by SV3ORA
Introduction
I always wanted to build a simple computer system out of common TTL chips but the complexity and cost was the main obstacle. Eventually, I found a project on the homemade CPU ring which described the construction of a 4-bit CPU out of TTL chips. Despite most of such articles are quite complex to understand and confusing to navigate through their pages, this article was so well written and so complete that I could entirely understood how the CPU worked on a single website reading. This inspired me to build my own version of it.
This CPU used mostly common TTL chips apart from the EPROMs and the RAM chips. I decided to build my own militarized version of it first and later on to try to make modifications to it. The main modifications which I felt I needed to make, were to replace the EPROMs and RAM, to use entirely TTL or more standard parts. This is a project that is probably going to take years to complete properly.
This is a slow and very limited usage computer system and it has been basically rebuilt by me for educational purposes and components reliability demonstration. There are not such things as keyboard and screen, if you want a such computer, build the ZX-80 that I have in another section on my website, or other such computers that I will include on the website. You may think of it more like a microcontroller rather than a computer. It is programmed to do a single job, using machine language code entered mainly using front panel switches, much like the MITS Altair 8800. There is no prepackaged CPU though. The CPU is mage out of common TTL chips.
There is no need to repeat everything here. Download an read the project articles first, to understand how the computer works. Then follow my page below to see how to make a rugged version of this computer, that uses more standard parts, it is failure immune and can work for many years of reliable operation.
Electronics Parts List
There was no electronic components parts list for this project. There were some gates that were not marked in the schematics the author provided, so I decided to make my own parts list by carefully examining these schematics. Since I was interested in immunity to failures due to physical factors, I used more rugged components. Here is the parts list including the "missing" TTL chips. The actual components used are listed. You may substitute some of them with your versions if rigidity is not your goal. For example you can substitute the F series semiconductors with the LS series and the military switches with cheap ordinary ones.
Part | Quantity | |
10uF | 1 | |
0.47uF | 1 | |
0.001uF | 1 | |
1-pole 3-position switch Grayhill 3 position switch. Model: 51MY23691. Has 12 tabs and center ground. NSN: 5930-01-304-4991 Position 1: Pin 2 to center ground. Position 2: Pin 3 to center ground. Position 3: Pull center rotor and rotate. Pin 1 to center ground - other positions locked out. |
2 | |
Push-to-make program
write switch. Auto-return switch Cutler Hammer, Part Number: 173K202A13 NSN: 5930-01-336-7686
National Item Identification Number
Reference Numbers (Part Numbers) |
1 | |
ON-OFF power switch
NKK locking switch |
1 | |
ON-OFF data,
instruction and address switch.
Torbal DP3A or DP3. VERTEX INDUSTRIES INC Part numbers: DP3A |
16 | |
LED | 32 | |
0.1uF | 30 | |
0.22uF | 4 | |
Beckman 898-3-R300 (or 899-3-R300) | 1 | |
Beckman 898-3-R3K (or 899-3-R3K) | 1 | |
Beckman 898-3-R4K7 | 2 | |
Beckman 898-3-R1K | 4 | |
54F244 | 6 | |
54F240 | 3 | |
54F32 | 1 | |
54F08 | 3 | |
54F193 | 1 | |
54F85 | 1 | |
54F670 | 1 | |
54F175 | 4 | |
54F624 | 1 | |
54F93 | 1 | |
54F161 | 1 | |
54F374 | 1 | |
54F74 | 2 | |
2114 (or 2112) | 2 | |
27C16 | 2 |
Schematics and PCB
There were schematics on the author website but these were only rough images, pin numbers, coupling capacitors etc are missing. Also there was no PCB for the project. I contacted the author if he had any CAD files, but I did not get a response. For these reasons, I have decided to make my own PCB. My version will include monitoring of the I/O with LEDs as well as some other possible changes.
There was a great decision I had to get in building this computer. Build the whole thing onto one PCB or split all computer modules into different PCBs. Splitting modules according to their role on the computer system gives greater flexibility. One can change modules by replacing them with more updated versions, as long as I/O connections stay the same, without having to replace the whole PCB of the computer. On the other side, lots of cabling is required between different modules. We are making a PCB for this computer because we want easier construction with less cabling, which will lead to a more rugged system. Why to build PCBs if you have lots of cables to interconnect them?
If you build the project as a KIT, then you should include everything in one PCB for cost reasons. But if you build the project only once like me, you should better split the modules into separate PCBs. This will allow you to alter them and experiment later on, without having to rebuild the whole computer.
Regarding the PCBs, I have tried to include as much traces at the bottom layer as possible. That way, if one decides not to build double sided PCBs, he could join the top layer traces using wires. This ensures that reproduction can take place even if an advanced PCB fabrication lab is not available.
Anyway, here are the PCBs and my modifications for the computer so far:
Programmer
ExpressPCB CAD file
Top copper layer (PDF)
Bottom copper layer (PDF)
Silkscreen layer (PDF)
Modified programmer schematic (WRITE switch disabled during RUN mode)