Re: Sharp MZ-80K help needed on repair
 

it is a difficult one I have it running the MROM at the moment I had left it off for about three hours turned it back on to a corupt screen press reset cleared fine to monitor screen.
The bending of the screen at the top if I press reset will not clear it but if I turn it off for about 30secs then back on we have straight text again but this could be a seperate issue with the monitor side.
Noticed that the keyboard sometimes we get most of the keys working then we lose that to only a couple working so wondering if thats a clue.
Put the ram tester rom back in and its running fine but still get the bending screen but computer still running at the moment the only constant I can think of is that when it does crash we lose the 5v on pin 14 on the 8255.

Re: Sharp MZ-80K help needed on repair
 

When it's doing that in a repeatable way, that is the time to try disconnecting pin 14 because that will tell us whether that mysterious 0V on V-Gate is being output by the 8255 or whether there is something else in play here - intermittent low resistance or short between V-Gate and ground.

From your most recent observations it appears that the V-Gate line can be at 5V even without the pullup resistor - this is the condition which we and the RAM TEST code expect, that is why the RAM test code does not bother to set up and use 8255 pin as an output. We need to know where and why low voltages sometimes appear on that pin.

Re: Sharp MZ-80K help needed on repair
 

Took out pin 14 of the 8255 Ram program still running.

Re: Sharp MZ-80K help needed on repair
 

Leave it out for now and check the voltages on the 8255 pin 14 and on IC25 pin 1, which are currently disconnected from each other, from time to time.

Re: Sharp MZ-80K help needed on repair
 

8255 pin 14 0v IC25 pin 1 5v

Re: Sharp MZ-80K help needed on repair
 

While it's in that condition use your little resistor rig to connect the disconnected 8255 pin 14 to nearby +5V (via the resistor). Now what voltage do you see on the 8255 pin, still 0V or some other voltage, with the resistor connected?

If you see 5V with the resistor connected leave the resistor in place and keep checking the 8255 pin voltage to see if it stays at +5V. If you see 0V on the 8255 pin even with a pullup resistor connected, report back.

Re: Sharp MZ-80K help needed on repair
 

Hi
sorry to be mr stupid and you are welcome to point out how stupid I am but if I connect 5v to pin 14 with the resistor then measure the same pin it will remain 5v.

Re: Sharp MZ-80K help needed on repair
 

Not necessarily, if the 8255 pin somehow gets switched to an output and drives low then the voltage on the pin will be about 0V or at least very low even when there is a pullup resistor connected to +5V.

The drive downwards to 0V by the 8255 pin in output mode would be stronger than the modest pullup to +5V by the resistor, so the 8255 output would win.

The purpose of the resistor is to allow us to see if / when the 8255 pin does go into output mode and drive low, for reasons unknown.

Re: Sharp MZ-80K help needed on repair
 

sods law it wont crash at the moment so dont know if thats good or bad but still get the pulling screen and the keyboard at the moment has decided only the few keys that are responding this time are not showing what they are surpose to show for example the bottom row instead of CR space etc are producing numbers instead.and not getting a flashing cursor.

Re: Sharp MZ-80K help needed on repair
 

When you are seeing these keyboard faults I expect that's with the MROM in and with the 8255 inserted normally?

While the keyboard fault is present, try scoping the following IC6 pins:- 15, 14, 13, 12.

Re: Sharp MZ-80K help needed on repair
 

I think that's down to the telly, safe to ignore for now.

Re: Sharp MZ-80K help needed on repair
 

The CRT display is actually an integral part of the machine, I don't think it is easy to connect to any other display as the video and combined sync signals are TTL-Level. It could be down to malformed sync pulses as well.

Re: Sharp MZ-80K help needed on repair
 

picture of pin 15 scope the other pins are simular and alter when a working key is pressed but pin 12 has 1v on it and does not move.flashing cursor has returned.

Re: Sharp MZ-80K help needed on repair
 

I think you meant that pins 15, 14, 13 of IC6 have activity on them but 12 is stuck at 1V.

If so, can you do the same check on pins 4, 3, 2, 1 of the 8255, especially pin 1, which generates the signal you should be seeing on IC6 pin 12?

Re: Sharp MZ-80K help needed on repair
 

Checked 8255 pins and they are fine but then rechecked pin 12 on IC6 and that is showing fine it might have been my fault it is in a hard to reach area.

Re: Sharp MZ-80K help needed on repair
 

If you read 1V on IC6 pin 12 originally you must have been making contact with it, or do you think that maybe you might have been on the wrong pin?

I've just been back through the thread and while I remember you chose to replace the 8255 because the original had badly corroded pins I didn't see anywhere where you said you had also changed the 8255's socket, sorry if I missed it?

If the connection between the 8255 pin 1 and IC6 pin 12 is coming and going that will have a direct effect on the scanning of the keyboard keys.

Re: Sharp MZ-80K help needed on repair
 

Yes I have replaced the 8255 socket. And I may have found the issue I have discovered if I press the 8255 down flexing the board I am losing then gaining pin 12 on IC 6 so I will in the morning get the board out and take a look.

Re: Sharp MZ-80K help needed on repair
 

Excellent, if you can get to the bottom of that, that will be one less intermittent problem. Hopefully that will be progress of a sort.

Re: Sharp MZ-80K help needed on repair
 

Sorry to have to suggest this, but it may be worth taking the 8255 out just to make sure that pin 1 is not folded over. Still happens to me from time to time, even now.

Re: Sharp MZ-80K help needed on repair
 

And ... also check the IC socket pins arent folded under, and soldered correctly.
You can normally check for this type of fault with torch and magnifier.
Good luck.