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.

Closed Thread
 
Thread Tools
Old 30th Sep 2021, 3:32 pm   #161
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Sharp MZ-80K help needed on repair

Another thing for you to try in addition to the above. With the power on, try locating capacitor C19 (near connector CN3) and shorting its +ve lead to 0V / Gnd. If you can't get to the +ve lead of C19, the anode end of either D1 or D2 - the end which does not have a line around it - is the same circuit point.

Taking D1 anode / D2 anode / C19 +ve momentarily to 0V / GND should force a power-on reset. If you monitor the Z80 pin 26 at the same time, you should see a definite change to low and then high again. This reset pulse should be much longer than the one generated by the reset-switch circuit.
SiriusHardware is online now  
Old 30th Sep 2021, 5:26 pm   #162
pudwink
Pentode
 
Join Date: Jan 2011
Location: Nottingham, Notts. UK.
Posts: 207
Default Re: Sharp MZ-80K help needed on repair

IC31 pin 1 is high IC31 pin 6 is low
pudwink is offline  
Old 30th Sep 2021, 5:40 pm   #163
pudwink
Pentode
 
Join Date: Jan 2011
Location: Nottingham, Notts. UK.
Posts: 207
Default Re: Sharp MZ-80K help needed on repair

https://youtu.be/fSoo9A2vG-o
Made a video of what it does on boot up then keep pressing reset button.
pudwink is offline  
Old 30th Sep 2021, 5:42 pm   #164
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Sharp MZ-80K help needed on repair

OK, as expected. If you logic-probe IC36 pin 6 you should see the PULSE Led flash when you press reset.

Could you now try making the machine do a power-on reset by shorting D1 anode or D2 anode momentarily to 0V, while monitoring Z80 pin 26 with your probe or scope. The purpose of this is to a) verify whether the circuit which generates the power-on reset is working, and b) to see if the Z80 starts consistently and always does the same thing after a longer reset pulse.

Edit: thanks for the video. Can you try the same thing while forcing a power-on reset over and over again without actually turning the power off? You don't need to make a video, just describe what happens.
SiriusHardware is online now  
Old 30th Sep 2021, 6:55 pm   #165
pudwink
Pentode
 
Join Date: Jan 2011
Location: Nottingham, Notts. UK.
Posts: 207
Default Re: Sharp MZ-80K help needed on repair

Hi
Shorted D1 and the scope shows pin 26 of the Z80 going high to low and keep shorting get the same efferent on screen as the video only quicker reset action.
pudwink is offline  
Old 30th Sep 2021, 7:19 pm   #166
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Sharp MZ-80K help needed on repair

OK, so we know that the power-on reset circuit is OK and the common section of the reset circuit (IC24 pin 6 onwards) is OK. The logic probe says that the switch reset is also working, I think you just need to be a bit more used to your scope before you can get the hang of seeing very short single events on the scope.

It looks as though the CPU is not reliably able to execute code, though.

How are you with building things, have you made a few electronics kits or built anything on veroboard? What we may ask you to do is to build a 'NOP circuit' which is a simple rig which replaces the CPU and makes regular signals appear on the system databus, address bus, chip select signals and some of the control lines, which in turn makes it easier to see if they are working normally.

We could also consider writing very simple bits of test code for you to programme into an EPROM but it seems doubtful whether even they would work at the moment.

Can I ask you to repeat an earlier test, try starting the system with no EPROM fitted in the MROM socket. If you do that do you still get a full screen of completely random characters every time you switch on?
SiriusHardware is online now  
Old 30th Sep 2021, 7:55 pm   #167
pudwink
Pentode
 
Join Date: Jan 2011
Location: Nottingham, Notts. UK.
Posts: 207
Default Re: Sharp MZ-80K help needed on repair

Took the M rom out and weirdly it shows the original problem of lines on the screen now.
pudwink is offline  
Old 30th Sep 2021, 9:04 pm   #168
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,264
Default Re: Sharp MZ-80K help needed on repair

Those vertical lines are caused by the z80 executing rst 38 (FF), due to no rom present, then at location 0038 it executes rst 38 again, so it keeps pushing the return address 0039 onto the stack, but never finds a return instruction so it just fills all ram with 0039, including the video memory. Note the pattern on the display is two characters wide.
Mark1960 is online now  
Old 30th Sep 2021, 11:12 pm   #169
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Sharp MZ-80K help needed on repair

That's making it sound like it can execute code, just not the code in the ROM, at least not reliably. What next?

I note there are buffers on all of the address lines (IC44, IC45, =74LS241) and on the data lines (IC36, IC37 = 74LS240) between the CPU D0-D7 lines and the buffered system data lines, also confusingly called D0-D7 rather than say DB0-DB7. There is also a buffer between MROM / the video system and the system data bus (IC50, 74LS244). Mark already suggested the replacement of one buffer (IC43, 74LS245) which has been done.

We know from experience that these 74LS24x buffers can be troublesome - given their relatively low cost, should we maybe just suggest to pudwink to replace these buffer ICs as a set? I'll wait for a consensus on this. There would be no guarantee that it would fix the problem.

We could try scoping all the Address, Data lines etc and hope to see something suspicious but it would be better then to force the system to generate predictable signals, ie, through a NOP test if we want to go that route, so that any problems stand out more.

Last edited by SiriusHardware; 30th Sep 2021 at 11:17 pm.
SiriusHardware is online now  
Old 1st Oct 2021, 12:19 am   #170
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,264
Default Re: Sharp MZ-80K help needed on repair

The ram checker was originally giving what seemed to be a correctly formatted display.

I think Pudwink should review any socket or chip replacements that he might have done since then. Checking connectivity of IC pins to any traces on top or bottom of the pcb to make sure there have been no damaged through hole vias and also no shorts on either side of the pcb, including under sockets. If turned pin sockets have been used make sure the gold insert contact is not missing and all pins are correctly inserted, not bent under the ic.

I think something has been damaged recently and we need to find that before swapping anything else and causing more damage.
Mark1960 is online now  
Old 1st Oct 2021, 12:31 am   #171
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: Sharp MZ-80K help needed on repair

Yes I agree with Mark every change should be scrutinized not to make the situation worse - think and check very carefully of what was done just before the failure.

On the other hand there is a recent failure but, that damage being the buffers is a possibility and having them socketed could lead to fast resolution of a common problem - I have had many fail or cause errors.
Timbucus is offline  
Old 1st Oct 2021, 1:04 am   #172
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Sharp MZ-80K help needed on repair

We've actually been here before but I don't know whether the suggestions were followed through.

Here are the actions as stated by pudwink, in order, around the time that the display problem appeared. Before that point we had a working display but some other problem with the RAM TEST program which we were about to try to get down onto.

Quote:
I took the board out to wire in a reset button all done correctly put the board back in and got the lines on the screen back.

I redid the eprom and now all I get is this (The scrambled output from the RAM TEST program)

I took the board out again I have replaced the M-Rom socket again just in case and reflowed the Z80 socket
I see two potential problems in there, one is that while the board was out to have the reset switch fitted an original physical fault may have been disturbed which caused the reappearance of the white lines fault.

In an attempt to fix this, pudwink then re-did the EPROM which I take to mean erased and reprogrammed the same device, in the immediate aftermath of which of which the RAM TEST program stopped working the way it had been and has not worked that way since. However, this could be due to the same fault which pudwink disturbed when he had the board out, and not necessarily due to corrupt code in the RAM TEST EPROM.

Taking that question first, I really think the RAM TEST needs to be programmed into another, previously unused device to see if that reverses the problem with the RAM TEST. Otherwise, we will never know if a new problem was introduced when the EPROM was reprogrammed.

We did also suggest checking continuity from all of the IC pins of the IC plugged into the MROM socket to the associated PCB pads - again, I don't know if this was done.

To be really thorough the connections between each pin of the MROM and every other place that each pin of the MROM should be connected to (according to the diagram) should be checked for continuity in case one of the incoming or ongoing address or data lines has been severed from one of the MROM pads. The same may also apply to the CPU since that socket has been changed(?) / reflowed too.
SiriusHardware is online now  
Old 1st Oct 2021, 6:11 am   #173
pudwink
Pentode
 
Join Date: Jan 2011
Location: Nottingham, Notts. UK.
Posts: 207
Default Re: Sharp MZ-80K help needed on repair

Hi all
Thinking about it this morning then looking back at the posts I think we have got a little confused.
SiriusHardware asked me to remove the Mrom as earlier test to see if we still got a screen full of characters but that's not what we did before we actually removed the Z80 to get the character screen.
So getting the lines on the screen with the Mrom removed is correct it's my fault should have relized sooner.
pudwink is offline  
Old 1st Oct 2021, 9:07 am   #174
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Sharp MZ-80K help needed on repair

OK, that fits in with Mark's idea that the Z80 is seeing FF on the databus and interpreting that as an RST 38 instruction, resulting in the lines you see. Removing the Z80, you should see random characters because nothing gets written to the screen RAM.

Can you:

-Do a careful physical / visual examination of all rework which you did starting from the time you took the PCB out to fit the reset switch. Check for shorts, possible missed or dry solder joints, solder hairs or splashes between tracks.

-Program the RAM TEST code into another EPROM, not the same one, if you have not tried that, to rule out a bad / intermittent EPROM.

-With MROM in the socket and power off, meter out all the connections from the IC pins to the associated PCB pads. By 'meter out' I mean check that the connection between each IC pin and its associated PCB pad is near to zero ohms.

These suggestions are based on the fact that you did a bit of work on the machine and then things changed for the worse. Although it could be purely unlucky coincidence that something new went wrong just at that time, the balance of probability is that something you did, intentionally or not, caused this new problem, so you have little choice but to rule that out first.

If everything meters out so far we may have to look for lost connections in the wider area.
SiriusHardware is online now  
Old 2nd Oct 2021, 5:36 pm   #175
pudwink
Pentode
 
Join Date: Jan 2011
Location: Nottingham, Notts. UK.
Posts: 207
Default Re: Sharp MZ-80K help needed on repair

Hi
Still waiting for the 8255 which is annoying and I have just ordered a couple of the SN74LS00N so I have them.
I have checked the socket through from top to bottom with the multimeter no issues I had 10 2532 used eproms all program and give the same screen.I have looked over the board cant see anything out of the ordinary.
I have supplied a pic of where I was soldering the reset just in case someone spots something I missed.
Attached Thumbnails
Click image for larger version

Name:	reset.jpg
Views:	49
Size:	85.0 KB
ID:	242539  
pudwink is offline  
Old 2nd Oct 2021, 9:03 pm   #176
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: Sharp MZ-80K help needed on repair

Quote:
Originally Posted by pudwink View Post
Hi
Still waiting for the 8255 which is annoying and I have just ordered a couple of the SN74LS00N so I have them.
I have checked the socket through from top to bottom with the multimeter no issues I had 10 2532 used eproms all program and give the same screen.I have looked over the board cant see anything out of the ordinary.
I have supplied a pic of where I was soldering the reset just in case someone spots something I missed.
If they are used you have given them a good 35 minute erase as previously discussed anything less risks some bits bleeding?
Timbucus is offline  
Old 2nd Oct 2021, 9:59 pm   #177
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,264
Default Re: Sharp MZ-80K help needed on repair

Quote:
Originally Posted by pudwink View Post
Hi
Still waiting for the 8255 which is annoying and I have just ordered a couple of the SN74LS00N so I have them.
I have checked the socket through from top to bottom with the multimeter no issues I had 10 2532 used eproms all program and give the same screen.I have looked over the board cant see anything out of the ordinary.
I have supplied a pic of where I was soldering the reset just in case someone spots something I missed.
Difficult to see with the low resolution and also quite a lot of flux residue which might be hiding something. It maybe worth removeing the flux, I find it easier to scrape off the board with the point of tweezers when the flux has had a day or so to dry out rather than using a solvent that tends to spread it around in a sticky mess. Be carefull not to use something sharp or apply too much pressure. Maybe also worth doing the other areas of the board that show signs of hand soldering.
Mark1960 is online now  
Old 3rd Oct 2021, 10:35 am   #178
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Sharp MZ-80K help needed on repair

Considering the next move after all avenues presently available to Pudwink have been explored, should we now look at knocking up a NOP test? That will create steady signals on many of the system lines and may reveal something if the various attempts so far have not.
SiriusHardware is online now  
Old 3rd Oct 2021, 12:25 pm   #179
pudwink
Pentode
 
Join Date: Jan 2011
Location: Nottingham, Notts. UK.
Posts: 207
Default Re: Sharp MZ-80K help needed on repair

The solder flux I use is no clean as I did not what the flux causing any issues.
I have had another look at the board today can't see anything that would cause an issue I even checked the Mrom socket to points on the board and that is fine no breaks.
I will try wipeing the EPROMs for 35 minutes just in case but they do test blank before au use them and even tried writing the bin fille 3 times at programming stage to the EPROM just to make sure.
pudwink is offline  
Old 3rd Oct 2021, 5:32 pm   #180
pudwink
Pentode
 
Join Date: Jan 2011
Location: Nottingham, Notts. UK.
Posts: 207
Default Re: Sharp MZ-80K help needed on repair

Tried both eproms again still the same just every now and again if I keep pressing reset I get the correct dram testing screen.
pudwink is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 4:11 pm.


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 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.