UK Vintage Radio Repair and Restoration Discussion Forum

UK Vintage Radio Repair and Restoration Discussion Forum (https://www.vintage-radio.net/forum/index.php)
-   Vintage Computers (https://www.vintage-radio.net/forum/forumdisplay.php?f=16)
-   -   I found it! A very sorry looking MK14. (https://www.vintage-radio.net/forum/showthread.php?t=170727)

SiriusHardware 7th Nov 2020 11:55 pm

Re: I found it! A very sorry looking MK14.
 
Having downloaded the image and viewed it in a viewer I would say it is a 7442. The number is printed so close to the bottom edge that the lower bar of the '2' is lost but the middle section of the 4th character has a distinct 's' curve which the first character (7) does not, so I think it is a '2'.

Buzby123 8th Nov 2020 1:33 am

Re: I found it! A very sorry looking MK14.
 
2 Attachment(s)
Hello all,

Sorry, not had a chance today to even power it up.

Yes, the chip is a 7442. Not sure if that was original from SoC, or if I put it there. Some of the other chips are definatley not SoC, 'cos they've got RS part numbers !.

When the chips were in Micky there was my off-board display, with transistors and FND500's I think, so maybe the on-board chips were not so important then.

Regarding the display, is it the correct one, and have I wired it correctly ?.

The PCBs, both original and replica, have two empty holes at the left end of the display connector. ( I used these forty years ago to supply power to my off-board display. )

At least with this PCB I'm not debugging the wiring, just the chips.

Hopefully I'll get some time tomorrow.

Cheers,

Buzby

Mark1960 8th Nov 2020 1:48 am

Re: I found it! A very sorry looking MK14.
 
That looks like the correct display and the connections on the display look correct.

Connections on the main pcb should be to the rightmost 16 pins, two on the left with no connection.

SiriusHardware 8th Nov 2020 2:23 am

Re: I found it! A very sorry looking MK14.
 
1 Attachment(s)
This has been posted before but here it is again for convenience. If you have a DMM with a 'diode test' feature you can use that to quickly test out the display, it will provide just enough enough power to light each display segment (one at a time) dimly.

That display certainly 'looks right', very similar to my original MK14 display.

circuitryboy 8th Nov 2020 9:04 am

Re: I found it! A very sorry looking MK14.
 
I'd forgotten you'd had extra circuit/big display - 7442 would be fine for that.
The R->L scan is not 'linear'. Varying brightness and some segment h activity.
From the right, are digits 3 and 7 completely off?
Does swapping IC9&IC10 or IC14&15 change the display?

Slothie 8th Nov 2020 10:35 am

Re: I found it! A very sorry looking MK14.
 
According to the data sheet the 7442 has normal totem-pole outputs which have a short-circuit current of 18-55mA as apposed to the 7445's open collector outputs with 80mA sink capacity. This would imply that the digits will be limited to this level, which would cause the digits to vary in brightness depending on how many segments are lit. The 2 chips are pin compatible, but the design really needs the open collectot outputs of the 7445. That said, it should work to some degree and so the symptoms you are experiencing are probably something else. Do the chip legs just need a clean?

SiriusHardware 8th Nov 2020 11:14 am

Re: I found it! A very sorry looking MK14.
 
I can see why, when you were driving extended / external display hardware, you changed that IC from an open collector type to a TTL-output type but for the 'classic' MK14 display setup you probably need to change it to a 7445 - note this was an 'original' 7445 and not 74xx45.

That still won't be the root cause of the problem though.

Buzby123 8th Nov 2020 11:30 pm

Re: I found it! A very sorry looking MK14.
 
Well, I've spent the last few hours checking the address decoding.

I removed ROMs, RAMs, and CPU. Put wires from a breadboard micro to drive A11-A8 and WDS/RDS.

Wrote code to wiggle pins cyclicly at a few hundred Hz, and prodded my scope where the decodes should come out. They all looked OK. ( I didn't actually check the addresses, just that the CE pins wiggled up and down nicely. )

So it looks like ROMs and/or ROMs are faulty. I've got four 2111s, and swapping them in all combinations made no difference at all. Using 'old' ROMs gives a different wrong display.

Tomorrow I'll rig up a RAM tester, and a ROM dumper.

I'm assuming those of you with VI boards just plugged all the bits in and it worked straight off ?.

Cheers,

Buzby

SiriusHardware 9th Nov 2020 12:02 am

Re: I found it! A very sorry looking MK14.
 
Yes, that was certainly my experience, I put the parts in and it 'just worked'. My board and Tim's both came from a single batch of 5 that I had made by uploading the unmodified project files from Slothie's post early on in the 'vdu' thread to JLCPCB.

It's impossible to know whether whoever had your board made carried out any changes but why would they when they know, from reading the relevant thread here, that the boards work exactly as designed. If you are spectacularly unlucky you could have an etch fault or some other manufacturing fault on your PCB.

To read out the PROMs, if you have an EPROM programmer make up a jig which places the two PROMs side by side as D0-D3 and D4-D7, commons up the A0-A8 pins, _CE pins and power pins on both ICs, maps them onto the matching pins on the footprint of a 2716 EPROM and drop the 'jig' into your EPROM programmer. On the programmer, choose 2716 and then limit the address range to 0000-01FF and read the contents out of the '2716'.

One other issue we have been made aware of somewhere in these threads, and this is not just relevant to the issue VI - some versions of 74LS157 don't work in the basic MK14, apparently even in original machines, so if those RS brand ICs you mentioned happen to include the 74LS157s then maybe try a pair of other-brand 74LS157s. I didn't bring this up before because it seemed you were using parts from a once working MK14 but since you seem to have several non-original parts in there this may be worth looking at.

Originally all the chips supplied in MK14 kits were National Semiconductor (DM or MM prefix) but a few other brands were eventually substituted for some of the chips - notably Texas (SN prefix).

In my issue VI I have a mixture of DM and SN prefix types, the 74LS157s are SN (Texas) types in mine.

Elsewhere in these threads there are posted Arduino Uno sketches for (a) reading out the contents of a 512-byte ROM and (b) doing basic functional tests on 2111 RAMs. If you don't already have a specific way to test the PROMS and RAM I'll try to track those posts down for you.

SiriusHardware 9th Nov 2020 12:39 am

Re: I found it! A very sorry looking MK14.
 
Looking again at the image buz posted earlier it looks as though the 157s have the original NS 'Wave' logo on them, so they are likely to be DM prefix - should be OK.

Slothie 9th Nov 2020 12:43 am

Re: I found it! A very sorry looking MK14.
 
I used Texas SN74LS157's in my original prototype which worked correctly.

SiriusHardware 9th Nov 2020 1:00 am

Re: I found it! A very sorry looking MK14.
 
Arduino Uno based 2111 tester in post #403 of this thread:-

https://vintage-radio.net/forum/show...145663&page=21

I still haven't fathomed out Slothie's trick of posting a link to the exact post on the page.

Mark1960 9th Nov 2020 1:14 am

Re: I found it! A very sorry looking MK14.
 
Quote:

Originally Posted by SiriusHardware (Post 1308476)
Arduino Uno based 2111 tester in post #403 of this thread:-

https://vintage-radio.net/forum/show...145663&page=21

I still haven't fathomed out Slothie's trick of posting a link to the exact post on the page.

I think the trick is to click on the number at the top right of the post, opens into a new window with only that post, then copy the internet address of the post.

https://vintage-radio.net/forum/show...&postcount=403

Slothie 9th Nov 2020 1:41 am

Re: I found it! A very sorry looking MK14.
 
Quote:

Originally Posted by SiriusHardware (Post 1308476)
Arduino Uno based 2111 tester in post #403 of this thread:-

https://vintage-radio.net/forum/show...145663&page=21

I still haven't fathomed out Slothie's trick of posting a link to the exact post on the page.

Right click on the number at the top right of the post and select "copy link address"

SiriusHardware 9th Nov 2020 1:55 am

Re: I found it! A very sorry looking MK14.
 
Got that. :)

circuitryboy 9th Nov 2020 3:15 pm

Re: I found it! A very sorry looking MK14.
 
Have you 'scoped the A3-A0 inputs to IC12 and all o/ps from IC13? If these look OK then the KYBD routine in either Monitor is executing correctly.

Buzby123 10th Nov 2020 3:04 pm

Re: I found it! A very sorry looking MK14.
 
It gets better, or worse, depending on your view !.

I didn't give much attention to the display driving chips, as all segs and digits were active at sometime during testing.

Today I received a TS200 chip tester, and it showed my 7442 is bad. ( It also showed my other 7442's are bad as well. How unlucky is that ?. )

So, without 100% confidence in the display mechanism, I am going to have to get another 7442 or 7445 before anymore testing.

Cheers,

Buzby

Mark1960 10th Nov 2020 3:34 pm

Re: I found it! A very sorry looking MK14.
 
Quote:

Originally Posted by Buzby123 (Post 1309015)
Today I received a TS200 chip tester, and it showed my 7442 is bad. ( It also showed my other 7442's are bad as well. How unlucky is that ?. )

How do you know if the TS200 is working correctly?

Buzby123 10th Nov 2020 4:41 pm

Re: I found it! A very sorry looking MK14.
 
That's a good question !.

I've got three 7442's, all test as BAD, and all behave same in Vicky.

I tested a few other of my stock of chips, simple gates mostly, and found a faulty 7404. Other 7404s test OK, so I know the TS200 can tell the difference in simple gates.

It does seem strange that three 7442s all tested bad, so I've not binned them yet !.

Cheers,

Buzby

SiriusHardware 10th Nov 2020 5:16 pm

Re: I found it! A very sorry looking MK14.
 
Issue...VIcky. I think I see what you did there.

My partner has your habit of giving names to gadgets. I have to keep myself constantly updated otherwise I have no idea who or what she's talking about when she asks where 'Serena' is.

I too find it suspicious that your chip tester is failing your entire stock of 7442s and to be honest the job of that chip is to connect the cathode of each display cell to 0V in turn, which it did seem to be doing (the displays were being scanned).

At the same time, the segment-driving hardware should be outputting a pattern specific to each display cell as the 7442 / 7445 steps through them, and that's the bit which does not appear to be happening.

You were always going to have to fit a 7445 eventually, so you might as well get that over with.


All times are GMT +1. The time now is 9:32 pm.

Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.