Since I'm designing one board, why not design two? I give you the "TestFlight" board.
What's a TestFlight board? I am glad you asked. It allows you to test your Mega32 code on a general purpose system without having to build the rest of the RodneySPR system first. With a little bit of buffering you can drive LEDs, relays etc and test code.
It's also a good way to test things before committing the final RodneySPR PCB Design. Maybe there are problems which we need to fix, and once you have PCB's it gets harder to change things.... read more
In the files section, you will now find a folder called "PCB". Underneath this will be a collection of numbered folders which correspond to hardware versions.
I have just uploaded 001 to this tree. It contains zipped files for use with CADSoft Eagle. This is a prototype PCB design.
This morning I had something of a breakthrough. The Alpha-1 algorithms have successfully run through on at Atmel simulator. The code has now been pushed to the git repository
Now, I realise this is a long way from RodneySPR laying down rubber on concrete, but it's a good place to start. I can now safely assume that the project can work. All I need now is the CPU and mechanical bits to go with it.
I know there's been a hiatus in this project while I worked on my paying job, but now I have some time to allocate to this project so I am back into it again.
The first step is to finalise the electronics design for the system. I have decided to go out on a limb, and create a custom CPU board for RodneySPR, based around the ATMega32. This is fundamentally the same chip as the Mega16, but with 32K of memory instead. The main memory is implemented on-board via a 25512 Serial EEPROM chip, so we have 64K x 8 main memory, which is outside of the main CPU and accessed in a manner not dissimilar to the original design, albeit using SPI rather than parallel I/O.... read more
There are two streams of work going on at present:
At the moment, the new generation hardware is not final, so the new generation code is written to be very generic, with lots of defines to cater for any future hardware changes.
Because it might be useful to add an SPI EEPROM to the system, it's recommended that you keep PORTB on the ATMega16 free if possible. This means we can't expand to an ACTH port, but I don't think that's a problem at this point.