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 13th Apr 2018, 12:40 pm   #1
Slothie
Triode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 12
Question MK14 schematic revisions

I'm in the process of making a replica MK14 (I know its been done before, but I've always wanted one and I'm using the exercise to learn/practice how to use Kicad). I've been comparing the schematics for the V1 and V5 boards I found on the interweb, and the main difference seems to be the circuit driving the PROM enable pin. Originally it was tied to A11, but now it's using the once spare NAND gate of IC17 to NAND the "RDS or WDS" line with the inverted A11 signal. This means that a Write to PROM space will enable the PROM outputs causing both the SC/MP and the PROM to drive the data bus.

OK, so before any time the A11 line was 0 this happened too, so I don't see how this is an improvement. I could understand if they gated with the RDS only, or am I missing something obvious?

Anyhow as a newbie on the forum I'd like to say hi and I'm enjoying reading through posts here!

Slothie
Slothie is offline   Reply With Quote
Old 13th Apr 2018, 5:37 pm   #2
SiriusHardware
Octode
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 1,679
Default Re: MK14 schematic revisions

Do you have a link to the late version of the diagram? I'm only familiar with the older version.

From what you say it sounds as though the mod partly addresses an earlier oversight, in so far as the PROMs should never have been activated solely when A11 was in one state or another, it should always have involved at least the RDS signal, just as you say.

It is slightly baffling that the design allows a write to PROM space which ends up activating the PROM outputs. Maybe the thinking was just that nobody would ever try to write to a PROM address, as that would be an illogical (!) thing to do.

I've read somewhere recently that the late(st?) issue MK14 board included some kind of improvement to the address decoding - On early through middle issues there are multiple images of the RAM, the PROMs, etc throughout the address range because the address decoding is only partial. In particular, the PROMs appear no less than 4 times in the first part of the address range.

So maybe this change - adding WDS and RDS to the enable signal for the PROM block - is used in conjunction with a change somewhere else which stops the PROMs from appearing as multiple images throughout the first 2K of the address range and frees up some of that range so that more RAM can be mapped into it - in which case it does make sense to have the WDS signal involved as well, as the chip select for RAM will obviously need to be activated by a write signal as well as a read signal.

Last edited by SiriusHardware; 13th Apr 2018 at 5:48 pm.
SiriusHardware is offline   Reply With Quote
Old 14th Apr 2018, 3:22 am   #3
Slothie
Triode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 12
Default Re: MK14 schematic revisions

I can't find where I got it from but heres a link to an image hosting site (its just too big to attach to a post!)....
https://preview.ibb.co/ifKgZ7/mk14_v5.jpg
Slothie is offline   Reply With Quote
Old 14th Apr 2018, 9:44 am   #4
SiriusHardware
Octode
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 1,679
Default Re: MK14 schematic revisions

...I'm away from base just at the moment, but just before I came away I had a hunt through my archive of MK14 related stuff and found a single sheet document, sent at my request from S.O.C. describing fairly extensive modifications for early issue MK14s like mine, the purpose of which were to remove the unwanted PROM images from the address range 0200H to 07FFH.

If these mods were done, it was then possible for the user to map 1.5K of added-on offboard memory into the address block no longer occupied by the PROM.

On reflection, this is probably one of the rarest original MK14 documents around, so I'll scan it and post it here - my scanner is so old that it only works with Win XP, so I'll have to drag out another semi antique in order to do that.

My thought is that the late(est) issue of MK14 PCB may actually have incorporated these mods. The simplest way to find out is for someone with an issue V machine to have a look through 0200H-07FFH and see if there are PROM images present there.
SiriusHardware is offline   Reply With Quote
Old 14th Apr 2018, 4:40 pm   #5
Slothie
Triode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 12
Smile Re: MK14 schematic revisions

Quote:
Originally Posted by SiriusHardware View Post
...On reflection, this is probably one of the rarest original MK14 documents around, so I'll scan it and post it here - my scanner is so old that it only works with Win XP, so I'll have to drag out another semi antique in order to do that..
I'd certainly like to see that when you get the time!

I think you are right that this difference is to prevent the PROM from being enabled every time A11 goes low, possibly because it might interfere with the VDU board, and they used the "read or write" line because it was there and the inverted RDS wasn't! I also noticed that the two spare AND gates are marked "These gates are used on rev 5 boards" so maybe the schematic is of rev 4 and the 2 gates were used to more completely decode the PROM address as you described. Its odd they didn't show them on the schematic when they added this remark but it was Science of Cambridge after all

I think that clears up the mystery for me at the moment but as I said I'd love to see that letter when you have time.

Ian aka Slothie
Slothie is offline   Reply With Quote
Old 16th Apr 2018, 12:07 am   #6
SiriusHardware
Octode
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 1,679
Default Re: MK14 schematic revisions

Hi Slothie - thanks for the link to the later version of the MK14 diagram, which is a new one for my collection. I didn't see your post #3 until after I had posted #4, due to your posts currently being delayed by new member moderation.

So here it is (Merry Christmas...) a scanned copy of Science Of Cambridge's official modification to early issue MK14s to remove the unwanted PROM images from the 0200-07FF address range.

I think the background to this was that I had written to them asking for details of how to expand the memory beyond its already maxed out 640 bytes, and they sent me this sheet with the handwritten comment scribbled across the top.

The first drawback (for anyone hoping to incorporate these mods into a replica) is that they are not shown in circuit diagram form, so reverse engineering the final diagram of the PCB as modified will involve looking at these physical mods and applying them to the early version MK14 circuit diagram until you have, in effect, the issue V circuit diagram.

The document does not go so far as to describe how to fit extra memory, it only describes how to free up some of the memory range so that extra RAM could be mapped into that range.

I never actually applied these mods to my own MK14: Intead, I uninstalled the VDU which was chewing up nearly all of the memory.
Attached Files
File Type: pdf MK14_EarlyIssue_AddressDecoderMods.PDF (30.7 KB, 17 views)
SiriusHardware is offline   Reply With Quote
Old 16th Apr 2018, 10:30 am   #7
Slothie
Triode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 12
Default Re: MK14 schematic revisions

An additional difficulty is that SOC changed which gates within packages were used, so the pin numbers in the v2 schematic are not the same as in the note... perhaps they were referring to a v1 board!
However I've worked it out (please excuse the rubbish image editing!)
Basically the reset circuit is reworked to free up a NOT gate and the spare gates used to NAND the inverted A9,10,11 signals with RDS or WDS.
Attached Thumbnails
Click image for larger version

Name:	prom_mods_1.png
Views:	17
Size:	24.9 KB
ID:	161211  
Slothie is offline   Reply With Quote
Old 16th Apr 2018, 12:06 pm   #8
SiriusHardware
Octode
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 1,679
Default Re: MK14 schematic revisions

Fast work

I didn't have time to try to relate the mod info to the issue II diagram but it doesn't surprise me that the gate pin numbering does not match the mod document.

Typical.

S.O.C. were fond of other stunts as well, such as ignoring the D0-D7 and A0-A7 order of the data / address pins on RAM ICs and tracking / using them in whatever order it was most convenient to do so.

This didn't affect the operation of the circuit (it doesn't matter if bit 0 of your data is stored in physical bit 6 of a completely different physical RAM location, as long as it is also recovered from bit 6 of the same RAM location) but it made tracing / faultfinding bus faults tiresome in the extreme.

So, will you now work these changes into your replica?
SiriusHardware is offline   Reply With Quote
Old 16th Apr 2018, 12:31 pm   #9
Slothie
Triode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 12
Default Re: MK14 schematic revisions

Quote:
Originally Posted by SiriusHardware View Post
So, will you now work these changes into your replica?
To be honest I'd worked out how they might have done it whilst not getting to sleep (insomnia isnt all bad!) before I read your document, but it confirmed my thoughts.

I'd like to roll these changes in, as I might as well build a rev 5, but I might need one of the AND gates for the RAM enable line, actual MM2111 ram chips are proving hard to find and the pin-compatible(ish) IM65X62 that I have managed to acquire from Poland requires the enable pin to be pulled low in read and write cycles. Unless I can liberate another of the inverters from the reset logic and invert the 'rds or wds' line, but I'm not sure the SCMP will like a slow rising reset pulse!
I might have to breadboard up a test circuit to check it can be made to work timing wise!
Slothie is offline   Reply With Quote
Old 16th Apr 2018, 1:17 pm   #10
SiriusHardware
Octode
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 1,679
Default Re: MK14 schematic revisions

I see P2111-1 available for $6 each in the USA just now, but the postage is, as always, prohibitive.
SiriusHardware is offline   Reply With Quote
Old 16th Apr 2018, 1:40 pm   #11
Slothie
Triode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 12
Default Re: MK14 schematic revisions

Yes, and he only seems to have one... I need at least a pair

It might just be a waiting game, it depends how patient I feel
Slothie is offline   Reply With Quote
Old 16th Apr 2018, 1:50 pm   #12
SiriusHardware
Octode
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 1,679
Default Re: MK14 schematic revisions

Sorry, didn't realise that source only had one.

I also see 4 * AM9111, second hand and two different speeds. Not sure if these are fully compatible with MM2111 / P2111, although the AM9111 devices often have 'P2111' marked on them alongside the AM9111 number, as if to say 'equivalent to P2111'.
SiriusHardware is offline   Reply With Quote
Old 16th Apr 2018, 1:58 pm   #13
Slothie
Triode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 12
Default Re: MK14 schematic revisions

The AM9111 has only one chip select, so would require the extra logic anyway, and its a 16 pin chip so wouldnt leave the opportunity to make the board compatible with MM2111 chips.
Slothie is offline   Reply With Quote
Old 16th Apr 2018, 2:14 pm   #14
SiriusHardware
Octode
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 1,679
Default Re: MK14 schematic revisions

Sorry again... good luck with the project, anyway.

Edit: the AM9111s I've just been looking at are 18-pin, but I suppose the issue with the CE pins is a deal breaker anyway.

Edit #2 : Also two chip selects on the AM9111 datasheet I have just looked at. Was this chip available in two outlines?

Last edited by SiriusHardware; 16th Apr 2018 at 2:21 pm.
SiriusHardware is offline   Reply With Quote
Old 16th Apr 2018, 2:39 pm   #15
Slothie
Triode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 12
Default Re: MK14 schematic revisions

My fault, I looked at the wrong section in the data book, I saw the outline for the AM9112

Looks like the AM9111 would be fine, just need to get a reasonable price. Thanks for pointing this out!
Slothie is offline   Reply With Quote
Old Yesterday, 1:18 am   #16
Slothie
Triode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 12
Default Re: MK14 schematic revisions

Well I've got the schematic finalised and PCB layout made, and ordered the PCB's! Heres the 3D model kicad generated....
Attached Thumbnails
Click image for larger version

Name:	mk14.jpg
Views:	21
Size:	60.5 KB
ID:	161356  
Slothie is offline   Reply With Quote
Old Today, 12:02 pm   #17
SiriusHardware
Octode
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 1,679
Default Re: MK14 schematic revisions

Nice! It's not so long ago that there were people wistfully wishing that someone would make some MK14 reproduction PCBs and now like buses, they all seem to be coming along at once. (You haven't said that you plan to sell any of your PCBs but at least two other sources recently started to sell replica MK14 PCBs).

Like yours, my (original) MK14 has a male pin-row connector fitted where the display would be and the display has been reconfigured that it can be plugged into / unplugged from that. The flat 0.1" spaced 'stiff' ribbon cable originally used for the display seems quite hard to come by now in any case.

I think I would have allowed myself the luxury of nicer keyswitches, but I can tell you that any kind of keyswitch at all is a huge improvement over anything which S.O.C. originally fitted to the MK14. That's why so few original MK14s have their original keypads.
SiriusHardware is offline   Reply With Quote
Reply

Thread Tools



All times are GMT. The time now is 11:54 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 - 2018, vBulletin Solutions, Inc.
Copyright ©2002 - 2018, Paul Stenning.