ATMega 32

An ATMega32 board has finally arrived in one piece and it seems to work. Well at least the LED lights come on when the power is attached. I'm going to have to go through a big learning curve to get this processor to do my bidding for me. On initial investigation it looks very possible to have the board monitor three temperature sensors, one in the flight box, one external and one with the camera as well  as driving the servo controller. To conserve power in the camera the ATMega32 will also control the power to the camera using a relay. This will have the added benefit of resetting the camera each time.

Had a breakthrough on the issues with the RS232 cable. I complied up the code for the Gumstix Verdex and installed the package and discovered that the servo control code worked perfectly with the supplied serial connection that came with the Verdex. This leads me to believe that either the serial cable from Maplins is faulty or there is a problem with the RS232 port on my Linux box. Taking the voltage drop between the outer casing on the cable and the Pololu servo contoller gives a 63V AC current which strikes me as wrong. It is about 3V with the butchered serial cable and I haven't yet measured the difference on the Verdex serial cable. The next step is to get the canon capture software compiled and to construct a cradle for the camera out of aluminium.

The Beginning

Well I have got the first bits of kit together and the trick now is getting them to work. The first exciting part to arrive was the very small Gumstix Verdex XL6P computer. The size of this small embedded Linux computer has to be seen to be believed! The imagery on the Gumstix website really doesn't do it justice.

The next bit of kit to arrive was a Futaba 3003 standard servo and a serial Pololu servo controller. Initially the servo is to be controlled by the Verdex via one of the serial ports thought if possible the control will move to an ATMega32 board to release the serial port on the Verdex. The servo will be attached to the side of the camera and will provide movement in the vertical plane. This will allow pictures to be taken of the horizon, straight down and straight up.

The development of the code to drive the servo has turned out to be a little more complicated than I anticipated. It must be a good few years since I last coded anything in C and programming the serial port communications is not the easiest place to drop back into it. I have reached the point where everything seems to work fine with a butchered RS232 link cable but when I attach a standard straight through cable it goes to pot. I think that it is probably something to do with more of the pins been connected straight through. There is a bit of code I can put in to make sure that these extraneous connections are turned off. However this involves setting lots of flags and using ioctl functions in C which I know little about and there is scarce documentation on the internet about programming the serial ports.