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 Dec 2020, 4:33 pm   #141
Buzby123
Hexode
 
Buzby123's Avatar
 
Join Date: Oct 2011
Location: Culcheth, Cheshire, UK.
Posts: 297
Default Re: I found it! A very sorry looking MK14.

Quote:
Originally Posted by SiriusHardware View Post
What's the single digit, a reverse 'C' or a '0'?

If '0', would suggest the system is being held in reset.
Sometimes it's a T-dot, the exact opposite of reverse 'C'. Sometimes it's 8-dot, sometimes it's just a dot. !

Can't do any testing on it till later.
Buzby123 is offline   Reply With Quote
Old 8th Dec 2020, 4:38 pm   #142
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 6,529
Default Re: I found it! A very sorry looking MK14.

Hmm, actually, you only get a single 'O' (or with the old OS, '-') if the system is held in reset after it has been running correctly, since the segment output hardware remembers the last character which was sent to it.

Edit: Just caught your response. Sorry to have to ask, but you've checked the +5V rail is spot-on volts wise?
SiriusHardware is offline   Reply With Quote
Old 8th Dec 2020, 6:33 pm   #143
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 6,529
Default Re: I found it! A very sorry looking MK14.

Check list:

1) Visual inspection, no cut-off component leads curled over and getting to know each other on the track side of the PCB? (The sort of thing which could happen after the PCB has been pressed down on a few times).

2) Check the regulated supply voltage if not already done.

3) Remove and check all ICs for possible folded-over pins. While doing so, use your IC tester to re-check all the TTL (except the 7445 which it does not seem to do correctly)

4) Try all possible combinations of two of the four RAMs in the 'standard' RAM sockets, which on the issue VI are the two closest to the PROMs.

5) Try your other PROMs

6) Try another SC/MP.
SiriusHardware is offline   Reply With Quote
Old 8th Dec 2020, 9:13 pm   #144
Buzby123
Hexode
 
Buzby123's Avatar
 
Join Date: Oct 2011
Location: Culcheth, Cheshire, UK.
Posts: 297
Default Re: I found it! A very sorry looking MK14.

Check list:

1) Visual inspection, no cut-off component leads curled over and getting to know each other on the track side of the PCB? (The sort of thing which could happen after the PCB has been pressed down on a few times).

No curly leads.

2) Check the regulated supply voltage if not already done.

4.96v

3) Remove and check all ICs for possible folded-over pins. While doing so, use your IC tester to re-check all the TTL (except the 7445 which it does not seem to do correctly)

I sent the TS200 back, going to buy an identical cheaper one. No bent pins.

4) Try all possible combinations of two of the four RAMs in the 'standard' RAM sockets, which on the issue VI are the two closest to the PROMs.

Not doing that yet.

5) Try your other PROMs

Yup ! '---- --' works, but '0000 00' doesn't !.

6) Try another SC/MP.

Not yet.




Arggggh !
Buzby123 is offline   Reply With Quote
Old 8th Dec 2020, 10:06 pm   #145
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 6,529
Default Re: I found it! A very sorry looking MK14.

I listed those checks in 'risk order' hence trying the PROMs and SC/MP last as they are the most unique parts which you wouldn't want damaged by, ie, undetected faulty TTL IC.

So it's looking like you have failure of your new-OS PROMs now? If you want, send just those ones back to me and I'll check them again. If one has now failed I can try to over-programme it (reprogram it with the same code) on the off chance that one of the fuses might have managed to re-grow (apparently that can happen, they have a similar problem to the well known 'tin whisker' disease which commonly affects certain types of germanium transistor).
SiriusHardware is offline   Reply With Quote
Old 8th Dec 2020, 10:46 pm   #146
Buzby123
Hexode
 
Buzby123's Avatar
 
Join Date: Oct 2011
Location: Culcheth, Cheshire, UK.
Posts: 297
Default Re: I found it! A very sorry looking MK14.

Hi Sirius,

I was surprised to find the ROMs had failed, but that was the easiest test to do first.

The 're-grown whisker' problem, is it related to absolute age, or to hours in action ?.

I'll pop them in the post again, just to see if a reprogram fixes it.

One day I'll have a stable Vicky, then I'll move on to Micky.

Cheers,

Buzby

Edit : If you remember, I put labels on the ROMs. These were from a thermal printer, and the '2o' label is discoloured more than the '1o' label. Maybe this is clue to something, but I don't know what. It will be intersting to see which nibble ROM has failed.

Last edited by Buzby123; 8th Dec 2020 at 10:52 pm.
Buzby123 is offline   Reply With Quote
Old 8th Dec 2020, 11:48 pm   #147
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 6,529
Default Re: I found it! A very sorry looking MK14.

To be honest I'm not sure what the failure mode or mechanism is.

If one of them appears faulty there would be nothing to lose by trying to over-programme it to see if it recovers. You do, as you say, seem staggeringly unlucky sometimes - worked fine for me here and then for you for the space of just one evening.

There is one relatively high-risk stunt you could try, the 'rap' method which is sometimes used to clear tin whiskers in germanium transistors. Put each IC in turn at one end of a longish piece of IC tubing (with a bung in the tube at the same end), hold the tube by the other end and rap the IC end of the tube smartly on a bench top. This supposes that any regrown whisker will be much more fragile than the links which were never intentionally blown, and will break open.

The risk is that too much shock force may break one of the 'good' fusible links, and if you do manage to shatter the (assumed) regrown, unwanted link there is no telling where the remnants of it could end up.

Using a programmer to try to re-blow a suspected regrown link is a much more targeted and less stressful method.
SiriusHardware is offline   Reply With Quote
Old 9th Dec 2020, 12:37 am   #148
Buzby123
Hexode
 
Buzby123's Avatar
 
Join Date: Oct 2011
Location: Culcheth, Cheshire, UK.
Posts: 297
Default Re: I found it! A very sorry looking MK14.

Hmm ?. I don't think I'll try that, knowing my luck I'll probably break the table !.

I don't know the detailed structure of a fusible PROM, but from reading how the programming machines work ( multiple little shots at the same fuse, not one big shot ) it does look like there is a possibilty that a re-program might work.

Although, knowing my luck, it's possible a fuse has gone open by itself. No chance of recovering that.

Before you re-program, maybe take a copy of the faulty code and see if you can deduce which bit(s) have gone wrong.

At a very long shot, it might be possible to re-program a slightly different, but functional code, by blowing existing fuses to make NOPs or a GOTO to skip the bad code. It would lose some functionality, but maybe nothing that would matter short term.

As it is, as long as my '---- --' ROMs keep working, I've got enough to start on Micky.

Cheers,

Buzby
Buzby123 is offline   Reply With Quote
Old 9th Dec 2020, 2:13 am   #149
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 6,529
Default Re: I found it! A very sorry looking MK14.

Yes, if the IC in question is not completely duff I'll try to see where it differs from what it should be. Will let you know when they arrive and what I find.
SiriusHardware is offline   Reply With Quote
Old 9th Dec 2020, 2:27 pm   #150
Buzby123
Hexode
 
Buzby123's Avatar
 
Join Date: Oct 2011
Location: Culcheth, Cheshire, UK.
Posts: 297
Default Re: I found it! A very sorry looking MK14.

It's a good job you took file copies before this failure !.

Can never have too many copies of these old ROMs, just can't tell when they might come in useful.
Buzby123 is offline   Reply With Quote
Old 9th Dec 2020, 5:03 pm   #151
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 6,529
Default Re: I found it! A very sorry looking MK14.

I actually already had file copies of the 'new' OS PROMs read from my own SOC supplied originals quite a few years ago now - but until I had your PROMs the first time a few days ago, I had never had file copies of the 'Old' OS read directly from SOC supplied PROMs.

I do now though, and as you say, they are well worth having just in case.
SiriusHardware is offline   Reply With Quote
Old 9th Dec 2020, 11:41 pm   #152
Buzby123
Hexode
 
Buzby123's Avatar
 
Join Date: Oct 2011
Location: Culcheth, Cheshire, UK.
Posts: 297
Default Re: I found it! A very sorry looking MK14.

I'm hoping this weekend to put the Vicky board into the Micky cabinet, so I can test my peripherals.

First up will be the display. I don't evisage any incompatible connector issues here, as the bubble LED is working fine, so the layout of the LED connector must be the same as a real MK14.

But what about the keyboard connector, is that the same ?.

I know I'm going to have to tread carefully around this connector, as I can see I've cut a track and soldered a link on Micky, but I think they're both something to do with the reset button. ( I seem to remember the SoC instruction text for removing ROM images also did something with the reset circuit. It might be that. )

So, apart from the GND and reset pads ( the first two near the reset button ), are the other pads in the same order as a real MK14 ?.

Cheers,

Buzby
Attached Thumbnails
Click image for larger version

Name:	Reset_btn_mod.jpg
Views:	30
Size:	43.5 KB
ID:	222241  
Buzby123 is offline   Reply With Quote
Old 10th Dec 2020, 8:58 am   #153
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 6,529
Default Re: I found it! A very sorry looking MK14.

The mods for removing the unwanted PROM images 'stole' one of the two inverters which were originally in series on the input of the reset circuit, so the reset -switch- circuit is then modified to take the reset-circuit input high when pressed rather than low.

The issue VI keypad connector is the same as on all original MK14s, thankfully, except that one of the two 0V fingers at the top end has been repurposed as an external reset input. Since external keypads don't usually use those top two connections, that would not usually be a problem but maybe one of your many add-ons picked up a common 0V from there?
SiriusHardware is offline   Reply With Quote
Old 10th Dec 2020, 9:52 am   #154
Slothie
Heptode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 825
Default Re: I found it! A very sorry looking MK14.

I just thought I'd clarify that on my issue Vi the external reset connection on the keyboard connector is active low, i.e. pull to 0v to reset - the reset circuitlry is the same as issue I-IV and different to issue V.
Slothie is offline   Reply With Quote
Old 10th Dec 2020, 11:18 am   #155
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 6,529
Default Re: I found it! A very sorry looking MK14.

Yes, sorry, I was describing the likely modified reset setup on Buz's 'Micky' PCB. Of all the original and replica PCBs, only the original issue V and the Martin L replica have 'active high' reset input circuits by default. Earlier issue PCBs which were modified to remove the unwanted PROM images (like Micky) will also have active-high reset inputs.

The issue V came with this modification already implemented on the PCB.
SiriusHardware is offline   Reply With Quote
Old 10th Dec 2020, 6:27 pm   #156
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 6,529
Default Re: I found it! A very sorry looking MK14.

Buz's new-OS PROMs have arrived back and I'm just about to have a look at those - as he pointed out, the thermal label on one IC looks rather dark, and I don't remember having noticed that when I last had them. Attached, a photo I took of the PROMs working in my issue VI the first time they were here a week ago. There's no sign of heat related darkening on either PROM label at that time.

Even when working normally these devices certainly are hot-heads, when I programme pairs for other people I always place the blue and green sticky dots at one end of the devices unlike the SOC practice of putting them in the centre, right over the hot core of the IC. They don't need anything to make them run hotter than they already do.

In other news, my 4 * AM9111s arrived from the source in Germany - I've done a quick test on them all and while three work, one is unfortunately faulty. They all look absolutely brand new, legs still splayed suggesting that they had never been in a socket before I put them in my tester.

I'll get back to the supplier later to see how they handle returns. I doubt whether they have the means to test these themselves and I'm sure they were supplied in good faith.
Attached Thumbnails
Click image for larger version

Name:	Buzby_New_OS_PROMs_FirstTime.jpg
Views:	63
Size:	66.4 KB
ID:	222278  

Last edited by SiriusHardware; 10th Dec 2020 at 6:38 pm.
SiriusHardware is offline   Reply With Quote
Old 10th Dec 2020, 7:31 pm   #157
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 6,529
Default Re: I found it! A very sorry looking MK14.

...Not good news on Buzby's new-OS PROMs unfortunately.

It turns out that the 'darkened' IC ('2o') is actually the good one, that reads / verifies perfectly against the code file I have here. But when I read '1o', the first 256 nibbles read correctly but every address from 100-1FF reads as 'F', so that is looking like a failure of the internal address decoding logic to me. I've done basic checks like resistances of all pins to the VCC and 0V pins but A8 on the dud IC reads no differently to pins A0-A7.

The only noticeable difference is that where '2o' has a resistance of >6M between its supply pins, '1o' has a resistance of 171K.

If it had been a case of finding a single bit stuck low somewhere I would have tried overprogramming but I don't think there will be any benefit to doing it in this case, where clearly the whole upper half of the IC has gone Missing In Action.
SiriusHardware is offline   Reply With Quote
Old 11th Dec 2020, 10:18 am   #158
Buzby123
Hexode
 
Buzby123's Avatar
 
Join Date: Oct 2011
Location: Culcheth, Cheshire, UK.
Posts: 297
Default Re: I found it! A very sorry looking MK14.

Quote:
Originally Posted by SiriusHardware View Post
...Not good news on Buzby's new-OS PROMs unfortunately.
As I said, the only luck I have is bad luck !.

I'm not too bothered at the moment. Now that Vicky is running, even with 'old' ROMs, I've got a basis to start work on Micky.

Hoping to get started working on it tomorrow, I'll keep you informed.

Cheers,

Buzby
Buzby123 is offline   Reply With Quote
Old 11th Dec 2020, 11:26 am   #159
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 6,529
Default Re: I found it! A very sorry looking MK14.

Well, the good(?) news is that only one of two is duff, so if you can source a single replacement Nat Semi DM74S571 and get it sent to me I will programme it for you. That will probably be cheaper for you than buying a pre-programmed pair. Just try to avoid any which are suspiciously cheap, as they may well already be programmed.

I've also had a response from the German seller of the AM9111s, they are sending me a replacement for the one dud IC out of the four I received.
SiriusHardware is offline   Reply With Quote
Old 17th Dec 2020, 10:34 am   #160
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 6,529
Default Re: I found it! A very sorry looking MK14.

A quick update on the AM9111s, the replacement for the dud IC has now arrived and is OK. The supplier was very civilised about it so I'd be more than happy to use them again.
SiriusHardware is offline   Reply With Quote
Reply

Thread Tools



All times are GMT. The time now is 12:58 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.