My original plan was to develop the CPU board before the video board, so in the early stages of development, there was likely to be an operational CPU board before a video board was available and I wanted to be able to determine whether the CPU appeared to be running "stand-alone" with no operator interfaces connected.

The MTXPlus Bus Diagnostic Card was designed to achieve this.

As Martin and I have created MTXPlus+ to MTX interface boards as part of the Diagnostic board development, we now have the ability (by using temporary I/O port addresses) to test a video board on a standard  MTX. So, in a change to the planned development schedule, the video board has been "promoted" to Phase III and is being developed ahead of the CPU board.

The video board will contain the VDP, VRAM, video output conditioning and the Sound Generator. My long term goal is to have a VGA output from the machine, but the VDP has a low frequency output that it not VGA compatible. Video up-scalers are not easy to build, and as I did not just want to use an external up-scaler like the HD9800 described on my MTX Video page, the output from the VDP will be conditioned to output RGB signals compatible with the SCART interface of modern TVs. In another design development, the video board output circuit components will now be installed on a daughter board - very similar to the original Memotech idea. This configuration will allow different video configurations to be developed and tested without changing the base board design.

CPU Board - CPU, CTC, ROM, RAM, RTC, memory decode logic and a "local" I/O connector.

The intention is to create a Z80 CPU board with functionality that is as close as possible to the MTX, to such an extent that an MTX OS/BASIC ROM can be dropped into the ROM socket and run MTXPlus+ as an MTX. The features of MTXPlus+ mode will be made available by selection of its own ROM.

To support this basic design and to enable testing before the I/O board was available, I had intended to include an interface connector on the CPU board that would allow some minimal I/O to be connected to the system. The plan was to add a connector to the PCB to enable "local", i.e., not backplane connected, I/O to be built for testing etc.

However, with the video board having been developed before the CPU board, a considerable amount of testing has already been possible. There is now little benefit in including the "local" I/O connector on the CPU board and it has been removed from the design to free up the allocated board space.

Rather than the proposed 20MHz, the CPU clock speed will now be 16MHz as described on the notes page

Making use of another couple of components from Lez, the board will also include :

• A Real Time Clock using a Dallas 12887, which is a direct replacement for the IBM/AT clock/calendar. Martin has proposed that the RTC uses the same I/O ports (70h & 71h) as an IBM PC, these ports are unused by any Memotech hardware and may make it easier to refer to IBM documentation when programming the RTC.

• A MAX705 microprocessor supervisory module to implement the RESET functions

Enhancements : The draft design includes an 8255 Programmable Peripheral Interface (PPI) chip, intended to be used to provide an IDE interface to a Compact Flash drive for "mass" storage.

Aspirational : Ideally, it would be possible to build a PS/2 keyboard decoder that would allow MTX type I/O signals to be fed to the MTX ROM from a PS/2 keyboard. The design includes 2 CPLDs intended to perform this function.

Once a PS/2 keyboard interface has been built, and after creating a new OS for MTXPlus+, the enhanced capabilities of a PS/2 keyboard could be used.

Although they won't fit on one of the existing boards, the MAGROM components do not require a full size Eurocard board, but the extra space could be used for other things, for example, a floppy disk controller.

Add a Floppy Disk Controller, with SDCard support - not required, see I/O Card CF card

System Overview

To retain software compatibility with the original MTX design, the upgraded system will need to be I/O port and graphics compatible with the MTX, but will also be capable of operating in an enhanced configuration with a higher processor speed, more memory and enhanced graphics capabilities.

This block diagram is from the MTX Operator's Guide and gives a high level overview of the MTX design, including the I/O port allocation. I have shown the cassette interface in grey as, at this point, I am not sure whether I will implement it and may build a floppy disk controller instead. Similarly, I will probably do away with the TV interface and upgrade the graphics output from the VDP to VGA, but the rest of the ports will be unchanged, details of the I/O for "MTXPlus⁺" can be found on the I/O Ports page.

Hardware Design

If I was going straight from KiCad to a finished design, I would be able have quite a high component density on each PCB, but for development, I plan to use Eurocard sized prototype boards.

The system backplane will be fed with +3.3v, +5v and +12v regulated supplies from an PC ATX style PSU. To stabilise the on board power, 0.1uF decoupling capacitors will be connected between VCC and GND, installed as close as possible to each IC.

A 100uF electrolytic capacitor will be placed between VCC and GND, close to the power input pins of each PCB. The bulk capacitor is intended to overcome voltage slumps caused by PC board trace inductances. To handle noise and fast transients, a 100nF capacitor will be connected in parallel with the electrolytic.

For initial testing and use, the backplane will sit flat on the desk and the cards will stand up vertically from it, but I need to start thinking of a permanent solution.

This is the sellers photo of the actual item, you should be able to see that the items is brand new - it is still in the original box with it's original packing materials.

Since I am not using a standard Eurobus power supply, this case also has ample space behind the card frame to install my ATX PSU.

Although not necessarily true for the desktop case like this, many of the Schroff and other vendor's, equipment sub-racks have lots of options available to suit the end-users requirements.

Consequently, the basic parts kit is very spare and the kit needs to be built up at extra cost to build an enclosure that meets your needs. This was always likely to be a problem when buying a used rack - it might have been missing essential components that I needed. Even though I have still to confirm what is in the box, I think (hope) that I pretty much have everything that I need and this was a great buy at the price.