Is it an error in the datasheet of lsm6ds3tr-c?

[BChao]

As the datasheet wrote, the full scale of angular rate measurement range is 2000 dps, and the output value is a 16-bit word in two’s complement, so when we set the angular rate measurement range to 2000 dps, the sensitivity (unit mdps/LSB)could be caculated : 2000 * 1000 / 32767 = 61.03. （1 dps = 1000 mdps）

But the datasheet wrote it is 70 (Typ.) Why it is 70 but not 61.03？

The same way we caculate the sensitivity of acceleration(unit mg/LSB): 16 *1000 / 32767 = 0.488, it match the value wrote in the datasheet.

[BChao]

the same way we caculate the sensitivity of acceleration(unit mg/LSB): 16 *1000 / 32767 = 0.488, it match the value wrote in the datasheet.

[Eleon BORLINI · Accepted solution]

Hi @BChao​ ,

your calculation is right.

Basically, the fact is that the sensitivity for the gyroscope is a little higher than the LSB calculated on the basis of the "pure" full scale. This is motivated by design for guaranteeing the linearity of the calculation of the LSM6DS3TR-C angular speed: in practice, the actual full scale (the one declared in the datasheet), is lower than the theoretical full scale to avoid effects due to "high-values" (i.e. close to the full scale nonlinearities.

So please consider the declared sensitivities for your calculations. See also these Github C drivers lsm6ds3tr-c_reg.c.

If my reply answered your question, please click on Select as Best at the bottom of this post. This will help other users with the same issue to find the answer faster. 

-Eleon