Projects

AT&Y SPEC-MATE Re-Creation

This has been a long quest for me. I’ve been searching for a SPEC-MATE for several years, since the advent of the Spectrum Next rekindled my interest in Spectrums in general. Back in the heyday of the Sinclair Spectrum and other 8-bit computers that kick started the home computing movement, I owned a Sinclair Spectrum 48K Plus that I upgraded from my original rubber key Spectrum with a kit from Sinclair. Along with my friends help, I used to hack games for infinite lives etc. This was a very common practice made easier by certian interfaces. The interfaces would freeze the game/program you were running and allow it to be saved to some sort of media, Tape, Sinclair’s own Microdrive or floppy disc for example.

Of course, these interfaces were quite expensive and above the budget of your average schoolboy hacker! I used to hack long-hand, that is to say, from the loader up. Some programs defeated my limited knowledge. Handily though, I had a friend who was into computers and had a more affluent family than I. He got a SPEC-MATE for his birthday. Later on, he loaned it to me so I could capture my game collection to Tape/Microdrive. This was a revelation! And one of the really useful features was that the SPEC-MATE made two files of main memory. This meant that you could load a dissassembler into one part of memory and a half of the programs code in the other. Hacking became exponentially easier!

Fast forward to the present day where my knowledge of hardware has come on a bit and PCB manufacture is a breeze with suppliers providing an amazingly high quality services for economic rates. I wanted to recreate a SPEC-MATE, but try as I might, I couldn’t find one to buy or loan. I hadn’t realised just how rare they had become. Then, my Facebook contact and friend Davide Barlotti offered me the loan of his. I accepted and he sent it over. This sadly coincided with a period of burnout in which I shelved the Spectrum and associated projects and left Davide’s interface safely tucked away on my shelves.

Later on, a chap by the name of Andy Lewis sent me his reverse-engineered schematic. I stored this for later use. Meanwhile, recently Davide requested I send his interface back, which I duly did and he notified me of a SPEC-MATE that had come up for sale on eBay. It was very pricey, but after searching for so long and knowing how rare they were, I snapped it up.

This fired me up into finally recreating this elusive piece of hardware. I first tested it briefly to check it worked, it did. The next step was to compare the ROM image with Davide’s. Sadly, it is the same. There was talk of an Opus Discovery compatible version of the SPEC-MATE but I believe this was never produced.

With my own device in hand, I was now confident in stripping it down to bare board in preparation to trace the circuit and verify it against Andy’s schematic. Unusually with this kind of hardware, all the ICs were socketed. I removed them and carefully placed them in a box on non-conductive sponge. I then removed all the sockets and other components with my trusty ZD-915 de-soldering gun. Lastly, removal of the edge connector and through board. Both had seen better days so I decided to replace them when I come to rebuilding it.

SPEC-MATE Stripdown

SPEC-MATE Stripdown

Once the board was clear, it was time to start tracing the circuit. Although I had Andy’s version, I wanted to verify it completely. I started by placing all the components on a schematic in KiCad EDA software. I then added my custom edge connector component and started to trace the connections. I started at one end of the edge connector pads and worked along, first the top row, then the bottom row. After that, I traced each connection on each IC in turn, using the boards own designations for each. Lastly, the other discreet components. This creates the schematic which I was able to match with Andy’s.

SPEC-MATE Schematic

SPEC-MATE Schematic

Following on from the process of circuit tracing comes PCB design. I made a concious decision not to follow the pattern of the original for the sake of size. Although I did start with a similar layout, I found that I could route the connections easier by repositioning various ICs. I also decided to have two trigger switches, one flat to the board and one at right-angles. The routing took several days with a number of design sessions. The result is a more compact board than the original.

New SPEC-MATE PCB Front and Back

SPEC-MATE PCB Front and Back

I recreated the AT&Y Logo to go on the board as an homage to the original. It is on the side facing the Spectrum along with the edge connector. The boards were fabricated by JLCPCB.com to their usual quality standards. Building is a simple process. I started with the IC Sockets and then added the discreets, followed by a single edge connector. I haven’t included a through board on the first build since the edge connectors I have don’t have pins that are long enough. I have ordered some long pin versions to facilitate this for the next build.

Recreated SPEC-MATE

Recreated SPEC-MATE

In the above image, you can see the ROM from the original interface. I later programmed my own ROM to replace it. Not having any 2764 EPROMs to hand, I used a 27C128, which is pin compatible with this board. I loaded the ROM image into the first 16k of space.

I was very pleased to find that it worked first time! I did a few basic tests saving to tape and Microdrive (using the excellent vDrive ZX). The SPEC-MATE isn’t as sophisticated as the Multiface and sometimes falls foul of Spectrum system variables, including RAMTOP. A fact I was reminded about when I tried to save a machine code program that had moved RAMTOP too low. In this instance, I set it to 65535 (With the CLEAR command). I also remembered that not all programs could be saved, depending on how it ran. Sometimes, you have to catch the program at a certain place! Oh the joys of 8-bit hacking 🙂

Here are all the files for fabricating the Recreated SPEC-MATE PCB:

  • Recreated AT&Y SPEC-MATE Schematic .PDF
  • Recreated AT&Y SPEC-MATE Gerbers .ZIP
  • Recreated AT&Y SPEC-MATE BOM .TXT
  • Recreated AT&Y SPEC-MATE Images .ZIP
  • Recreated AT&Y SPEC-MATE KiCad Files .ZIP
  • Recreated AT&Y SPEC-MATE Rom Image/Doc WEB
  • Recreated AT&Y SPEC-MATE Instructions .TXT

[UPDATE 10th Feb 2019]

It seems that the interface is a little picky about what EPROM is used. I had an issue with my own build where the EPROM failed after a few uses, but it is unclear whether this was because it went out of tolerance or failed completely. Another constructor tried several EPROMs before he found one that worked satisfactorily. Investigation continues and I will publish any findings later. Suffice it to say, at the moment, you may have to try several EPROMs until you find one that works consistently.

Claudius reports Pawel’s interface is built and works well. He sent me a couple of images:

Pawel's SPEC-MATE #1

Pawel’s SPEC-MATE #1

Pawel's SPEC-MATE #2

Pawel’s SPEC-MATE #2


Romantic Robot Multiface 128 Re-Creation

A while ago, I remembered what fun I had on the Sinclair Spectrum, an AT&Y SPEC-MATE and Microdrives, hacking games for infinite lives and such. Sadly, try as I might, I can’t find a SPEC-MATE for sale anywhere (UPDATE: See Above!). This led me on to the Multiface. A device produced by a company called Romantic Robot. With one attached, the user is able to freeze a program, save it to a variety of formats (Tape, Microdrive, Disk etc.) and add ‘pokes’ (change RAM values) for said infinite lives.

I first tried to re-create a simpler interface, the Spanish Phoenix III. My prototype appears to work with the Phoenix ROM but can’t page out at the moment. It also seems to work with the SPEC-MATE ROM but with the same issue. Research on that one continues.

In the meantime, I sought to investigate how these interfaces work and see if I could re-create one from the hardware itself. I chose to work on the Multiface 128 since it is compatible with the 48K/128K and +2 models of Spectrum and I am in contact with someone re-creating the Multiface 3 which is compatible with the +2A/B and +3 models. Between us, all the models are covered!

The first step was to photograph the board in detail to use as a reference for re-building later and for chip placements etc. I will post the images I captured along with the design files below.

Following this, I stripped my Multiface down to bare board. I acheived this using a de-solder gun, a most useful piece of equipment. I tried very carefully to de-solder all the chips without damage. Sadly, the 74LS27 was very stubborn and got a bit mangled! On the positive side, the PCB wasn’t damaged at all during the process which means I should be able to re-build the interface at a later date.

MF128 Stripped

MF128 Stripped

I placed all the chips on an non-conductive sponge in a box, along with the rest of the components. The only part I didn’t remove was the edge connector and passthrough board. This would have been a difficult job to do without causing unnecessary damage.

I then proceeded to lay out a set of components in the EDA Program KiCad. A free Schematic capture and PCB Layout program. I recently converted from EagleCAD and I have grown used to the different methodology without too much difficulty. Once I had laid out a full set of components, some of which I had to draw from scratch, I added the connections for the edge connector. I prefer to use net labels to indicate connections rather than the more conventional bus lines. I find it makes for a cleaner diagram.

When I had completed the edge connector, I traced each connection along it. I started with the upper connections since they were easier to access. I followed the tracks with a pointing device, in this case a pencil! Where I was unsure a connection got to a particular destination due to interconnecting vias, I would double check with a multimeter set on it’s lowest resistance range. Most modern meters have a continuity/diode check that can be used for this purpose. My aged Fluke 73 has to make do with resistance ranges.

This process was quit painstaking but essential to get correct. Lots of the connections go to more than one chip’s pin so checking multiple times is essential. When I had gone through all the edge connector connections, I continued by tracing every chip pin one by one. The board isn’t marked so I added my own component designations to one of the images I had captured and copied those in KiCad.

Component Designations

Component Designations

Each chip pin was traced and gradually, I assembled the circuit diagram, line by line. Again, checking, double checking and even triple checking was necessary to avoid mistakes. After all the chips were done, I added the two resistors, NMI button and the protection diode on the edge connector. This diode is ommited on the prototype since I wasn’t intending to put a passthrough on it but can easily be added. Finally, after a good few man hours, the job was done.

Recreated Multiface 128 v1.00 Schematic

Recreated Multiface 128 v1.00 Schematic

Time to design a PCB. I intended to get the prototype PCBs fabricated by EasyEDA due to the fairly fast turnaround and good price point for prototype boards. The final boards I will have made by Hackvana since they are of very high quality. To hit EasyEDAs Prototype price point, the PCBs needed to be within 100mm x 100mm in size. The original Multiface board is 114mm x 70mm, so I shuffled it around slightly. Later on I will expand the board out to fit the original case.

The PCB was quite challenging to lay out. It required mutiple vias and some creative routing. I hand routed all of the connections one by one. This is my favorite part of any design though.

Spectrum Multiface 128 v1.00 KiCad Preview

Spectrum Multiface 128 v1.00 KiCad Preview

Having routed the board, I did a DRC check to make sure the connections were all complient with the design rules I had set up. All was fine. I then plotted Gerbers and drill files which would be submitted to the fabricator. Rather than trust these files are OK, it’s good practice to check them externally to KiCad. I use Gerblook at the moment but I have noticed a message on the site to say it has been bought. Hopefully the service will remain free of charge.

Spectrum Multiface 128 v1.00 Gerblook Front

Spectrum Multiface 128 v1.00 Gerblook Front

Having checked the board looked correct on Gerblook, I uploaded the ZIP of Gerber files and drill file to EasyEDA and ordered the prototype boards. Time passes….

…And the boards arrive. they took around 2 weeks from the time I ordered them. Not bad for a budget service.

Recreated Multiface PCB Front and Back

Recreated Multiface 128 PCB Front and Back

A quick visual check found the boards in good order. I assembled all the components, including a few from the original device since I had none of those particular ones in stock. I soldered the de-coupling capacitors first since they are the lowest profile. I followed these with the resistors and NMI button, then all of the chip sockets.

Fully Built Interface

Fully Built Interface

The above image is the completed interface mounted on the rear of a 48K Spectrum Plus. You will note that the edge connector faces the Spectrum and the chips are on the reverse of the board. This is in keeping with the original design. It also means that the board has extra room behind between it and the Spectrum. You will also notice that the ROM chip U1 has no Erase window. This is because I had no 27C64 EPROMs in stock. This is an OTP chip (One Time Programmable).

I was very pleasently surprised to see the interface work first time!

Multiface 128 Menu

Multiface 128 Menu

One press of the NMI button brings up the familiar menu. I have gone through the motions of testing the unit by initiating a tape save and poking 16384 with 255. It all seems OK. I can’t test with Microdrives as yet since I have none working. I will update when I have the chance. I have since tested the interface with both a standard Microdrive and the fantastic vDrive ZX and both work properly.

In the meantime, if you would like to investigate this first iteration of the design or build one for yourself in it’s present state, the design files are below. I will probably be updating the design in due course by extending the PCB to fit the original case.

Here are all the files for fabricating the Recreated Multiface 128 PCB v1:

  • Recreated Multiface 128 PCB .SCH schematic file
  • Recreated Multiface 128 PCB .PCB board file
  • Recreated Multiface 128 PCB .PRO file
  • Recreated Multiface 128 PCB .LIB file
  • Recreated Multiface 128 PCB Gerber .ZIP file
  • Recreated Multiface 128 PCB Schematic .PDF file
  • Recreated Multiface 128 PCB BOM .TXT File

The entire directory of resources is available at: Recreated Multiface 128 PCB

Please note, due to the way the website server stores and accesses files, I have added underscores to filenames. Also, early versons of the Schematic PDF and the Schematic image above contained an error showing pin 11 of the SRAM chip being connected to A0 instead of D0. These files have since been replaced. The error never got to the gerber files.

I haven’t included the ROM file due to the unknown copyright status. however, it is available from multiple sites. The version that works in this hardware is 87.12 Search for ‘Multiface 128 ROM version 87.12’

This project was a proof of concept for me. I wanted to challenge myself in re-creating a fairly complicated piece of retro hardware and use it to aid in my understanding of how such hardware worked. I also wanted to preserve the design for retro enthusiasts enabling them to re-create their own. I’d welcome any contact with the original IP holder/s In the meantime, I make no claim as to the validity of my re-creation or fitness for purpose. I have checked it to a point on my own hardware but this is no guarantee of it working on any other hardware. In short, you use these files at your own risk and any loss or damage caused is entirely your own responsibility.

If you do make one of these interfaces, I’d love to know.

Multiface 128 recreations in the wild.

I have been contacted by several enthusiasts who have successfully built and are using the design. I’d love to see images of any interfaces in action. Please send them to info (at) projectspeccy (dot)(com) and I’ll add some to this page.

6 Boards by Claudius

6 Boards by Claudius

More Of Claudius's Boards

More Of Claudius’s Boards

Claudius Fatla had 6 boards made and visited me after requesting I autograph the boards for him and his Speccy friends. One of them is Paweł Osipowicz who built this one:

Paweł Osipowicz's MF128

Paweł Osipowicz’s MF128

Stefan Schomberg successfully built a recreated MF128. Apparently, he ordered ten PCBs and sold the spares off so there should be nine more out in the wild 🙂

Stefan Schomburg's MF128

Stefan Schomburg’s MF128

One of which is Erwin Rattinger’s:

Erwin.Rattinger's Multiface 128

Erwin Rattinger’s MF 128

Erwin's Interface In Action

Erwin’s Interface In Action

Guenter Bruetting is building a couple of MF128s and a simple 8-bit IDE interface (see below). Guenter has added a throughport on the back of one of the MF128s but is yet to add the diode as on the original.

Guenter Bruetting's Interfaces

Guenter Bruetting’s Interfaces


Shadow of the Unicorn Interface Re-Creation

I reverse engineered this interface as a reaction to prices I saw online for the original Mikro-Gen hardware. It’s not a very complicated piece, just an EPROM and an IC with a joystick connector at the top with a few diodes. The original is a single sided PCB and is routed for convenience (more of that later).

I was supplied some really good photographs by someone who I have sadly forgotten the name of, please let me know if it was you. I labelled on of the images to create a guide to make my job eaiser.

SotU Interface Labelled Image

SotU Interface Labelled Image

I then layed out the components in a KiCad schematic and started to trace each connection. I tend to do the edge connector first and any point-to-point connections later. This builds up a complete schematic. Then it is a simple process to lay out a PCB. I stayed pretty close to the original in this case but I did cheat and use a few second layer connections for my own convenience. I also wanted to line the diodes up to make the board neater than the original too.

SotU Interface on Gerblook.org

SotU Interface on Gerblook.org

I also took the opportunity to recreate the original logo to add to the board. KiCad makes it easy to make graphics for the silk screen layer. KiCad include an application where you can import one bit BMPs and turn them into library parts.

I had the boards made by JLCPCB.com as usual and they arrived in a few weeks as is normal. I built one up and programmed an EPROM up from the ROM image at World Of Spectrum. But, the interface failed to work. This was at a time when there was a lot going on personally and I simply thought I’d made a mistake. I shelved the interface for a while in frustration.

I came back to it recently and re-checked the tracing. As you can see in the first image, it’s not too complicated and I was puzzled why it wouldn’t work. I then showed it to my friend Russ and it dawned on us that because the address lines were scrambled, then the ROM would be too. This explained the ‘convenient’ routing on the PCB. The ROM image on WOS must have been descrambled for use with emulators!

I posted this info on Facebook in the Spectrum Forever and Spectrum For Everyone groups and several people lent a hand in re-scrambling the ROM. Many thanks go to everyone on the two Facebook pages – but especially Russ, Jerzy and Matthew Logue who supplied a ready-to-run scrambled ROM image which you will find below.

Finished Shadow of the Unicorn Interface

Finished Interface

On my own copy of the interface, I have used an Atmel 27C128 one time programmable EEPROM but you can use a UV Erasable EPROM in it’s place like the original interface. You can find all the instructions for the game and other documents on WOS. Please bear in mind that the ROM image on WOS is not the one you should use with this interface. Use the file posted below.

Here are all the files necessary for fabricating the Shadow of the Unicorn Interface:

  • Shadow of the Unicorn Interface Schematic .PDF
  • Shadow of the Unicorn Interface KiCad files .ZIP
  • Shadow of the Unicorn Interface ROM .ZIP
  • Shadow of the Unicorn Interface Images .ZIP
  • Shadow of the Unicorn Interface Gerbers .ZIP
  • Shadow of the Unicorn Interface TZX .ZIP
  • Shadow of the Unicorn Interface MDR .ZIP

I have included the TZX files of the game for convienience and also an .MDR file for those lucky enough to have a vDrive ZX.

I hope this project enables those who could never obtain the original or refuse to pay the ‘Retro’ prices demanded these days, to play the game as it was originally intended.


Sinclair Spectrum Edge Connector Key

Guenter Bruetting kindly sent me an STL file for 3D Printing which is for placing in the edge connector slot position.

Slot Key In Place

Slot Key In Place

Anyone who has had to make edge connectors to be compatible with Sinclair Spectrums (Or ZX80/81s) will know how difficult it is to find a suitable key. Indeed, I have been using large paperclips for the purpose until now. However, having used these 3D Printed Slot Keys, I doubt if I’ll ever use anything else from now on.

A Strip of 3D Printed Slot Keys

A Partial Strip of 3D Printed Keys

I printed mine on an Anet A8 3D Printer using EFORINK PLA at 200deg C with a 60Deg C Bed. My 3D Printer usually requires a raft, which you can see in the above image. I will include the original STL file and my own GCode file created in Cura, but feel free to tune your preffered slicer to suit your own printer.

  • Guenter Bruetting Spectrum Edge Connector Slot Key .ZIP
Slotkeys without raft

Slotkeys without raft

A big thanks go to Guenter Bruetting for allowing me to add this useful resource to the site.

[Update] It has been noted that since these keys are not fixed in and rely on friction, that they can fall out. I have found them to be reliable but I also have got into the habit of always checking for a key upon insertion or removal of an interface.


Sinclair Spectrum Backplane

A long time ago, there was a fabulous magazine devoted to the Spectrum called Your Spectrum (Later becoming Your Sinclair). In it was published an article about add-ons for the machine. The article was illustrated using a Backplane to support a variety of peripherals.

Your Spectrum Issue 18 (Pages 20/21)

From Your Spectrum Issue 18

Although the article doesn’t mention the backplane, I always remembered it and thought it was a good idea. This led me to designing one for my own use.

Spectrum Backplane v2.00 - EasyEDA Gerber View

Spectrum Backplane v2.00

It took a couple of versions to get the connectors correct! All of the edge connector pins are straight through, meaning that it should be compatible with all Spectrums, and aside from the on board power supply, also be compatible with other machines that use 28×2 edge connectors (this has not been tested!).

The optional on board power supply is purely for the peripherals and does not feed into the machine it is connected too. There is a jumper to select whether the peripherals are powered by the connected machine or via the on board power supply. It is based around a Traco unit that can supply 1.5A of current. It requires at least 6.5v to operate properly but any center positive supply of 6.5v – 12v should be fine. Included is a power LED to indicate external power is being used. There is also a convenient reset switch.

There are two straight through edge connectors to mount interfaces on in the traditional fashion. One on the left hand edge and one on the rear. The other four connectors can either be used with IDC pin headers or sockets or via adaptors. Included in the design files are PCBs for adaptors for converting these to standard edge connectors, albeit vertical. The adaptor board is simply soldered between the pins of an IDC plug to make up the adaptor.

Spectrum Backplane v2.00 With Vertical Adaptor

Backplane v2.00 With Adaptor

Although I have found them to be ridiculously expensive, it’s a good idea to put polarising pins in to the IDC socket in the ‘slot’ positions to avoid the adaptor being plugged in the wrong way around.

I have carried out the minimum of testing on this board so far and you will notice that there are no buffers included as yet, this may be rectified in the future. It has operated fine with my hardware. So, for now, the design is there to experiment with as is. If you choose to build and use one, it is entirely at your own risk and no responsibility will be taken for any loss or damage.

Here are all the files necessary for fabricating the Spectrum Backplane v2.00:

  • Spectrum Backplane v2.00 Schematic .PDF
  • Spectrum Backplane v2.00 KiCad files .ZIP
  • Spectrum Backplane v2.00 Gerbers .ZIP
  • Spectrum Backplane v2.00 Images .ZIP
  • Spectrum Backplane IDC-Edge adaptor Gerbers .ZIP

I’d be very interested to see what uses the Backplane is put to. Please feel free to drop me a line with pics etc.

Horizontal Adaptor

  • Horizontal Adaptor Schematic .PDF
  • Horizontal Adaptor KiCad files .ZIP
  • Horizontal Adaptor Gerbers .ZIP
  • Horizontal Adaptor Images .ZIP
Backplane Horizontal Adaptor

Backplane Horizontal Adaptor


TZXDuino Compact

Because of the Spectrum Next, I got interested in Sinclair Spectrums after a long (15 year) break. I have acquired a few of the original models and enjoyed reaquainting myself with the Spectrum system and programming. Although I have a DivIDE and other such interfaces, there’s nothing like the nostalgia you get when you load an original program from tape.

Although I own several Spectrum tapes from the old days, there are files available online in TAP and TZX formats which can be played as if you had the original tapes. These are redily loaded into Spectrum emulators but not so easily into the real thing. However, a couple of very clever guys, Duncan Edwards and Andrew Beer came up with a device to do just that. The TAP or TZX files are copied onto a Micro SD Card and placed in the device which, via a menu, you can play the files into a real Spectrum machine. The device can be used with several other 8-bit micro computers but I will be concentrating on the TZXDuino for Sinclair Spectrum.

TZXDuino Compact v1.01 (Gerblook.com)

TZXDuino Compact v1.01 (Gerblook.com)

The circuit is fairly simple. An ATMEGA328 is connected via I2C to an SD1306 OLED Display and via SPI to a commonly available Micro SD Card module. Five tact switches are used for selecting the file to play and an audio output is driven directly from pin D9 of the microcontroller. The audio output seems adequate for most files but occasionally fails. I will be looking into adding a small amplifier of some sort to boost the audio a little.

The PCB was designed to be easily mounted in a box with 3.2mm holes in each corner. The display and micro-sd card module sit on the top. I have found it necessary to put a small piece of foam behind each to prevent them from bending towards the components mounted on the PCB.

TZXDuino Compact v1.01 PCB

TZXDuino Compact v1.01 PCB

I had the PCB fabricated by the ever excellent Hackvana.com who always give a great service and produce really good quality boards.

As per normal with building any project, start with the low profile components first, Resistor, Diode, Crystal etc. And gradually build up to the higher components. The only difficult part to solder is the USB Mini port, in this application it is there just to supply power. You can use the two header connections directly behind the SMD pads if you prefer. The diode is for polarity protection.

The SD-Card module I purchased had a straight header already soldered on so it required de-soldering and replacing with a strip facing down, under the board. The display came without a header strip installed so it was an easy case to add one.

Programming the ATMEGA328 is made easier if you have an Arduino Uno. I extracted my Uno’s ATMEGA328 and temporarily installed a new one. I then programmed it with the ICSP header on the Uno’s board. If you have an ATMEGA328 with a bootloader, you can use the USB input as normal. If neither of these options is available, you will need to make a small adaptor board for ICSP programming, use a breadboard or perhaps, use a minDUINO as a base.

Here are all the files necessary for fabricating the TZXDuino Compact v1.01:

  • TZXDuino Compact v1.01 Schematic .PDF file
  • TZXDuino Compact v1.01 KiCad files
  • TZXDuino Compact v1.01 Gerber .ZIP file
  • TZXDuino Compact v1.01 Sketch .INO file
  • TZXDuino Compact v1.01 Images .ZIP file
  • TZXDuino Compact v1.01 Parts List TXT file

The entire directory of resources is available at: TZXDuino Compact v1.01

Because I used a previous version of the KiCad files, the newer v1.01 directory requires the older ‘rescue’ files. If you intend to use the KiCad files, include all the items in the zip file.

Future development may result in an amplified version of this board.

Many thanks to Duncan Edwards and Andrew Beer for designing the original.


Sinclair Spectrum Composite Mod PCB

I designed a PCB to house several versions of the Spectrum Composite Mod. Details of how to implement the PCB can be found in my series on Refurbishing a 48K Rubber Keyed Spectrum on my companion site, Projectavr.com. The PCB simply replaces the existing Modulator PCB. I would recommend that you remove the original carefully and put it in storage, in case you ever wish to re-install it for authenticity.

The PCB itself is very simple, requiring just a few components to operate. Choose which of the three options you would like to use.

Option 1 is an Inverted Emitter Follower recommended to me by Andrew Gostling in an attempt to get a Dell UltraSharp 2007FP Monitor to work with a 48K Spectrum. Sadly, the monitor is extremely picky about it’s composite input and is well documented for it. In this instance it didn’t work. However, the composite mod works with all the other monitors I have tried it with.

Option 2 is a simple NPN transistor and resistor mod. the NPN Transistor can be any general purpose device. I have used a BC547 in the past.

Option 3 is a single 100uF capacitor which is seen to be the minimum required.

I have had several conversations around the need for other components with regard to the composite mod. Some say that it is not necessary at all and others say that, in case the monitor goes faulty or is very badly designed, it can feed voltage back to the Spectrum. I recommend you do your own research and choose the option that best suits you (including a direct connection and/or suspending the components outside of the modulator box retaining the original PCB inside the modulator).

Spectrum Composite Video Mod PCB

Spectrum Composite Video Mod PCB

Here are all the files necessary for fabricating the Spectrum Composite Mod PCB:

  • Spectrum Composite Mod PCB .SCH schematic file
  • Spectrum Composite Mod PCB .PCB board file
  • Spectrum Composite Mod PCB .LIB file
  • Spectrum Composite Mod PCB Gerber .ZIP file
  • Spectrum Composite Mod PCB Schematic .PDF file

The entire directory of resources is available at: Spectrum Composite Mod PCB

To create the board, you can simply copy the Gerber ZIP file and submit to your favourite fabricator, I use Hackvana for my boards.

Spectrum Composite Mod PCB In Place

Spectrum Composite Mod PCB In Place

All these files are provided freely in the spirit of Open Hardware. Feel free to copy and modify the design for your needs. I have checked operation on my own hardware but this is no guarantee of it working on any other hardware. In short, you use these files at your own risk and any loss or damage caused is entirely your own responsibility.

Having said that, I’d love to know what option you have found to work best and in what combination of Sinclair Machine and display device.


Simple 8-Bit IDE Interface for Spectrum +3e

A long time ago (2002) I came across a design for an IDE interface for the +3e. I loved this idea, a Spectrum with a Hard Drive, even if this particular interface only gave you half the capacity of the hard drive. At the time, I was just getting into PCB design and, with a dodgy version of Ares/Proteus, I designed a gosh-awful board covered in links (since I couldn’t produce a double-sided PCB at home). However, it worked! I was very chuffed with myself and sent the design, along with an internal version to Garry Lancaster who kindly published it for me. I included images of my interface captured with the then brand new Casio QV-10 Digital Camera! You can still download the original archive from the WOS page.

Original Simple 8-bit IDE Interface (Casio QV-10 Image)

Original Simple 8-bit IDE Interface(Casio QV-10 Image)

Since getting to grips with KiCad, I have designed a few Spectrum related PCBs so I thought I’d re-design the IDE Interface with a double-sided board. The process was fairly simple since the interface only uses 3 components.

Simple 8-Bit IDE Interface KiCad 3D Preview

Simple 8-Bit IDE Interface KiCad 3D Preview

The board presented here was designed a while ago and bears the ProjectAVR logo rather than the newer ProjectSPECCY logo. I included holes for mounting the PCB in a case if required and the IDE and Spectrum Edge connectors can be straight or angled depending on the builders preference. To use this interface, you will need to replace the two ROM chips in your Spectrum +3 with versions on the WOS +3e pages. Full instructions are given there.

Completed Interface with Hard Drive

Completed Interface with Hard Drive

Here are all the files necessary for fabricating the Simple 8-Bit IDE Interface for +3e:

  • Simple 8-Bit IDE Interface for +3e Images .ZIP file
  • Simple 8-Bit IDE Interface for +3e KiCad .ZIP file
  • Simple 8-Bit IDE Interface for +3e Gerbers .ZIP file
  • Simple 8-Bit IDE Interface for +3e Schematic .PDF file

The entire directory of resources is available at: Simple 8-Bit IDE Interface for +3e

All these files are provided freely in the spirit of Open Hardware. Feel free to copy and modify the design for your needs. I have checked operation on my own hardware but this is no guarantee of it working on any other hardware. In short, you use these files at your own risk and any loss or damage caused is entirely your own responsibility.

[Update]
Originally, back in 2002, I did design an internal version of the interface which I didn’t fabricate myself. Now, Claudius Fatla has designed a neat internal solution. Files for this can be found on this WOS page.