what is the sampling frequency in each axis for IIS3dwb digital accelerometer sensor

ASriv.3 · ‎2021-09-01

I see that ODR mentioned in datasheet is 26.7Khz but my doubt is that whether it is in each direction (x,y,z) or combined for all 3 axis. This is because when i try to measure the time taken to take samples then i get it around 9000Hz in each direction. I am using data polling technique via STM32 F4 controller

If it is so then how BW of 6.3Khz is achieved

Eleon BORLINI · ‎2021-09-02

Hi @ASriv.3 ,

the 26.7kHz ODR is for all the 3 axis, in the sense that you will receive a batch of (X,Y,Z) every 1/ODR, meaning in first approximation the .

You can configure the single axis mode so that you can get 26.7kHz on the single axis.

XL_AXIS_SEL[1:0] = xxb (00 = 3 axes; 01 = X-axis; 10 = Y-axis ; 11 = Z-axis)

The 6.3kHz bandwidth refers mostly to the frequency response flatness (±3 dB point) of the mechanical element, as you can see from the Figure 15., 16., 17. of the datasheet.

This means that the device can detect a vibration up to 6.3kHz without loss in sensitivity.

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

View solution in original post

Eleon BORLINI · ‎2021-09-02

Hi @ASriv.3 ,

the 26.7kHz ODR is for all the 3 axis, in the sense that you will receive a batch of (X,Y,Z) every 1/ODR, meaning in first approximation the .

You can configure the single axis mode so that you can get 26.7kHz on the single axis.

XL_AXIS_SEL[1:0] = xxb (00 = 3 axes; 01 = X-axis; 10 = Y-axis ; 11 = Z-axis)

The 6.3kHz bandwidth refers mostly to the frequency response flatness (±3 dB point) of the mechanical element, as you can see from the Figure 15., 16., 17. of the datasheet.

This means that the device can detect a vibration up to 6.3kHz without loss in sensitivity.

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

ASriv.3 · ‎2021-09-09

Thanks a lot for your feedback
So in case we take samples on single axis it should be 26.7K right?
But i tried to calculate using HAL_gettick() in STM where i was getting around 13k . Is this because we read 2 registers H and L for getting the samples.
Please confirm that Sampling freq is 26.7 KHz in one direction and if we are taking on all 3 axis then it is approx 26.7/3 Khz
Thanks

Eleon BORLINI · ‎2021-09-09

Hi @ASriv.3 ,

>> So in case we take samples on single axis it should be 26.7K right?

Right, this should be the case, not 13kHz.

So that's strange you get only this value. In any case, it is enough by Nyquist to exploit all the sensor band, which is 6.3kHz.

>> Please confirm that Sampling freq is 26.7 KHz in one direction and if we are taking on all 3 axis then it is approx 26.7/3 Khz

Yes, this is correct.

-Eleon