Wednesday 30 September 2009

29th September 2009 - Remote Compass

Like all tubular metal aircraft, it is difficult in the Tailwind to find a site for a normal compass where it isn't influenced by the magnetic field from the airframe. This will probably be particularly the case for my build as I TIG welded the airframe and TIG is known for leaving steel magnetised. The solution is to have a remote compass situated in one of the wooden wings well away from the fuselage. Most modern electronic flight instruments (e.g. Dynon) have an option for a remote compass but they are expensive and I want to stick with the conventional panel layout. So, I decided to build my own remote compass. The picture shows the actual remote unit I've constructed; for reference the holes in the PCB are 0.1" apart so the entire unit is about 1" x 2.5" and weighs less than 1/2oz. Eventually it will be mounted in a small plastic box and screwed to one of the wing spars. The heart of the unit is an amazing chip from Honeywell - the HMC6343. This has in it 3 magnetometers to measure the earth's magnetic field and 3 accelerometers to measure gravity. These together allow the unit to not only report magnetic north irrespective of the orientation of the unit, but also to output the pitch and roll angles of the unit (and therefore the aircraft). The chip communicates using the I2C standard defined by Philips which uses a two wire serial bus. The four pin connector includes the serial bus wires together with a ground and 5V supply. It will be controlled by a microprocessor circuit installed behind the panel and the compass direction displayed on the display screen in the panel above the radios http://tailwindbuild.blogspot.com/2009/09/14th-september-2009-panel-painted-and.html and also on a separate dedicated 3 digit LED display - location to be determined. The actual orientation of the sensor is not important as calibration of the unit is extremely simple, first tell the chip whether it is on edge or flat, then point the aircraft in one known direction, compare the direction with the compass heading to calculate the deviation and program this into the chip, everything else it does itself and Honeywell guarantee less than 2 degrees of error between the compass output and the true heading with a repeatability error of less than 1/2 degree.
Posted by Picasa