IIS2DH Anti-Aliasing Low Pass Filter by Design in comparison to IIS3DWB and what are the resonance frequencies of each axis of the IIS2DH?

[FPal.1]

Hi,

so far we want to compare several MEMS accelerometers and their specifications. From ST, two sensors are on our list, the IIS3DWB and the IIS2DH.

For the IIS3DWB, the data sheet offers a frequency response curve at page 19 and regarding aliasing the following information (see IIS3-Block-Diagram-Frequency-Response.PNG):

"The IIS3DWB has been specifically designed to provide a wide bandwidth with very flat frequency response in the

pass band and a very high attenuation in the stop band so to virtually eliminate any frequency aliasing."

1) So for the IIS3DWB their is no "true" analog low pass filter before the ADC in order to prevent aliasing, its simply done by the frequency response of the senor itself right? Because the LPF1 is a digital low pass filter, in theory not capable of preventing aliasing effects.... Or am i on a wrong direction :) ?

2) Does the frequency response for the IIS2DH provides a low pass curve progression by the sensor itself too in order to eliminate aliaising, simillar to the IIS3DWB?

The IIS3DWB on page 5 offers the following information about the resonance frequency for each axis (see IIS3-Axis-Resonance-Frequencies.PNG).

3) What is the resonance frequency for the axis of the IIS2DH?

Hope my questions are correctly written, because I am not a native english speaker. Thanks for your help.

Greetings,

Joe

PS: for me, it was not possible to add inline pictures (error: "Object of InlineImage message segment is missing expected property...") ;) maybe i did something wrong or its an issue while submitting...

[FPal.1]

I would expect a signal chain for the IIS2 and IIS3 similar to this one....

[Eleon BORLINI]

Hi Joe @FPal.1​ ,

No, there is no analog anti-alias filtering just before the ADC of the IIS3DWB and IIS2DH.

In both cases, the mechanical structure of the accelerometer MEMS sensor acts as a low pass filter, preventing a significant fraction of the aliasing issue.

The fact is that the anti-alias filter is -at a practical level- limiting the bandwidth of the sensor, and for the vibrometer having a broad band (up to the 7kHz resonance frequency) is fundamental.

-Eleon

[FPal.1]

Hi Eleon @Eleon BORLINI​ ,

thanks for your quick reply ;) I had thought so that the mechanical structure of the accelerometer would act as a low pass filter. But for me it was important to have this confirmed again, thx.

Am I correct that LPF1 for the IIS3DWB is neccessary to limit the signal bandwidth on less than the resonance frequency of the sensor axis? Because an actual gain caused by resonance is not desirable and could be misinterpreted?

Do you have any updates regarding question 3) ? (resonance frequency of each axis of the IIS2)

Thx in adavance.

Joe

[Eleon BORLINI]

Hi Joe @FPal.1​ ,

Although it is a simplification, your guess has a grain of truth, but I cannot disclose the internal device architecture.

I previously missed the question 3), sorry:

>> 3) What is the resonance frequency for the axis of the IIS2DH?

It's above IDR/2 when ODR = 5.3 kHz ;)

-Eleon

[FPal.1]

Hi Eleon @Eleon BORLINI​ ,

thanks for your answer and your information so far ;) I totally understand that it is not possible to disclose all of the internal architecture of your sensors.

My work right now as a PhD student from the University of Applied Sciences in Karlsruhe, Germany, incorporates a comparison about different MEMS sensors and their specifications. The aim is to work out requirements for the sensors for vibration analysis. But so far, every data sheet - as you might know - and sensor architecture is (totally) different and sometime not easy to compare.

I have two questions left (hope thats the right place to ask):

1) Does ST provide any whitepaper (or similar) about Aliasing and the signal chain architecture for MEMS sensors in general?

2) Is it possible to get a contact email address if I would ask for sample sensors and eval-kits to test?

Greetings,

Joe

[Eleon BORLINI · Accepted solution]

Hi Joe @FPal.1​ ,

thank you for your the explanation of your research purpose.

Let me try to answer to your questions:

1) Does ST provide any whitepaper (or similar) about Aliasing and the signal chain architecture for MEMS sensors in general?

Unfortunately, these are confidential information and I cannot disclose them :(

2) Is it possible to get a contact email address if I would ask for sample sensors and eval-kits to test?

You can find our free samples on the page itself of the product you need or you can also insert a request for purchase on the eStore page searching for the specific component.

-Eleon