Maker Pro
Maker Pro

Looking for 3-axis magnetometer chip

Status
Not open for further replies.

evol_w10lv

Feb 19, 2013
73
Joined
Feb 19, 2013
Messages
73

evol_w10lv

Feb 19, 2013
73
Joined
Feb 19, 2013
Messages
73
Xtrinsic MAG3110 - output data rate only 80 Hz
HMC5883L - 12-Bit ADC (I guess 16-bit would be more precise)

As I said, haven't work with them.. so I don't know how really good they are. Also think AMR (anisotropic magnetoresistance) type would be most appropriate for inertial navigation project. Any other opinions?
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,878
Joined
Jun 21, 2012
Messages
4,878
A three-axis magnetometer is not the best choice for a navigation aid. The Earth's magnetic field varies all over the map (literally!) with changes in latitude, longitude and altitude. Plus you need some mechanism to "null out" the effect of nearby ferromagnetic elements, assuming those remain fixed with respect to the sensor.

A three-axis laser-ring rate-gyroscope coupled with a three-axis linear accelerometer would be better, backed up by a GPS receiver as a sanity check because all inertial navigation systems are susceptible to drift. IIRC ring gyros are now commercial off-the-shelf items ready for incorporation into an inertial position-measuring system. If you cannot afford a ring gyro, there are some fairly accurate MEMS three-axis gyros available. And just about every cell phone has a three-axis accelerometer chip inside, so those should be inexpensive and available too.
 

evol_w10lv

Feb 19, 2013
73
Joined
Feb 19, 2013
Messages
73
Thanks for replying.
The aim is to use all three sensors (accelerometer, gyro, magnetometer). We know that cheap gyro sensors drift over a time significantly when standing still. So the task is to develop algorithm, that recognize the moment when gyro starts drifting and then use magnetometer as an aid sensor to make corrections.
Yes, unfortunately magnetometer calibration is needed every time when used.
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,878
Joined
Jun 21, 2012
Messages
4,878
@evol_w10lv : It sounds like you know what you are doing, and are aware of the limitations of the sensors required for an inertial nav system. Depending on your application and reliability/accuracy requirements, a GPS receiver could be much more reliable than a magnetometer to make corrections. Of course if this is an underwater nav system then GPS is not going to be an option. There, you might want to add gravitometers to identify known mass anomalies, which I believe the U.S. Navy already does to move our "boomers" around in a very stealthy manner.

BTW, gyros drift whether "standing still" or not. They also have angular acceleration limits if part of a high-performance mobile platform. The rate gyros we used with our B-52H defensive fire control system in the last century were caged and mounted in a 3-axis gun control line platform, which was also locked until a radar "lock and track" solution was obtained. Only then were the gyros un-caged and the analog ballistic computer enabled to steer the control line platform to aim the Gatling gun. The two-axis gun mount was slaved to the control line platform azimuth and elevation axes until radar lock was lost or the target destroyed (whichever came first). There was also a roll axis on the control line platform because in a real live-fire defensive situation the bomber pilot would be jinxing and doing whatever else he could so as not be a target. A typical defensive encounter only lasted a few seconds, hardly enough time to accumulate any significant drift for this application, but those Sperry rate gyros were also used in nav systems.

Sorry I can't provide you with any help on sensor selection.
 

evol_w10lv

Feb 19, 2013
73
Joined
Feb 19, 2013
Messages
73
... a GPS receiver could be much more reliable than a magnetometer to make corrections.
Yes, but the main aim is to develop system that works when GPS not available (underground parking zones, signal obstructed by tree canopy, buildings, other obstructed areas, etc).

Also, when using magnetometer, we can determine heading respect to magnetic north. Unfortunately magnetometer is not the most reliable thing ever even after calibration... Probably fluxgate type could be most reliable (http://www.gmw.com/magnetic_measurements/Bartington/Mag-03_Pricing.html), but high price and not very small. I guess AMR could be best inexpensive variant. It would be great to hear expert opinion before buying the sensor.

Sorry I can't provide you with any help on sensor selection.
That's ok.
 
Status
Not open for further replies.
Top