Interak keyboard – proof of the pudding

This is just a quick post.

I’ve spent some of the afternoon in a happy state re-living the 1980’s by typing in a hex dump from a magazine and spotting my mistyping.

161023-img_20161023_201745The magazine in question was Interaktion – newsleter No. 2 from the Interak user group and as you can see, the program is a version breakout.

You can see in the photo that I have added the USB serial mod I hinted at in the last post. This enables me to use a PC to squirt data through a virtual serial port and have it appear at the Interak as though it had been typed in.

Thanks go to Alan Paton at for putting the newsletters onto the ‘net.


Retrochallenge 2016/2 – Interak keyboard wrap-up

160915-img_20160915_204655In my last two posts I wrote about how I made an adapter for my 1980’s Interak-1 computer so that it could make use of USB keyboards as the original Alphameric keyboard had failed.

Those posts were a bit rushed and the adapter was still a work in progress. It’s finished now and so in this post I’ll try to tell the full story and in more detail so if anyone else has the need for a USB keyboard to 7-bit parallel ASCII keyboard they will have something to follow.

The problem.

The Interak comes from an era before USB became ubiquitous, when a parallel cable running from a semi-intelligent keyboard to the main CPU was commonplace.

161005-img_20161005_192546With these keyboards, when a key is pressed, the binary representation of the key, in ASCII, is presented on seven data lines. An eighth bit called Strobe is then pulsed high (5V) and then low (0V) to indicate to the CPU that a character is ready.

Sadly, the keyboard that came with my Interak had developed a fault and several columns of keys didn’t work at all. I tested the socketed chips with my IC tester but they seemed OK and so I suspected the micro-controller and I didn’t want to go too far down that road and so I decided to do away with the  original keyboard.

What to do now?

I looked around for a while and found a number of solutions that will use PC keyboards with the PC-mini-DIN connector but they haven’t been mainstream for over ten years and I didn’t want to end up with the same problem later.

USB is the way to go.

When USB is the solution… Crikey.

161001-img_20161001_181323In all honesty, I don’t know a great deal about USB and the though of writing a USB host to interface with a keyboard seems like no kind of fun (Retrochallenge is meant to be fun) and so I looked around for solutions and found HobbyTronics who produce a small USB host board that just needs 5v and will produce a serial output.

Although a serial stream isn’t what’s needed, it is easy to pipe it into an Arduino and get that to output the ASCII in parallel on its IO pins.

161001-img_20161001_205049I made a prototype using an Arduino UNO connected to the Hobbytronics board.

To make debugging easier I used a software serial library to keep the hardware serial port free. More on that later.

The real thing.

The Arduino UNO worked very well. Well enough to convince me to use a small Arduino Pro Mini and make the thing real.

The Interak uses a 15 way D type on the front for the keyboard connector and so I soldered the Pro Mini as close as I could to the connector and the the three wires to the USB board.

161003-img_20161003_210244Next, get it into a small case… 161005-img_20161005_200524

Proof if proof were needed.

161005-img_20161005_200700So there you go. My beloved Interak has a new lease of life.

One more thing…

There is a nice little addition waiting in the wings here. The Arduino has another serial port, the hardware one and so It would be quite easy to add the standard USB to serial board and then the Interak’s keyboard port would appear to my PC as a serial port and I could squirt data into from a terminal emulator. There would be no handshaking and so I’d have to artificially drop the data rate with delays but it would work.

The code.

* Interak-1 USB keyboard adapter
* By Andy Collins.
* This code is free to use.
#include <SoftwareSerial.h>

#define CR 0x0d
#define LF 0x0a

SoftwareSerial mySerial(11, 12); // RX, TX

const int STROBE = 9;
const int D6 = 8;

int inByte = 0; // incoming serial byte
int lastInByte = 0;

void setup()
// Open the hardware serial port for debugging

// Start the SoftwareSerial port
DDRD = DDRD | B11111100; // Using d2-d7 as output. Saving d0-1 just in case.

pinMode(D6, OUTPUT);

digitalWrite(STROBE, LOW);

void loop() // run over and over
inByte = -1; // Nothing yet

if (mySerial.available()) // USB Keyboard
inByte =;
else if (Serial.available()) // USB Serial port
inByte =;

if(inByte != -1)

if( (inByte == LF) && (lastInByte == CR) )
PORTD = inByte << 2; // Shift left to avoid TX, RX;

// Catch the overflow.
digitalWrite(D6, inByte & 0x40);

// Bits set up. Now Strobe;
digitalWrite(STROBE, HIGH);
digitalWrite(STROBE, LOW);

lastInByte = inByte;



Retrochallenge 2016/2 – Oh-oh we’re half way there

161003-img_20161003_210244In my last post I started setting up my stall and wrote a simply Arduino sketch to read a character from the Hobbytronics USB keyboard adapter and squirt it to my PC.

My Interak-1 has a 15 way D-type connector for the keyboard as I’ve decided that my keyboard solution will plug into that rather than be buried inside the machine itself.

I have  also decided that the final project will live on an Arduino Pro mini as they are tiny and so the photo above shows the Arduino soldered onto a 15 way D-type plug linked to the USB keyboard PCB.

Cut to the chase.

161003-img_20161003_210058This photo shows things are basically working! Happy Happy Joy Joy etc.

There is some boxing up to be done and some writing up to be do so that if there is another Interak-1 user out there who needs a keyboard, they can see what I’ve done.

It’s not so daft really as plenty of old machines used 7 bit parallel ASCII keyboards and plenty of those will have been lost.

More to follow.


Retrochallenge 2016/2 – Interak keyboard

So RC is upon us and I have decided to replace the failed keyboard on my newly acquired Interak-1 with a modern USB item.

The Interak keyboard uses a 7 bit parallel interface to present an ASCII character to machine itself with a strobe line to say that the data is ready. It’s a bit much to ask the Interak to cope with USB directly and it’s a bit much for me to learn enough about USB to be able to program a Z80 to cope in the time allowed and I don’t want to.

161001-img_20161001_181323Enter HobbyTronics who produce a small USB host board that just needs 5v and a USB keyboard and will produce a serial output.

The serial output can’t go straight in to the Interak either but serial to parallel isn’t too bad.

Enter Arduino.

161001-img_20161001_205049So, here we have an Arduino UNO with a prototype shield and a tiny USB keyboard. At the moment it’s just reading the serial input from the keyboard adapter and squirting it down the serial port to the PC but it’s a start.161001-img_20161001_205115 Next is to set up some pins for the parallel interface.

Keep watching (if you want).

Retrochallenge 2016/2 – Throwing my hat into the ring.

It’s Retrochallenge time again and how quickly it comes around and how ill-prepared I am.

Recent competitions have seen great entries by talented retro-folks the world over whereas my contributions have been less glorious but filled with enthusiastic zeal.

This time around will be no different 🙂

I’m not sure what I will do to be honest. I may take a broken retro system down from the “shelf of good intentions” and try to fix it. I may try and interface a ZX printer to something non-Sinclair or perhaps get the Tandy 4 colo(u)r plotter talking to my PDP-11. Who can say?

Whatever happens there will be joy and frustration but mostly joy.

In the words of poppy-funster “Jem”

It’s just a ride, it’s just a ride
No need to run, no need to hide

Though there may be a need to solder…

See you on the other side.


Interak 1

160917-img_20160917_101041Back in the day, manufacturers of small computer were legion. There were myriad little companies trying their hand at getting into micros. If you look back through the pages of Electronic Today International, Practical Wireless etc. you will find adverts for these small companies selling their kits, PCB etc.

It was usual in those pre-web days to send off a “Stamped-Addressed-Envelope” (SAE) along with some unused postage stamps and in due course a catalogue would arrive  detailing the product available.

On such company was Greenbank Electronics of Merseyside here in the UK. Their machine was the Interak 1, a Z80 based computer composed of a 19″ rack with a backplane and a collection of cards that could be purchased individually as funds allowed.

Back then there were many companies vying for my money and whilst I got as far as sending off the stamps to Greenbank and getting the Interak catalogue, I got no further.

However, as you can see above, thanks to the generosity of a friend of mine, I now have an Interak1 and I am very happy about it.


New arrival – Sharp MZ-80K

160802-IMG_20160802_202212Like so many of my machines, the Sharp MZ-80K was out of my reach the first time round and by the time I could have afforded it, I was on to something else.

This example was passed on to me recently and had previously been used for developing a pharmacy system years ago.

The years have not been too kind to this one and there is some surface rust but it’s not too bad cosmetically.

160805-IMG_20160805_185525It’s not without problems however. When I first gave it some power I saw the screen on the right. As well as the general mush there are vertical lines running up through the characters. That made me suspect the character generator ROM and sure enough, it has a broken pin.

160805-IMG_20160805_203905 I snipped a pin from a surplus IC and solders that in place. The vertical line has gone and I’m just left with the mush.

MZ-80KSignalsDiagnostics continue.

Fortunately I have a copy of the service manual and on page 11 it shows a series of expected ‘scope traces. At pin 8 of IC7 there should be an 8Mhz wave. Sadly, I see about 760Hz. Way off.

MZ-80KClockCircuitIt doesn’t seem possible as the video is rock solid and all of the video signals are derived from that 8Mhz clock signal. I’ve checked several times and I always get the same result. I was suspecting the crystal, the 74LS93 or the 74LS04 so I replaced all three but to no avail.



Retrochallenge 2016 – End of the line. All change.

Well, another RC comes to an end an if there’s any truth in the saying “It’s not the winning, it’s the taking apart” then all is well. If, however, it is the winning then we have another “success deficit” on our hands.

So far I have the VT-180 in pieces. I looked at a few signals on the video board and compared them to those marked on the schematic but they all looked fine (see the photos).

160116-IMG_20160116_181840 160116-IMG_20160116_181754 160116-IMG_20160116_181725So next up I wondered if the 2114 static RAM chips have failed. I’ve seen them go on my DEC Rainbow and also on a Commodore PET a friend of mine has, so it’s quite common really. It would explain the garbled output.

One thing I like about DEC kit is it’s well made. The downside is that they used a lot of solder and getting chips out of any DEC board I’ve ever worked on is very difficult. So, as the last few hours of RC2016-1 tick away I have3 chips out, new ones on order and crud to clean out of the holes in order to add sockets.

The nice thing about Retrochallenge is there’ll be another one passing by any minute now.

Retrochallenge 2016 – Week 1 done. Some progress.

So here we are, one week in to RC2016-01 and what do we have to show for it?

As you saw in my previous post, I have taken the lid off the VT180 and it sits as a bare metal cage while I try to figure out what’s wrong.

151231-IMG_20151231_174248The initial display (shown on the left) has changed and is less verbose. 160106-IMG_20160106_194827 I’m not sure what happened to trigger the change.

The screen display is no longer in sync and rolls very quickly.

In order to help with the repair I have taken the VT100 board out of the card cage and I have it plugged in to the cable that was powering the card edge connector on the card cage. It’s clearly straight through so should be fine.

160106-IMG_20160106_194705That means that I can get to the board with my meter and ‘scope etc. I have also removed the AVO (advanced Video Output) board and the small inter-connect board. This hasn’t changed the symptoms at all.

The first real diagnostic work was to measure the voltages from the PSU. They all seem fine.

I have been able to use a logic analyser (borrowed – I really want one of these) and I’ve been taking a look on the bus.

have attached three screen shots taken of the logic analyser screens. I have 18 channels and so I have the data bus, the lower 8 bits of the address bus and the synch pin. All signals are from clips on the 8080.

I have the triggers set to stable high on sync (start of new instruction) and 0 on the address bus. I put the analyser to wait for trigger and turned on the VT180.
The three shots are from the same run and I’ve just stepped through the time a little to show what happens at start up.

vt180-digiview-3vt180-digiview-2vt180-digiview-1I get this with or without the keyboard. With the keyboard, all of the lights are lit. When I first started on this repair, a few, normal looking lights came on first and then, after a while, they all came on. It’s a funny old world.


Retrochallenge 2016 – First look.

151231-IMG_20151231_174248Let’s start at the very beginning. It’s a very good place to start – Apparently.

According to the documentation, the VT180 is a VT100 serial terminal with a VT18X add on board. This add on is a Z80 single board computer that talks to the VT100 though one of its communications ports. this make for quite a nice combination for the day.

You can use the VT100 as just a terminal to your corporate mainframe or mini-computer and when required, fire up the Z80 into CP/M-80 and compute like there’s no tomorrow.

Mine is computing like Yesterday was a little rough.


At this point I am working without the VT-18X board present. I don’t think I need it yet. My understanding is that the VT100 side should come up cleanly without it. I could of course be wrong but I’m going down this road for a while.

The photo above shows what I’m dealing with. The keyboard has sensible lights on but I don’t know yet if that means anything.

My first job as always is to try and re-seat all of the socketed chips on the board. I’ve don’t that and it made no difference. Oh well.

Going in.

Next step, are the voltages correct? The VT100 has an Intel 8080A processor on board and that needs care and feeding.

Off with its head.

One thing I’ve always liked about DEC equipment is the way they built it. Their kit always feels like it was ment to br built and stay together but they also kept in mind that you will need to service it.

160102-IMG_20160102_140931The photo on the left shows one of four plastic “poppers” that hold the case halves together. There are also four bolts to hold the steel frame in but that’s it. Dead easy.


Check the power supply.

Using the diagrams in MP00633_VT100_Schematic_Feb82.pdf I checked the voltages on all of the pins. Everything is OK with the exception of the 12V line. I can see that there should be 12V on the orange wire but there is hardly anything there at all 0.4V and just ain’t enough.

Is that my problem?

Hmmm. The CRT uses the same 12V supply and you can see from the top photo that the CRT is working. When I tested the voltages I didn’t have the CPU board in and the CRT didn’t come on. When I plug the CPU board in, the 12V comes back.

Doh! Red herring.


A short step from where you are.