Networking

Work in progress, running notes

I want to start with a baseline of a working hello radio. I've machined four boards. The first two I lost some pads. I was following the directions to set the cutter diameter to 0.38, however this is too small, which means more trace gets cut than should. I decided this is part of the reason i was losings pads. So on the third I set the diameter to 0.397. Unfortunately, on that one I forgot to load the outline program and i destroyed the board by using the 1/32 cutter on the traces path. doh. The fourth board i cut everything out fine, but then found i lost a pad going to the ground on the MRF IC. But, I think it will be salvageable. So I'm now cutting a second board, since i need some sort of communication network to test. If it comes out okay I'll begin populating the boards.

Some notes:

  • Looking at the datasheet, it recommends a ground via directly under the transceiver IC, hello-radio has it off to the side.
  • I want a direct read of the ARSSI pin. hello-radio does not pull out the analog signal. In digital mode it is a threshold measurement, not a linear analog value.

definitely worth cutting with sharp cutters. i spent a bunch of time clearing out little slivers of metal.

I've machined four boards. The first two I lost some pads. I was following the directions to set the cutter diameter to 0.38, however this is too small, which means more trace gets cut than should. I decided this is part of the reason i was losings pads. So on the third I set the diameter to 0.397. Unfortunately, on that one I forgot to load the outline program and i destroyed the board by using the 1/32 cutter on the traces path. doh. The fourth board i cut everything out fine, but then found i lost a pad going to the ground on the MRF IC. But, I think it will be salvageable. So I'm now cutting a second board, since i need some sort of communication network to test. If it comes out okay I'll begin populating the boards.



Placing Parts a note.
soldering the tiny 0603 parts near the antenna. they float away fast. it's best to put a drop of solder on opposing ends so heating up one component doesn't make the other one float away.

New tip on the solder iron makes a HUGE difference. i was finally able to get a "drop" of solder on the tip and heat transfer was lightning fast (well almost). this really worked super well.

Soldered everything. Vias I did by sticking solder through and melting on the groundplane side. I then melted the solder to the front side. This is kinda risky. Better to strip a wire, pass it through, solder it on each side, then trim it down. this way you don't risk melting away the solder from the middle of the board, since it is not inspectable.

Video

Programming
Now to try programming
I went to Amir's website:
While in the directory with all of your files type:

make

note: firefox/windows downloads makefile and turns it into makefile.txt. avrdudeI want to start with a baseline of a working hello radio.

I got an error:

avrdude: Device signature = 0x1e950f
avrdude: Expected signature for ATMEGA168 is 1E 94 06 Double check chip, or use -F to override this check.

Looking at my chip, it's actually the ATMEGA328P. I should've checked this, the drawer I pulled the parts from must have becoming mixed up, or I must have become mixed up. Luckily these two chips are nearly the same. now i just have more memory. But i need to update teh makefile.

uncomment line 6, and comment line 7 in makefile MCU = atmega328p #MCU = atmega168

Success with loading the data onto the chip.

my ftdi driver stopped working, i was hearing a triple beep(bounce) when i plugged it in and got no simulated serial port. I reinstalled the drivers and it is working again.
The system wasn't working. I noticed I was using an 8mhz system and the make file was programmed for 16, so I commented out line 16 and made a new line setting the freq = 8000000

I then received notification through serial that the chip was alive. i uncommented the printf statements inside hello_radio.c to find out what was happening
it looks like i get all the way into the code, (line 787, where we wait for a receive packet from a second device. so now i need to populate a second device.

Made a second board. it works. well the same as the previous one.

433mhz is a long antenna. so i updated line 94 in the code to switch to an 868Mhz band, since that gets me a decent antenna lenght of 8.6cm.

i think one of my transceiver crystals is not seated correctly. i'm not seeing a signal off of it on the sccope.
yup, i reflowed it with the heat gun and now i'm getting a 3.3 oscillation. and i'm getting communication. sweet!