4. Implementation

Building this project was very enjoying. It was also time consuming, and since I didn't have too much free time, I usually ended up soldering at nights for 1 to 3 hours each session. I couldn't afford to work every day so it ended up working for ~10 hours per week. The more I felt I'm getting closer to the end of it, I added more effort and worked more time per day. I estimate the work at about ~40 hours.In-between I started writing the code for the PIC microcontroller. I decided to use the assmebly language because I was already familiar with it when learning about PIC programming, and the MPLAB as a free environment also includes a free assembly compiler - I didn't have any free C compiler (not a demo or trial versions). I also thought that it is good to learn about PICs using assembly - it is exactly shows you how every bit in the microcontroller works - no hidden stuff under the hood (or language compiler). I tried testing it using the MPLAB's simulator which can even simulate COM on RS-232.

When I started building this project, I got myself three kind of boards:

Breadboard - This was used at the begining to test a few concepts by combining some parts, and later on as a test bench for components (testing if a transitor worked or a led was broken etc). It's also good for measurments etc.

Stripboard - I tried building a little circuit on it and eventually decided it is very difficult - The thing is that strip board is made out of long copper stripes. It's somehow a bit similar to the breadboard, only when you build a project that involves many components, it's very difficult to design their position on the board, and you need to make cuts on the strip using a knife and also use many wires to connect among strips. I abndoen it.

Perfboard - Perforated board - This board has copper only around each hole - each hole is isolated from the other holes (unlike stripboard), so if you want to make connection between holes you can use wires, but you can also use soldering to create long strips. I used this technique a lot in my circuit, so most of the connections are actually made under the board - On the surface above you see mostly the components and sometimes wires.

It's a matter of practice to decide when you should use soldering to create stirps and when you should use wires.

I already knew, by my scheme, that there are actually 3 circuits involved, but I decided to put all of them on the same perfboard. The perfboard size I decided to use is 230mm x 100mm. I followed the layout of my schematics and tried to build it using some visual estimation - I can't say it's the best engineering way of doing things, but I didn't use a software for circuit layout and it was very difficult if I would have to measure all the componets by hand.

Although all the circuits were actually laid on the same board, I actually treated them as separated, so each circuit ends up with a connector, and starts with a connector - I later made cables that connect each circuit to the other. In fact, I also drew the borders of each circuit, and even today, If I'm brave enough, I can just cut along the lines and get the three circuits apart - everything will still work because there's no a single hard connection between them!

(But I won't do it - I'm afraid of breaking the perfboard while cutting).

To be continued...