How does calibration work and what do the numbers mean?

[Vishal Shah]

When you press calibrate, a known magnetic field is momentarily applied with the internal coils to calibrate the magnetometer response. The calibration process yields a calibration number which is multiplied with the raw output in a way that scales the analog output to be 2.7 V/nT (in default mode) and the digital output to be represented in pT.

In the high gain mode (3x), the analog output is 8.1 V/nT and in low-gain mode (0.3x) the conversion factor is 0.9 V/nT

If the calibration number shown on the user interface is small <1, then the magnetometer is relatively healthy, but if the displayed number is high (>1), then there is some problem, most likely residual field or the gradient field too high. The calibration number itself is arbitrary and it is capped at 15.9. 

If the magnetometer is left uncalibrated, the calibration factor is set at one, and the raw uncalibrated output of the magnetometer is displayed on the digital as well as the analog port. 

[Ekaterina Skidchenko]

I am not quite sure what is meant by "known magnetic field". Could you please explain that? And another question: Is a minimum of 2 magnetometers required for calibration or one magnetometer can be calibrated independently?

пятница, 12 апреля 2019 г., 4:34:26 UTC+3 пользователь Vishal Shah написал:

[James Osborne]

Hi Ekaterina,

By "known magnetic field" we mean that we apply current on the internal coils and then measure the response of the magnetometer and this provides the calibration factor. Each sensor is calibrated independently.

Best regards,

James 

[Ekaterina Skidchenko]

In "QZFM Quick Start Guide" you stated that "calibration values less than 1.5 are considered normal and indicate optimal performance". What number (1 or 1.5) is the limit for healthy performance? Is it good or bad when the calibration value is exactly 1/1.5?

пятница, 12 апреля 2019 г., 4:34:26 UTC+3 пользователь Vishal Shah написал:

When you press calibrate, a known magnetic field is momentarily applied with the internal coils to calibrate the magnetometer response. The calibration process yields a calibration number which is multiplied with the raw output in a way that scales the analog output to be 2.7 V/nT (in default mode) and the digital output to be represented in pT.

[James Osborne]

The limit for healthy performance is generally a calibration of about 1.5. If the sensor is in a low ambient magnetic field environment, the calibration factor will be lower than if the internal coils need to cancel out a large ambient field. This is usually the most significant source of large calibration values (over 1.5). Sensors can be at optimal performance with a calibration above 1.0 (lower values are generally better), but around 1.5+ the performance is almost certainly not optimal. 

[Ekaterina Skidchenko]

Are the calibration numbers somehow connected with the value of the dynamic range of sensors? Do the Gen 2.0 sensors obey the same rule of calibration numbers for Gen 1.0 to be less than 1.5 for optimal performance?

пятница, 12 апреля 2019 г., 4:34:26 UTC+3 пользователь Vishal Shah написал:

When you press calibrate, a known magnetic field is momentarily applied with the internal coils to calibrate the magnetometer response. The calibration process yields a calibration number which is multiplied with the raw output in a way that scales the analog output to be 2.7 V/nT (in default mode) and the digital output to be represented in pT.

[James Osborne]

The calibration numbers do not have an effect on the value of the dynamic range of the sensors, the dynamic range value is fixed. The Gen 2.0 sensors do have the same rule of a calibration factor above 1.5 would likely indicate less than optimal operation. 

[Ekaterina Skidchenko]

The calibration number of one our Gen1.0 sensor was initially a bit higher than the calibration numbers of other sensors (approx. 1 against approx. 0.5) and was increasing during experiments to approx. 1.4. Is this OK?

пятница, 31 июля 2020 г., 21:24:18 UTC+3 пользователь James Osborne написал: