Questions about filtering & thresholds in the LIS2DH

[IC.1]

I have a question about the high-pass filter mode configuration. With the configuration at 0x00, it states in the data sheet that the high-pass filter mode will be in normal mode in which it will be reset by reading the REFERENCE register. Assuming the normal mode it is referring to is the three modes (low-power, normal, high) I assume I can still use the high-pass filter while in low-power mode correct and read the REFERENCE register to delete the DC component of the acceleration correct?

Another related question. What is an application of reading the REFERENCE register if I always have the high-pass filter turned on? Perhaps I have a misunderstanding of how it is used, as there doesn't seem to be any difference when calling it.

Also, I wanted to confirm that the thresholds for the interrupts are symmetrical around the zero-g level. For example, say my threshold was 1000 mg, my duration is 20 ms, and I am sampling at a rate of 100 Hz. If the accelerometer experienced a threshold of 1500 mg the first sample, and a threshold of -1500 mg the second sample, is my understanding correct that this would generate an interrupt? If this is true, would it matter which axis it experienced the acceleration in (for example, it experiences 1500 mg in the x direction and then -1500 mg in the z direction)?

Thank you for the help.

[Eleon BORLINI]

Hi @IC.1​ ,

About the REFERENCE topic, I suggest you to have a look to other similar products, such as the LIS3DH and the related application note AN3308. In particular, the use of the Reference register is detailed in page 11:

It is recommended to read register REFERENCE (26h) when the device operating mode is switched from high-resolution configuration (HR) to power-down mode (PD); this operation resets the filtering block before switching to normal/high-performance mode again.

and in page 18:

Also for the second question (about the threshold crossing) I suggest you to refer to the same app note, p.28. It is a bit different with respect to your statement, but it is correct that the threshold is defined symmetrically around the 0g level.

 

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

[IC.1]

Yes, I read both of these. I believe my questions were a bit more specific, which is why I consulted the forums. Thank you though.