Sunday, January 31, 2021

The inevitable MC68000 Single Board Computer

Since the mid-1980s and the announcement of the Sinclair QL, which 'never happened', I have always been interested in the Motorola 68000. My means did not allow me to buy a Mac, but I could have buy a QL. In short...


OldComputers.net

 And then the time passed and I got down to the PC. So I never owned, either, atari of the ST range :

Wikipédia

Then the years passed again and in the early 2000s, I decided to develop my own 68000 board. Nothing very complicated. A 68000, RAM, ROM, real time clock, LCD display and serial port plus a small extension connector. Which gives this schematic: 



This schematic has some small errors, especially on the side of the MC68B50 serial converter. However, nothing serious since once corrected on the board, the system worked fairly quickly.

The PCB : 


Originally, this board was designed to operate at 10MHz. Once the operation was validated, I have not touched this system since 2004. It has now been over 15 years. 

Since a few months, I decided to take up the subject of processor boards. This is the reason of this blog dedicated to retro-computing, with current technology.

Fortunately, I kept all of the project files. In addition, all the logic of the board is contained in a PLD so that it is possible to modify the behavior of the system. 

Wen I wanted to put this board back into service, I noticed that it no longer included the 10MHz oscillator. 

Impossible to find him. On the other hand I had a version of 16MHz. So I looked to see if it was possible to adapt the board to this new frequency.

As I am using EPROM emulators with very short access time, I have not changed the DTACK delay. I only modified the generation of the Baud Rate and took the opportunity to go from 9600 baud to 19200 baud. 

At 16MHz, the clock speed is a little too fast for the 68B50 which is given for 1.5MHZ. Well, 1.6MHz should pass. 

And here we are!



I was able to reprogram without difficulty the circuit EPM128SLC84-15 with version 13.01 of Quartus II. I short, the programmable circuit diagram with Quartus :




By resuming the study of this circuit, I noticed with surprise that I had used the graphics method plus the Verilog coding. I don't use these methods at all today. I go exclusively through VHDL coding. 

Anyway the modification of the source code took me only a few minutes, programming of the CPLD included. 

The modification of this system was also an opportunity for me to test the new Momik ROM emulators :

(Fee advertising)

This emulator can emulate up to 27C040 type EPROMs. Originally this board was intended to work with AT29C040A type Flashes. I had to accommodate the differences between these two types of circuits. 

The result is that this system works great at 16MHz with two of these EPROM emulators. After successfully restarting this board using the two Momik EPROM emulators, I programmed two FLASH AT29C040A for fully autonomous operation this time. 

In fact, this is the first time this microcomputer has operated in this way. I had never inserted a real AT29C040A until now : 



Despite the fact that the 68000 used is a 12MHz version and that with a frequency of 16MHz, the serial converter is also slightly over clocked, everything works perfectly. I am satisfied with the result. And the hours spent studying this system at the time are not wasted!


No comments:

Post a Comment