UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Computers

Notices

Vintage Computers Any vintage computer systems, calculators, video games etc., but with an emphasis on 1980s and earlier equipment.

Reply
 
Thread Tools
Old 8th Oct 2021, 4:44 pm   #241
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 7,521
Default Re: Sharp MZ-80K help needed on repair

To determine whether you have a keyboard problem or a keyboard circuit problem, first thing to do is to unplug the keyboard from the motherboard and use a short wire link to connect any of the keyboard connector pins 1-8 to any of pins 9-18. If the keyboard circuit is working you should be able to 'type' characters by joining pairs of pins together. If you find that this is working you could write or make a table of which pairs of pins 'type' which characters - some of them won't be obvious, like SHIFT which doesn't do anything by itself.

Don't connect any of the pins in the 9-18 range to each other though, could damage IC6.
SiriusHardware is online now   Reply With Quote
Old 8th Oct 2021, 4:49 pm   #242
pudwink
Pentode
 
Join Date: Jan 2011
Location: Nottingham, Notts. UK.
Posts: 164
Default Re: Sharp MZ-80K help needed on repair

oh yes I do it was booting now getting corrupt screen again put the dram tester program back on and corrupt screen on that again and if I do sometimes get a clear screen program wont run ahhhhhhh.
pudwink is offline   Reply With Quote
Old 8th Oct 2021, 5:08 pm   #243
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 7,521
Default Re: Sharp MZ-80K help needed on repair

It sounds like you have that most feared of all faults, a physical / intermittent fault. The good news is that the machine has shown you, briefly, that it can work.
SiriusHardware is online now   Reply With Quote
Old 8th Oct 2021, 5:16 pm   #244
pudwink
Pentode
 
Join Date: Jan 2011
Location: Nottingham, Notts. UK.
Posts: 164
Default Re: Sharp MZ-80K help needed on repair

unless IC36 and 37 have gone bad again
pudwink is offline   Reply With Quote
Old 8th Oct 2021, 5:22 pm   #245
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 7,521
Default Re: Sharp MZ-80K help needed on repair

Did you say that one of your device programmers has a chip test mode on it? (Just not for DRAMs, as I recall). If so try testing those ICs.
SiriusHardware is online now   Reply With Quote
Old 8th Oct 2021, 5:30 pm   #246
pudwink
Pentode
 
Join Date: Jan 2011
Location: Nottingham, Notts. UK.
Posts: 164
Default Re: Sharp MZ-80K help needed on repair

no unfortunetly not got a dram tester and a eprom programer they wont check the ICs I only brought 3 of the SN74LS241N so I had a spare tried replacing ether IC36 or 37 with the spare no difference so ether both have blown or its something else.
pudwink is offline   Reply With Quote
Old 8th Oct 2021, 5:48 pm   #247
pudwink
Pentode
 
Join Date: Jan 2011
Location: Nottingham, Notts. UK.
Posts: 164
Default Re: Sharp MZ-80K help needed on repair

If IC36 and 37 have blown again what would cause this.
pudwink is offline   Reply With Quote
Old 8th Oct 2021, 5:59 pm   #248
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 7,521
Default Re: Sharp MZ-80K help needed on repair

Well, believe it or not you have still made progress because the machine worked for (hopefully) just long enough to let you diagnose which of the RAMs were faulty so that's one less area of immediate concern.

Did you try the 'press test' Mark suggested earlier, apply pressure to each IC in turn while resetting or turning on the machine? You could get lucky and identify a chip or an area which is physically sensitive / intermittent.

You can buy testers, similar to your DRAM tester, which are intended for the purposes of testing TTL and CMOS logic ICs. Higher end device programmers sometimes have that as a secondary function.

One thing you could do is put each of the suspect ICs in a plug-in breadboard if you have one - make sure you place it with the legs on either side of the centre line so that the breadboard itself does not short any of the IC pins together. Measure the resistance from the 0V pin (pin 10) of the IC to each input on the IC

(pins 2, 4, 6, 8, 1, 13, 15, 17)

and from the 0V pin (pin 10) to each output on the IC

(Pins 3, 5, 7, 9, 12, 14, 16, 18)

And also from the 0V pin (pin 10) to the two enable inputs on the IC (pins 1 and 19).

The resistance from the 0V pin to all pins of a similar type should be about the same. Same goes for the resistance from the +5V power pin (pin 20 on this IC) to all pins of a similar type.

If you don't find a significant difference between two pins of the same type it doesn't necessarily mean that the IC is OK, the fault could just be buried too deep inside the IC to affect the readings on the pins.

On the other hand, if you do find a very significant difference on one pin compared to all the other pins which have a similar function, regard that as a probable faulty device.

Remember when making measurements like this on logic ICs, always keep the probes the same way around - as a guideline when checking from the 0V pin to other pins, always have the black probe on the 0V pin, and when checking from the 5V supply pin to the other pins always have the red probe on the +5V supply pin.

Edit: It's not actually certain that IC36 and IC37 have blown or even that they were originally blown. If the machine is damaging data buffers it could be because there is 'bus contention' - two things trying to drive the data bus at the same time - due to a circuit fault, but that can not be happening all the time because the machine would not have worked for the short time that it did.

Last edited by SiriusHardware; 8th Oct 2021 at 6:05 pm.
SiriusHardware is online now   Reply With Quote
Old 8th Oct 2021, 6:19 pm   #249
pudwink
Pentode
 
Join Date: Jan 2011
Location: Nottingham, Notts. UK.
Posts: 164
Default Re: Sharp MZ-80K help needed on repair

I have tried the press test no results I think the easy option is to get hold of some more SN74LS241N and see if that gets us back to booting up if not something else if yes we will have to work out what blows SN74LS241N ICs.
pudwink is offline   Reply With Quote
Old 8th Oct 2021, 9:01 pm   #250
Mark1960
Heptode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 687
Default Re: Sharp MZ-80K help needed on repair

It might be worth trying again after leaving it in a cold room overnight, just in case its a thermal intermittent.

Another thing worth checking is to measure pin 6 on the z80, make sure the clock input is driving a clean square wave as close to each supply rails as possible. I noticed this is driven by a 74LS244 with a 330 ohm pullup, so I wonder if that gate might be getting a bit tired due to old age. One of the quirks of the z80 was that the clock input is not ttl compatible, we used to drive the clock with 74S04 and 330 ohm pullup or with a transistor buffer.
Mark1960 is offline   Reply With Quote
Old 9th Oct 2021, 9:54 am   #251
pudwink
Pentode
 
Join Date: Jan 2011
Location: Nottingham, Notts. UK.
Posts: 164
Default Re: Sharp MZ-80K help needed on repair

Tried it again this morning still the same this is what I get on pin 6 of the Z80
Attached Thumbnails
Click image for larger version

Name:	IMG_20211009_105027[1].jpg
Views:	22
Size:	62.9 KB
ID:	242982  
pudwink is offline   Reply With Quote
Old 9th Oct 2021, 3:11 pm   #252
pudwink
Pentode
 
Join Date: Jan 2011
Location: Nottingham, Notts. UK.
Posts: 164
Default Re: Sharp MZ-80K help needed on repair

Hi
I have replaced IC 36 and 37 again and no difference just the corrupt screen so at least we know it's not that.
pudwink is offline   Reply With Quote
Old 9th Oct 2021, 5:41 pm   #253
Mark1960
Heptode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 687
Default Re: Sharp MZ-80K help needed on repair

The clock looks ok to me.

Most likely cause of an intermittent fault would be an IC socket connection, could be IC pins corroded, socket connection corroded, maybe IC pin bent under with a turned pin socket, or a missing gold insert on a turned pin socket.

As it seems to be in a permanent fail it might be possible to find the fault by measuring resistance between IC pins and the pad on the pcb, but take care not to apply pressure that might temporarily improve the connection.

Another possibility is failed through hole plating, I have seen this with boards from the 80s when the pcb manufacturing process was not so well developed. I have a reject z80 pcb from that time that I fitted links through each through hole to get it working. I would have expected this type of fault to have caused a much earlier failure in the life of the mz80 though.

Is there any signs of corrosion in any area of the board? Perhaps close inspection and checking continuity of through hole vias in any areas of corrosion might show a fault.
Mark1960 is offline   Reply With Quote
Old 10th Oct 2021, 8:54 am   #254
pudwink
Pentode
 
Join Date: Jan 2011
Location: Nottingham, Notts. UK.
Posts: 164
Default Re: Sharp MZ-80K help needed on repair

I am just wondering why when it goes corrupt that the reset side fails to work correctly and I get a lot of black screens where pin 14 of the 8255 stays at 0v until the picture comes back.
Now you will probably tell me this is correct just trying to narrow down what is causing this it was so sweet the brief moment we got it working would like to see it working again.
pudwink is offline   Reply With Quote
Old 10th Oct 2021, 10:19 am   #255
pudwink
Pentode
 
Join Date: Jan 2011
Location: Nottingham, Notts. UK.
Posts: 164
Default Re: Sharp MZ-80K help needed on repair

Now I have been continually pressing the reset button (not that it does a lot) and its started working again and most of the keys are working this time, but after about a min the top line text starts pulling to the left which is what it did last time before it stoped working but if I turn it off leave it a short time then turn it back on again the top line is perfect again for a short time and starts pulling again.
Attached Thumbnails
Click image for larger version

Name:	IMG_20211010_110642[1].jpg
Views:	17
Size:	40.8 KB
ID:	243058  
pudwink is offline   Reply With Quote
Old 10th Oct 2021, 10:59 am   #256
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 7,521
Default Re: Sharp MZ-80K help needed on repair

Quote:
the reset side fails to work correctly and I get a lot of black screens where pin 14 of the 8255 stays at 0v until the picture comes back.
No, this is a problem.

Pin 14 (V-Gate out from the 8255) is not correctly set up in the RAM test code to output a logic '1' to make V-Gate high - as you are in contact with the author of the RAM test EPROM maybe you could alert him to that problem or just draw his attention to this thread - he obviously has very good knowledge of these machines so we'd welcome his direct input. However, the resistor you added should compensate for that issue in the meantime.

It is possible that in the circumstances you describe the 8255 is being part initialised with that port pin set to be an output and then driven low, which will have the effect of driving the V-Gate line low and turning the video off, hence blank screen.

There are only officially two things connected to that line, 8255 pin 14 and IC25 pin 1 (plus the added pull up resistor). First check the resistor you added, make sure the soldering is good and that you did connect the other end of it to a known good +5V point.

To eliminate the 8255 as the source of the problem remove it, carefully bend pin 14 out to one side so it no longer goes into the socket and then re-insert the 8255 still with that pin hanging out to one side. Does that allow the screen to stay on consistently?
SiriusHardware is online now   Reply With Quote
Old 10th Oct 2021, 12:07 pm   #257
pudwink
Pentode
 
Join Date: Jan 2011
Location: Nottingham, Notts. UK.
Posts: 164
Default Re: Sharp MZ-80K help needed on repair

At the moment it is booting up corectly but not leaving it switched on long so the 5v on pin 14 is constant so untill it stops booting dont think bending the pin out is going to prove anything at the moment.
with the resistor just did a temporary measure.
Attached Thumbnails
Click image for larger version

Name:	IMG_20211010_130111[1].jpg
Views:	12
Size:	79.2 KB
ID:	243060  
pudwink is offline   Reply With Quote
Old 10th Oct 2021, 12:12 pm   #258
pudwink
Pentode
 
Join Date: Jan 2011
Location: Nottingham, Notts. UK.
Posts: 164
Default Re: Sharp MZ-80K help needed on repair

Actually just tried something else now it is booting I have disconnected the resistor and still getting 5v on pin 14 so I don't think the resistor is making any difference and not needed.
pudwink is offline   Reply With Quote
Old 10th Oct 2021, 12:25 pm   #259
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 7,521
Default Re: Sharp MZ-80K help needed on repair

Whew, this is a tough one because we seem to be trying to hit a moving target.

Just to clarify, are you saying you now think this is time related so that if you leave the machine on the video becomes distorted at the top and then eventually goes off?

If I have that right, let it happen so that you have a blank screen and then measure or scope the logic level on IC25 pin 1 and also scope IC17 pins 1, 2, 13 and 12. What waveforms or steady states do you see on those pins when the 'blank screen' fault is present?
SiriusHardware is online now   Reply With Quote
Old 10th Oct 2021, 12:45 pm   #260
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 7,521
Default Re: Sharp MZ-80K help needed on repair

The next time it boots up and seems to be running the RAM TEST code normally with the 1-2-3 rolling in the corner, try pressing and flexing the PCB - does that make it crash, does the number in the corner stop rolling?

Looking at the broader picture we can say that it didn't work, then it did very briefly, now it doesn't again. It would be unusual (although not impossible) for an IC to be faulty, fix itself and then break down again so it is more likely that you are looking for a physical-intermittent fault such as a bad IC socket contact or a bad VIA (a through hole connection between the upper and lower side of the PCB) or a fractured or damaged track. You'll only find a problem like that by exercising extreme patience and measuring from every point on each circuit node to every other point on the same circuit node.

This will be made even harder if the fault is connecting and disconnecting whenever it feels like it so it just happens to be working at the moment when you measure it.
SiriusHardware is online now   Reply With Quote
Reply

Thread Tools



All times are GMT. The time now is 9:06 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2021, vBulletin Solutions, Inc.
Copyright ©2002 - 2021, Paul Stenning.