Repairing the monitor of a Sharp MZ-80A

[tjewell]

Hi all, new user here. Forgive me if I've missed a thread or FAQ covering this already.

I've got a Sharp MZ-80A micro from the 80s which has a fault with the monitor - a single horizontal line. The computer itself seems to be working fine. After much investigation, I think the problem is in the vertical deflection circuit of the monitor's video board. My problem is that we've now moved from digital to analogue electronics and I feel a little out of my depth. I can go into more detail about what I've done already and I have full schematics for the board, but I don't want to go into too much detail until I'm sure I'm in the right section of the forum.

Many thanks in advance! Tony.

[drussell]

First of all, be sure you keep the brightness WAY down while you have no vertical deflection or else you will badly burn the CRT very quickly.

Vertical issues are usually pretty straightforward to troubleshoot.  First check the output transistor, driver transistor, oscillator transistor and any passives connected directly to them for obvious shorts / opens.  A huge percentage of the time the fault itself is rather obvious, then you just have to figure out WHY it failed. 

[tjewell]

Well, I've got a lot further!

I've identified the vertical deflection circuit and checked all the electrolytic capacitors - two had failed and have now been replaced. I found a replacement uPC1031 IC and replaced that (it appears to use an IC rather than distinct transistors). I've tested the vertical part of the yoke, and the resistance is the same as a known working machine (that sadly, I don't have to hand, otherwise I'd swap tubes). There's no obvious fusible resistors to replace.

I'm a little stuck what to do next - any suggestions? I still get a horizontal line, and if twiddle v-lin (I think) a bit, I can get the line to be about 5mm thick - at that point I can make out a bit of text, so I know the computer is fine, and presumably the horizontal circuits are too. I'm stuck!

Many thanks in advance.

[CJay]

Forget the tube, that'll be fine, the scan coils might be suspect but I doubt it.

Most likely fault on something that old will be dried up capacitors especially as it seems the drive output is fed via a large electrolytic which will be under quite a bit of load.

Datasheet of the chip here shows an example application and just how many caps there are in and around the chip.

http://www.unisonic.com.tw/datasheet/PC1031.pdf

[tjewell]

Ah, useful schematic - glad to see I identified all the capacitors correctly. Two of those were dead - as you suspected the big 2200uf (C10) and also C8, the 47uf that's also connected to the deflection coil.

I replaced the uPC1031 before I spotted the dead capacitors (I know, I did this in the wrong order a bit). Would putting a working IC into the circuit with the dead capacitors destroy a second IC? Is it worth me trying to find another?

[CJay]

I reckon it's unlikely to have damaged the first one but it's possible.

What test gear do you have and have you checked every one of those capacitors?

Unfortunately without a 'scope it might be a case of bulk checking components, checking for dry joints, also check the preset controls are 'smooth and don't go open circuit anywhere on their travel.

Is it the same monitor as the one in the MZ80K?

Check this video at 17:40, shows the monitor board



Service manual here which has the MZ80K monitor schematic with the same frame chip

https://www.sharpmz.org/mz-80k/download/sm80k.pdf

[tjewell]

I've got a multimeter and a capacitor meter. I've been tempted to get something like a Bitscope, because the service manual has a lot of detail on what signals to expect where.

As far as I can tell, it's pretty much the same monitor board as the MZ-80K, just a different shape (there might be subtle differences, but it's the same IC, and the core of the circuitry looks the same).

I got the service manual from here - https://mz80a.files.wordpress.com/2016/08/mz-80a_service_manual.pdf

So I'm still stumped - as you suggest, maybe it's time to move beyond capacitors and check the other components. I've checked for continuity and dry joints around the IC and the yoke connection, but I guess I need to widen my search.

Thanks for your help!

[CJay]

I've just printed thepair of schematics, close enough that they look interchangeable. The MZ80K one is useful as it's got voltages marked on various points.

And ooh that's nasty, they take composite sync and filter off the horizontal pulses to leave only the Vertical to be fed to the frame chip.

Do the DC checks, make sure the +12V is correct, check you've got 11.4V on pin 2 and 10 of the uPC1031, my money is still on it being a faulty cap.

A 'scope would be really useful, a very cheap 20MHz analog one would be fine for this stuff, about £20 on ebay plus carriage.

[tjewell]

Ah, yes, we might be on to something! Looking at the MZ-80K schematics, I can see 12v just after the input at C2004 and again on the emitter of Q2003 (the two points marked on the schematic). So the +12v is fine ... but ... when I measure at either pin 2 or 10 on the IC, I see just over 4 volts (I note, before I replaced the IC or the two capacitors, I only saw 1v). But 4v isn't 11.4v! Hmm, although I've tested/replaced all the obvious capacitors round there - time to double check soldering etc.

Thanks for the tip on ebay, I'll take a look.

[CJay]

so across the 3.3R resistor you are seeing 8V drop, hate to say it but either something is drawing approx 2.4 amps or that resistor has fried and gone high in value.

I'd check the resistor, make sure the caps are OK (broken record I know), double check your soldering on the new chip, make sure that the tab of the chip isn't meant to have an insulating washer and if that checks out OK, desolder the chip and check pin 2 and pin 10 for shorts to pin 8 (GND)

[Rasz]

so across the 3.3R resistor you are seeing 8V drop, hate to say it but either something is drawing approx 2.4 amps
..
desolder the chip and check pin 2 and pin 10 for shorts to pin 8 (GND)

or just measure in circuit between those pins

[CJay]

Fair comment, it's unlikely to have a short elsewhere

[anachrocomputer]

Sounds like you're tracking it down! If you need to use a scope, have you considered visiting your local Hackspace or Makerspace? We have several scopes of the right type at the Bristol Hackspace, for instance. Are you anywhere near Bristol, by any chance?

[tjewell]

Okay, pretty sure I'm close now. I pulled the 3.3ohm resistor, and it's obviously blown (it's got a black burn on it, and it's no longer reading 3.3ohms!). I hadn't spotted it earlier because it was tucked right behind the 1000u capacitor. I'm guessing it was low resistance, acting more as a fuse? Would I need to replace it with exactly the same resistor? (which, of course, I don't have to hand).

@anachrocomputer - I'm actually based in Cambridge, but I'm sure somewhere as hi-tech as this will have something similar. Thanks for the suggestion!

[Macbeth]

Good luck. I have an MZ-80A in the attic and when I last looked at its internals all the ceramic bypass caps had a white salt on them. It still worked just fine.

[CJay]

Definitely another 3.3 ohm (half watt I think?) if you can lay hands on one, a 4.7 might do but you may see some frame cramping or distortion

Or not.

Resistors that are meant to be fusible tend to have a coarse, matt finish to them, they're usually 'flameproof' and designed to die graciously without .

I would think the new chip is still ok too

[tjewell]

Ah, I was so hopeful this would work! Now I do read 11.4v at pins 2 and 10, but the output is still exactly the same (a horizontal line, that with a little twiddling is about 5mm high with a bit of readable text from the boot screen in).

I tried measuring for shorts between pins 8 and 2/10, and I'm not sure how to interpret the results. The old IC that I removed gives 12K ohm between 2 and 8, and an open circuit between 8 and 10. The new IC, in circuit, gives 34ohms between 8 and 2 or 10.

Many thanks for your help and patience!

[CJay]

Humour me and put the original chip back in.

[tjewell]

Happy to humour you! After a certain amount of swearing, I've swapped the chips - and oh my gosh, it seems to be working! The display is a bit wobbly, which I suspect I can solve with a bit of pot twiddling, and there's something odd about the boot message that I suspect is unconnected. What do you think happened CJay?

Well, thanks CJay and everyone else that's helped. If you're ever in Cambridge, UK, the beers are on me.

[CJay]

Brilliant, glad to have helped and happy there's another working vintage machine.

Change the rest of the caps, do the ones in the main PSU as well (expecially after watching that Dexter's Lab video), I doubt there will be a 'tweak' that'll fix the wobble.

I reckon there's a good chance that the 'new' chip was either fake or faulty from new.