between either a normal or inverted pixel.

[MaFli]

Quick and dirty snapshots of both of them. Timebase is set different on each, but you get the picture (pun intended )

[zeddex]

@ MaFli - yes the character set appears perfect on the non-inverted characters (at least, to me)

If I do a basic PRINT CHR$(N); program to display all the characters, it's only the inverted ones which go crazy. The regular characters are never affected.

If I attach the probe lead to the "bottom" leg of C11, or IC18-9, the flickering all stops and the image is perfect.

My oscilloscope knowledge is non-existent, I'm all about the Auto-Set at the moment, and I'm not sure how to match your time bases (don't laugh!) but I'm not seeing quite the same as you, particularly on IC16... however the waveforms look the same on both my clone and Real '80 (I've noticed your scope is a professional model, not my cheapo bargain basement Aliexpress edition )

@1024MAK - Yes I was testing the capacitor in circuit - another lesson learned! I replaced it anyway just to be sure... no difference which of course you predicted

[1024MAK]

IC18 pin 9 is one of the 3.25MHz clock lines. C11 AC couples this signal to IC11 & IC16.

Have a look at this schematic...
A non-highlighted version is here. Both from Grant's wonderful site

The actual pixel data comes from IC9 pin 7 or pin 9 (74LS165), but the inverted signal is done by a combination of IC11, 12, 13, 16 and 20. It is the signal on IC20 pin 5 that determines if the video is inverted. So please double check the connections in this whole area.

Mark

[MaFli]

Haha, nothing wrong with a Hantek. Should do the job just fine. What model is it btw?

Hmmm... are your chips socketed? It must be somewhere in the area that Mark suggested. You have already replaced the LS74 IC18. You could try that with the other ones. Or re-seat them again in their sockets. Is the distortion straight away or does it appear after a while? That could suggest something is heating up and is starting to malfunction. Al my LS logic remains cool to the touch. Does any of your LS logic appear warmer than the others?

Can you check power consumption? Maybe with a bench supply with current read out? My ZX80 clone draws around 275mA with a CMOS Z80, LS logic, modulator removed, ZX8-CCB attached with a bench supply set at 9V.

What still puzzles me is that when you probe C11 on either side ("bottom leg" or IC8-9), the image becomes normal. If you have a multimeter you could check for shorts or open circuits in the area Mark indicated. What I haven't checked is what happens if I probe the area while connected to my TV. TV is downstairs and nerd area is in a different room.

If your Hantek has dual channels, you could compare signals on both your original an clone at the same time.

[zeddex]

Just a mini update... gah, my clone now does not display anything at all! Just blank like before (nothing at the video out, plenty of activity around the CPU and other areas, including many of the previously checked ICs, so its not totally dead!)- so it's not an issue with the CPU as previously. none of the ICs get hot either. I suppose my soldering simply isn't as good as I thought it was... I even went back and reflowed all the pins, using flux, it all looks nice and shiny but who knows.

I think I will need to, as suggested, set them both up on separate channels and work through the schematic, hoping to find discrepancies, check for shorts etc. It will not beat me! A bit depressing though, I'd almost fully finished the thing and now seem to be working backwards

I've never looked properly at a schematic before but I think I'm getting the hang of it. I kind of understand the principles of logic ICs so hopefully I can trace it.

Thanks so much for all your help guys, once I have them set up properly with 2 seperate power supplies and probes I can thoroughly scrutinise and take proper notes. I think that would make it easier to review.

(Scope is hantek dso5102p, I don't have a bench PSU.. yet! Still using the old >9VDC Sinclair packs)

[1024MAK]

Do you have a pulsing signal on /IORQ on the CPU (pin 20) ?

If no, there will be no picture.

Mark