UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio 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 17th Sep 2020, 10:41 am   #41
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 4,997
Default Re: I found it !. A very sorry looking MK14

You're referring to the missing 'interconnection' blob on the junction of Read Data / IC7 OE / IC17 input, I take it. Yes, quite poor.

Also the misleading way in which the connections for the external keypad connector are shown, as they do not arrive at the edge connector in the order that the circuit diagram suggests they do. The addition of some pin numbers to the drawing of the keypad connector would have been very helpful.
SiriusHardware is online now   Reply With Quote
Old 17th Sep 2020, 11:37 am   #42
Slothie
Heptode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 573
Default Re: I found it !. A very sorry looking MK14

I "corrected" that blob without really noticing it because I had already decided the schematic was more a guide than a definitive reference One version of the schematic I used had the numbers for the common cathodes of the display backwards and guess which one I followed on my initial prototype! So when I finally got my prototype working the display was fully working and backwards! Much track cutting and bodge wire fitting followed that one.
All things considered its a wonder I got the keyboard connector correct as I seem to have done. I think its because by then I was using an issue V schematic I'd found and confirmed it with posts I'd seen on here (possibly your MK14 uploader device). Because the schematic was so vague I had to go looking elsewhere...
Slothie is offline   Reply With Quote
Old 17th Sep 2020, 11:55 am   #43
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 4,997
Default Re: I found it !. A very sorry looking MK14

I am fairly convinced that this was why the JMP replica PCB ended up with a different (and arguably more logical) order of keypad connections. If the originator didn't have an original PCB to trace from it would have been perfectly reasonable to take the diagram at its word.

But yes, you got the keypad connector right, as you did with so many other things.
SiriusHardware is online now   Reply With Quote
Old 17th Sep 2020, 1:13 pm   #44
Buzby123
Pentode
 
Buzby123's Avatar
 
Join Date: Oct 2011
Location: Culcheth, Cheshire, UK.
Posts: 115
Default Re: I found it !. A very sorry looking MK14

I'm beginning to think it would be easier to design and build a new MK14. This hacked board of mine is giving me grief.

I've taken the 8154 out, and swapped the ceramic 8060 for a plastic one, just in case I make a slip.

Powered up, drawing 0.42A, looks OK. Clock is running, 8060 data lines all wiggling about. DB0-3 solid 0-5v, but DB4-7 show some 'half-way' levels. Not too worried about them yet.

Most address lines on 8060 look reasonable, but AD09,10,11 don't look happy at all.

I now need to find where these go. The SOC drawing is no use, because I completely re-hashed the address decoding, and most of it is now on that verowired daughter board !.

Wish me luck !.

Cheers,

Buzby
Attached Thumbnails
Click image for larger version

Name:	DS0002.jpg
Views:	26
Size:	18.8 KB
ID:	215835   Click image for larger version

Name:	DS0003.jpg
Views:	22
Size:	18.1 KB
ID:	215836   Click image for larger version

Name:	DS0004.jpg
Views:	23
Size:	18.1 KB
ID:	215837  
Buzby123 is online now   Reply With Quote
Old 17th Sep 2020, 1:57 pm   #45
Slothie
Heptode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 573
Default Re: I found it !. A very sorry looking MK14

Its worth noting that on the MK14 only D0..D3 have pullups to +5v, so when the data bus goes tristate D0..D3 will go to +5v but D4..D7 will do whatever they like. Something else on the bus may be pulling these lines weakly to odd levels, or capacitive coupling or........
Slothie is offline   Reply With Quote
Old 17th Sep 2020, 2:15 pm   #46
Buzby123
Pentode
 
Buzby123's Avatar
 
Join Date: Oct 2011
Location: Culcheth, Cheshire, UK.
Posts: 115
Default Re: I found it !. A very sorry looking MK14

Yes, that's why I'm not too bothered about the wobbly signals on D4-7.

But the AD09-11 pins should only have hi/lo, and that's what I'm going to try and sort out before going any further.

Unlikely I'll be doing any more today, other duties call.

Cheers,

Buzby
Buzby123 is online now   Reply With Quote
Old 17th Sep 2020, 2:34 pm   #47
Slothie
Heptode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 573
Default Re: I found it !. A very sorry looking MK14

Disclaimer: I havent actually probed an MK14 with an oscilloscope recently
According to the data sheet the address lines are only driven while NBREQ is low, which is from just before NADS goes low to just after NRDS/NWDS goes from low-to-high; that means the address bus can be floating from (just after) the positive edge of NRDS/NWDS to (just before) the falling edge of the NADS.

Anyhow, its worth investigating and reverse-engineering your veroboard just to know what its doing!
Attached Thumbnails
Click image for larger version

Name:	Screenshot_2020-09-17_14-32-43.jpg
Views:	19
Size:	62.4 KB
ID:	215844  
Slothie is offline   Reply With Quote
Old 17th Sep 2020, 11:32 pm   #48
Buzby123
Pentode
 
Buzby123's Avatar
 
Join Date: Oct 2011
Location: Culcheth, Cheshire, UK.
Posts: 115
Default Re: I found it !. A very sorry looking MK14

I didn't realise that the address pins were occasionally 'floating', so it seems like my screenshots might not be as bad as what they look like.

But are all the address lines being driven at any time ?. I don't what areas the code is accessing.

What I need now is the gadget I built a hundred years ago, the 'NOPulator' !.

This was a gadget I built when I was working for an arcade machine company, fault finding in things like Space Invaders and pinball machines.

Finding faults in the address decoders was a PITA, until I built the NOPulator.

This was a Z80 CPU mounted on a 40 pin DIL, which plugged in where the original Z80 should be.

The Z80 on the NOPulator had it's data pins 'bent up' and they were hard-wired as a NOP. This made the Z80 run through the whole address range repeatedly, executing NOPs at every memory location.

A tiny switch gated the RDstrobe from the Z80 to either RD or WR on the socket, so the NOPulator would either read or write the whole memory range.

What I need now is a NOPulator based on an 8060.
Buzby123 is online now   Reply With Quote
Old 18th Sep 2020, 12:14 am   #49
Mark1960
Pentode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 125
Default Re: I found it !. A very sorry looking MK14

Rather than bend pins on an ins8060, try removing the proms and ram, then add pullups for the data lines that donít already have them. Should run CAD instruction using auto index P3 which should then cycle through memory, every third access should read from an out of sequence location but maybe not too hard to check the pattern.
Mark1960 is offline   Reply With Quote
Old 18th Sep 2020, 12:20 am   #50
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 4,997
Default Re: I found it !. A very sorry looking MK14

The hex code for NOP in SC/MP is 0x08, so in theory you can make something which will apply that hex code to the data bus all the time or just during _NRDS and the CPU should just sweep majestically through the address range executing all the NOPs.

You'd need to remove anything else which might try to place data on the bus in response to a read from any part of the address range - RAMs, PROM, etc.

If you make a breakout adaptor from a standard EPROM or EEPROM pinout to 2 * 16 pin headers to plug into the PROM sockets you can try writing little bits of test code to exercise various parts of the circuit to see if they behave as expected.

Like this: I have this one programmed with the 'Old' OS and 'New' OS, selectable, so I can run tests with either OS. I used ordinary 16-pin IC sockets as the 16-pin 'header plugs' which plug into the PROM sockets.

Edit: Cross posted with Mark.
Attached Thumbnails
Click image for larger version

Name:	eprom_adaptor_s.jpg
Views:	24
Size:	139.1 KB
ID:	215894  
SiriusHardware is online now   Reply With Quote
Old 18th Sep 2020, 12:39 am   #51
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 4,997
Default Re: I found it !. A very sorry looking MK14

Expanding on your own idea, do you have access to other micro toys such as an Arduino Mega? If you removed the SC/MP and connected a Mega to the CPU socket with a header you could write sketches to put any combination of address or scanned address range and RD or WD on the buses at any human-visible speed or static state you like, in order to see what is going on.
SiriusHardware is online now   Reply With Quote
Old 23rd Sep 2020, 12:12 am   #52
Buzby123
Pentode
 
Buzby123's Avatar
 
Join Date: Oct 2011
Location: Culcheth, Cheshire, UK.
Posts: 115
Default Re: I found it !. A very sorry looking MK14

I just thought I'd keep you updated on my progress with Micky.

First I traced the ribbon cables to the 32 pin plug, then lots of continuity testing to determine what the daughter board is doing.

It looks like the daughter board does the address decoding for :
0000-01FF : SCIOS ROM on main board ( which I'm fairly certain is the 'new' version. )
0200-05FF : 1K of battery-backed RAM on the daughter board.
0600-07FF : SCIOS ROM on daughter board ( probably the original version. )
? ?? ? : Optional RAM on main board.
? ?? ? : 8154 on main board.

The ? are because I think I've found the reason Micky was abandoned all those years ago. One of the signals for these decodes just connects to fresh air.

Don't laugh, this was my first experience with a microprocessor. I remember having the daughter board working completely ( even though the decode logic is nothing to be proud of ), but I must have done some more mods on the main board and didn't realise that a track cut left a wire floating.

Next job is to find where the floating wire should go, then fix some switches to wiggle the address lines and check the decoding.

A few more hours work, and Micky should be running again.

Cheers,

Buzby
Buzby123 is online now   Reply With Quote
Old 23rd Sep 2020, 12:25 am   #53
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 4,997
Default Re: I found it !. A very sorry looking MK14

Good luck with the ongoing repair, we are all hoping to see it come back to life.
SiriusHardware is online now   Reply With Quote
Old 23rd Sep 2020, 10:42 am   #54
Slothie
Heptode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 573
Default Re: I found it !. A very sorry looking MK14

The battery backed RAM is a great idea. I was going to do that on my 1.5k RAM expansion since the variant of 6116 I had was an IDT part designed to be backed up by a battery.
Slothie is offline   Reply With Quote
Old 23rd Sep 2020, 6:06 pm   #55
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 4,997
Default Re: I found it !. A very sorry looking MK14

My 'Maplin Z80 CPU' PCB used 6116 RAMs of any flavour and had the option to battery-back the RAM, so the principle may be applicable to any type of 6116.
SiriusHardware is online now   Reply With Quote
Old 23rd Sep 2020, 7:46 pm   #56
Slothie
Heptode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 573
Default Re: I found it !. A very sorry looking MK14

Quote:
Originally Posted by SiriusHardware View Post
My 'Maplin Z80 CPU' PCB used 6116 RAMs of any flavour and had the option to battery-back the RAM, so the principle may be applicable to any type of 6116.
Yes I think its just some are designed to use a very low current in standby (1mA or so) whereas standard parts use 10mA or higher when not selected.
Slothie is offline   Reply With Quote
Old 11th Oct 2020, 2:39 pm   #57
Buzby123
Pentode
 
Buzby123's Avatar
 
Join Date: Oct 2011
Location: Culcheth, Cheshire, UK.
Posts: 115
Default Re: I found it !. A very sorry looking MK14

I'm still here !.

Not had much time these last few weeks, but hope to get back on it soon.
Buzby123 is online now   Reply With Quote
Old 11th Oct 2020, 3:51 pm   #58
Timbucus
Heptode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 676
Default Re: I found it !. A very sorry looking MK14

That will be good still looking forward to seeing it running fully - in these times lack of updates sometimes leads to concern
Timbucus is online now   Reply With Quote
Old 3rd Nov 2020, 6:56 pm   #59
Buzby123
Pentode
 
Buzby123's Avatar
 
Join Date: Oct 2011
Location: Culcheth, Cheshire, UK.
Posts: 115
Default Re: I found it !. A very sorry looking MK14

I'm still working on Micky, there is something seriously difficult to track down.

However, I got a copy of a Slothie VI PCB, so I'm going to use that as a testbed for the components from Micky.

Last edited by Cobaltblue; 3rd Nov 2020 at 8:20 pm. Reason: of course you can mention eBay you can't pretend not to be mentioning it
Buzby123 is online now   Reply With Quote
Old 3rd Nov 2020, 7:28 pm   #60
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 4,997
Default Re: I found it !. A very sorry looking MK14

Good idea to acquire one of the issue VI PCBs to use as a test bed for the ICs from your original system. Hopefully you can be a third eye on the VDU project at some point as well. There is an extra IC in the issue VI, a 74LS20 (U20) so make sure you have one of those available as well.

The IC north-west of the display with the IC 'U' number and the device number tangled up on the screen printing is either an 80L95 (as per original) or can be substituted with a 74LS365. I think Tim has tried one or two other variants in that position as well.

The two 7408s behind the display have to be 'original' 7408s and not 74(something)08. It's OK to use a 74LS08 in the U16 position, as indicated by the screen printing.
SiriusHardware is online now   Reply With Quote
Reply

Thread Tools



All times are GMT +1. The time now is 8:04 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 - 2020, vBulletin Solutions, Inc.
Copyright ©2002 - 2020, Paul Stenning.