Minimalist Europe Card Bus (MECB) – Motorola I/O Card + Sound

With my first Minimalist Europe Card Bus CPU Card finally completed, to my satisfaction. Next up, is my Motorola based I/O Card, which will finally enable a complete modular 6809 based system.

In fact, my initial design has been completed, which also contributed a couple of recent posts / videos as I finalised the ACIA Baud Rate Generator, and the Interrupt Driver components of the design.

The Motorola based I/O Card includes the following Motorola peripherals:

• A MC6850 Asynchronous Communications Interface Adapter (ACIA)
• A MC6821 Peripheral Interface Adapter (PIA)
• A MC6840 Programable Timer Module (PTM)

In addition, I’ve added an optional Sound Interface, via a SN76489 Digital Complex Sound Generator, interfaced via one of the PIA ports.

The schematic and PCB layout for the initial prototype have both been completed, and the first prototype PCB’s have just arrived.

Join me, as I walk through the Schematic and PCB layout and then assemble and test my first card.

So, before we get too far ahead of ourselves, let’s start-off by taking a look at the schematic.

Firstly, I (once again) based the design on the MECB KiCAD Template, which I’d initially designed to make the design of subsequent boards easier and consistent.

So, as with my other boards, I started off with the MECB connector and Address Decode PLD already defined in the schematic, and laid-out on the PCB. As always, this is very helpful!

Here’s the completed Motorola I/O Card v1.1 schematic.

Adding the Motorola peripheral chips to the initial KiCAD template was straightforward, with each connecting up to a PLD chip select output.

Next was the Baud Rate Generator for the MC6850 ACIA, which I covered in detail in an earlier video.

Then, also covered in an earlier video, is the optional Programmable Timer Module Non-Maskable Interrupt driver.

This allows a Monitor program, that implements it (such as ASSIST09), to support single-stepping of your code and the setting of breakpoints, to allow for easy on-board code debugging.

Finally, I’ve used Port B of the PIA to optionally interface to a SN76489 Sound Generator. It’s optional, as I’ve also brought-out all of the PIA Port pins to a connector.

So, if the Sound is not desired, you can simply not insert a Sound Generator chip, so both PIA ports are available for external use.

So, let’s now take a look at the PCB layout.

Everything fits quite tidily on the (just under) 100mm square standard MECB card size.

Including connectors for all the I/O connections.

This includes, on the board edge, first a 24 pin right-angle male IDC connector bringing out all 20 I/O pins of the PIA. We also have a 6 pin FTDI standard TTL Serial connector for the ACIA. And finally, a 3.5mm Audio Out jack.

In addition, on the board we also have a 12 pin male IDC connector, to bring out the 9 optional interface pins of the PTM.

Finally, there is the single jumper for connecting the optional PTM Non-Maskable Interrupt connection.

And, of course, we also have the 4-way Dual In-Line switch for the ACIA Serial Baud Rate selection.

So, now let’s take a look at the manufactured prototype PCBs, that have arrived from JLCPCB.

The PCB looks great! So, let’s get our first one assembled.

For assembly, I started off with the bypass capacitors, and the two resistors.

Then I soldered in the IC sockets, including the 8-pin IC socket that I’d modified for the Crystal Oscillator, as I explained in earlier posts.

I then also soldered in the TO92 MOSFET and the 2 pin header for the Timer’s optional NMI connection.

Then, the 3.5mm audio connector, the right-angle header for the TTL serial interface, and the audio output electrolytic capacitor.

Then, the 4-way Dual In-Line switch for baud rate selection.

Then, the two I/O IDC connectors.

Finally, the male DIN 41612 MECB bus connector, which is secured in place with a couple of 10mm M2.5 bolts.

With the board assembly completed, it was time to get some IC’s inserted.

I started with the 74HCT393 Baud Rate divider.

Then, the three Motorola Peripheral chips. The SN76489 Sound Generator, and the 4.9152Mhz Crystal Oscillator.

Finally, I needed to program a PLD for the desired addresses assigned to the Peripheral chips.

As I’ll initially utilise the same ASSIST09 monitor that I was using on my 1980’s Wire-Wrap 6809 card, I define the I/O Address space on the CPU Card as E000 – E0FF.

The three Motorola peripheral IC’s are allocated an 8 byte address space each, starting at the I/O addresses 00, 08, and 10.

So this gives us the following device addresses:

• MC6840 PTM = E000
• MC6850 ACIA = E008
• MC6821 PIA = E010

This Chip Select address assignment is then setup in the PLD.

If you want to know more about the Minimalist Europe Card Bus use and programming (of the ATF16V8 PLD), then I recommend viewing my earlier post / video covering the MECB Address Decode / Glue Logic.

Here is our WinCUPL definition for our I/O Card PLD.

In the WinCUPL source you’ll first see the standard header and the Input pin assignments.

Then, in the second page of the source you can see the three standard chip select output pin assignments for CS1, CS2, and CS3.

These are all as per the standard PLD usage, that I originally defined for my Minimalist Europe Card Bus peripheral cards.

Finally, at the bottom of the source, there is the three lines of logic equations which define the chip select output logic.

Pretty straight-forward!

I then used WinCUPL to compile that source file into the required .JED file, for programming the PLD chip via my TL866 programmer.

Once the PLD was programmed, I then inserted it into our I/O card.

As mentioned earlier, for an initial test I’m using the same Motorola ASSIST09 Monitor that I was using in my original 80’s wire-wrap CPU card.

This will test both the ACIA and PTM peripheral chip’s operation, as both are used by the ASSIST09 monitor.

For this test I also installed the NMI jumper, to enable the PTM output to trigger a Non-Maskable Interrupt of the CPU.

Next, I program the original 2KB ASSIST09 monitor code into the top 2KB of a 28C256 FLASH ROM.

This ASSIST09 ROM was then installed into our 6809 CPU Card, replacing the ROM which held the TMS9929 Video Card test we had used in a previous post / video.

Finally, I checked that the CPU card DIL Switch was correctly set for the I/O Address space allocation of E000 – E0FF.

It was already set for this I/O Address space, as we’d used this same I/O address block for the TMS9929 Video Card, which was allocated an 8 byte I/O space starting at I/O Address 80, so effectively at memory address E080.

This address allocation also means that I can just use the existing Video Card alongside our new I/O Card, without any configuration changes being necessary (to avoid any address space allocation conflicts).

So with that all done, we were then ready for our first Motorola I/O Card test!

For the test, I first connected up the I/O Card’s TTL serial port to a TTL to USB adapter and then plugged in and also ran Tera Term on my PC.

With that sorted out, I then inserted the CPU Card and I/O Card into our backplane, alongside the existing Video Card.

Then, it was time to apply power and see what happened.

We got the ASSIST09 prompt, so we know the ACIA is working.

To test the PTM and the NMI Interrupt driver, we can verify that the breakpoint and single stepping functionality of ASSIST09 is working correctly.

To test this, I used the simple “Hello World” program that I demonstrated in an earlier post / video, which covered firing-up my old 80’s wire-wrap homebrew 6809 board. You’ll find that post here.

This earlier post / video also demonstrated both the operation and the available commands of the ASSIST09 monitor, so I won’t repeat that here.

So, I first issue the L (Load) command to ASSIST09. Then I used the Tera Term “Send File” option to send the assembled “Hello World” .S19 file to our CPU card.

This short program was assembled targeting memory location 0000, so by displaying memory address 0000 to 001F D 0 1F, we can see the program is present.

Calling the program with C 0, we can see the program run and display the expected “Hello World”.

Now, to first test that Breakpoints are working.

First I set a breakpoint for address 0000, so our program should immediately stop before executing the first instruction.
So, B 0

Then if we call the program again, with C 0, we see our breakpoint has worked. We have stopped with the Program Counter at 0000.

Now I set the Stack Trace level to the current Stack Pointer value, using the S command.

Then we try a single step using the period (.) command.

And we are looking good. The program counter is now pointing at address 0003, which is the second instruction.
The first instruction’s opcode (OP) was 308C followed by the 7 bit Offset Operand 05, which loaded the X register with the effective address of our “Hello World” string, being a Program Counter offset of 5, which is at address 0008 (PC = 0003 + offset of 05).

And, we can single step the rest of the program withe repeated . until we return from subroutine to the monitor (Opcode 39 – RTS).

So, Breakpoints and Single Stepping are verified working!

We’ve now verified our ACIA, PTM, and NMI Interrupt driver are all successfully working.

This just leaves our PIA and Sound Generator chips, which I’ll need to write some code in order to test.

For now, I think we’ll call it for today, with our Motorola I/O Card looking like a success.

At least, I now have a modular Minimalist Europe Card Bus (MECB) based 6809 system that updates my original 80’s wire-wrap card, with a fully configurable memory map. As currently configured, I have a full 56KB of RAM, a TMS9929 Video Card, and (subject to final testing), a PIA with a Sound Generator to complete a nice little retro 8-bit system.

Next, I might explore a different CPU card…

Discussion

Discuss this post on the Forum