Compass Calibration
Principles of Operation:
Compass Calibration
Contents
1. Compass System
Nortek's compass is a system including a 3-axis magnetoresistive magnetometer, a liquid tilt sensor, algorithms to convert data from these sensors into heading and instrument tilt, and a calibration algorithm.
Figure 1. Compass system.
The magnetometer measures the magnitude of three components of the earth's magnetic field, and the tilt sensor measures two components of tilt and detects up/down orientation. All Nortek products (with compass) combine this information to compute the instrument's tilt and heading, then uses the tilt and heading to correct measured velocities to earth coordinates.
Magnetoresistive (MR) magnetometers sense magnetic fields with thin strips of a metal alloy whose electrical resistance varies with a change in applied magnetic field. These sensors have a well defined axis of sensitivity and are mass produced in integrated circuits. Nortek's compass includes three MR sensors, one each for X, Y and Z components, plus a two-axis tilt sensor. The magnetometer that Nortek uses is produced by Nortek.
Each compass system is calibrated in the factory to quantify the characteristic response of the individual components and of the system as a whole. When it leaves the factory, each system can measure its tilt and the direction of its magnetic field vector accurately, anywhere in the world.
Users disturb the magnetic field near the instrument when they deploy. Adding a battery pack and mounting the instrument with deployment hardware adds magnetic materials that change the earth's field at the instrument. The user's compass calibration procedure quantifies this magnetic disturbance, and the instrument's compass algorithm then corrects for it to obtain accurate heading.
Nortek's MR compass is new, as of early 2003. Prior to this time, its compasses used flux gate (FG) magnetometers. The FG magnetometer also came with a built-in calibration procedure, which turned out to be largely ineffective. The calibration procedure developed for the MR compass also works for the FG compass, because the characteristic responses are similar. However, because the MR sensors are more stable, more linear and more sensitive, the MR compass should produce better results overall. The MR compass enables Nortek to improve its overall quality control because nearly all MR compasses will satisfy the requirements for accurate heading measurement.
2. Compass Calibration
The magnetometer calibration requires a special jig and procedure, and can only be performed by Nortek AS or NortekUSA. Users perform the compass calibration just prior to deployment to correct for the introduction of new magnetic materials. These magnetic materials typically include a battery pack and the mounting structure. The compass calibration procedure requires a single, slow rotation around the instrument's tilt axis. A rotation taking at least 60 seconds is sufficient.
For the calibration procedure to work, the compass and magnetic materials must remain fixed relative to each other. As long as this is the case, the calibration procedure can correct for magnetic field disturbances that are greater than the earth's magnetic field.
Because errors in MR and FG compasses are similar in nature, it turns out that our calibration procedures work with both magnetometers. With calibration, the FG compass meets our specification, and the MR compass appears to perform better than our specification.
3. Test results
We tested two new MR compasses and three FG compasses in an Aquadopp Profiler, after calibrating them with nearby magnetic materials. We used two different disturbances: a battery pack and a steel bar. The battery pack was chosen as a typical disturbance, and the bar (placed close to the compass) was chosen because its magnetic field should exceed the field of a typical deployment frame.
After calibration, we rotated the instrument (and battery or bar), comparing its measured heading against the heading from the rotating table. We performed the rotation with tilts of 0°, 5°, and 15°, using tilts in both the pitch and roll axis. Figure 2 shows the results of some typical rotations.
Figure 2. Example results of six rotations using an MR compass. Heading offset is the difference between the actual and measured heading. The three colors for each plot correspond to different pitch (results for varying roll were similar). The dashed lines represent the corresponding mean offsets. The top plot had a battery at the base and the bottom one had a bar near the center; the values in milligauss represent the approximate change in the measured magnetic field induced by the bar or battery. For reference, the earth's field is nominally 500 milligauss.
We performed a total of 48 tests similar to those in Figure 2, using the two MR compasses and two of the FG compasses (we excluded the third FG compass because it produced unacceptable results). The mean offset was 0.1° ± 0.6°, which is consistent with the mean offsets resulting from the experimental setup rather than the compass. For each of the 48 tests, we computed the rms error as the standard deviation of the error about its mean. The average rms error for the six tests shown in Figure 2 was 0.6°.
Figure 3 shows the results of all 48 tests. These results show that the compasses maintain good accuracy over tilts ranging from 0° to 15°, independent of what was placed next to the instrument.
Figure 3. RMS error for different tilts for both the MR and FG compasses. The black line gives the mean and standard deviation for all four compasses at each of the tilts. The maximum error for a given rotation is about 40% larger than the rms error.
Sample Time Length
