Hi, we want make meassurements around 10uT using the sensor LSM303D. The measurement is static, so f=0. Which uncertainity we have to expect such as gaussian noise etc.? Thanks...

KArar.1 · ‎2020-06-10

Eleon BORLINI · ‎2020-06-12

Hi @KArar.1 , considering a ±2 gauss FullScale, 10uT = 100 mgauss are sensed with a sensitivity of 0.080mg/LSB. From characterization tests, the standard deviation of such 100 mgauss measure is 0.27 mgauss (say 0.3 mgauss), so you can consider that the distribution obtained from reiterated measures of a field of exactly 100 mgauss (with the magnetometer's hard- and soft-iron compensated) will have a gaussian shape centered around 100 mgauss and with a sigma of 0.3 mgauss. Btw, which physical phenomenon would you like to measure? Regards

KArar.1 · ‎2020-06-15

Hi Mr.Borlini,

Thank you for your explanation. Basically we need to know magnetic noise density. As we understood, it is 5mgauss/rms which is 5uT with sensitivity 0.080mg/LSB (Is it true?). However, isn't this noise density too much for a measurement around 10uT? We will apply a monte carlo method to estimate the error rate of our simulation from matlab? That's why we need to know all uncertainties including magnetic noise density and also we need to know 1/f noise (pink noise). Thanks...

KArar.1 · ‎2020-06-15

As you mention before, we will use gaussian distribution too:)

KArar.1 · ‎2020-06-16

Hi Mr Borlini,

So what is magnetic noise density (5mgauss) here? Is it lowest detectable signal? Moreover what kind of uncertainties are included for standard deviation? Thanks...

Eleon BORLINI · ‎2020-06-17

Hi @KArar.1 , but 5mgauss are 0.5uT... Regards

KArar.1 · ‎2020-06-17

Hi Mr.Borlini,

Sorry for typo. Yes ıt should be 0.5uT, but stıll ısn't ıt too hıgh as a noise for a 10uT signal? Kınd regards...