Hmm... I am not sure if this calibration approach would work for magnetometer. This works for accelerometer since Earth's gravitational acceleration is constant (1g) as long as you don't accelerate/decelerate. Although, earth's total magnetic field is constant at a given location on the earth, your magnetometer can be easily affected other ferrous materials in the neighborhood and it is very difficult to factor them out from just Earth's magnetic field. The correct approach to calibrating magnetometer involves placing it inside a mu-metal shield to measure the bias voltage (output at zero input field) and then placing it inside a Helmoltz's coil in different orientations while given a known input (eg. 0.5 gauss) to calculate scale factor (output for incremental change to the input). Regards, Giri
________________________________ From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of sougata sen Sent: Wednesday, September 19, 2007 5:47 AM To: Urs Hunkeler Cc: Tan Simin Jeremy; [email protected] Subject: Re: [Tinyos-help] 2-axis Magnetometer MTS310CA hi jeremy adding to Urs' comments to caliberate, u should get a +1/-1 mGauss when the +ve/-ve X/Y axis is facing towards the earth or vice versa. if that is different, u have to caliberate the values to get this reading. On 9/19/07, Urs Hunkeler <[EMAIL PROTECTED]> wrote: Hi Jeremy, I see you have been posting this message several times and apparently have not received a reply. So here is my unqualified reply. Maybe this gets somebody so annoyed that they just want to correct me :-) 1. I think the sensors need to be calibrated. For instance this could be done by turning the mote slowly around while simultaneously sampling the magentic field. That way you'll get a maximum and a minimum value corresponding to perfect (mis)alignment to the earth's magnetic field. How you do this exactly I don't know (but would like to learn as well). 2. Based on the calibration you need to adjust the on-chip amplifier. There is are some functions in the Mag interface that seem to allow exactly this: gainAdjustX and gainAdjustY. 3. It should. But I seem to vaguely remember that the magnetic sensors on the MTS310 sensor board only measure differences in the magnetic field, so you could not measure a static field. I hope this helps a bit. Please let me know if you find out more about it. Maybe you could contribute these findings then also to the TinyOS wiki? http://www.eecs.harvard.edu/~mdw/tinyosdocs/index.php/Main_Page Cheers, Urs Tan Simin Jeremy schrieb: > Hello ppl, > > I'm currently working on the Honeywell HMC1002 Magnetometer which is located on > the MTS310CA sensor board. However I'm facing a few problems and hope that you > will be able to assist me. > > 1. Why is it that my readings for the x and y-axis are different when I use > different sensor boards (MTS310CA) on the mote? > > 2. My readings for the x-axis is changing but the readings for the y-axis does > not change in the room. WHy is this so? The y-axis only changes when I bring a > magnet near the magnetometer. > > 3. Will the magnetometer still give us a reading if we do not bring a magnet > near it? I'm sure there is still the earth's magnetic field present in the room. > > Thank you for your help. Have a blessed week ahead! > > Cheers, > Jeremy _______________________________________________ Tinyos-help mailing list [email protected] https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help -- Sougata Sen Bangalore
_______________________________________________ Tinyos-help mailing list [email protected] https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
