LSM6DSL + LIS2MDL noise in one acceleration axis

[Florin Stroiescu]

Hi,

I am using the setup with LSM6DSL + LIS2MDL. All works fine, with the exception that once started the conversion to FIFO with sensor fusion, at 104 sps, I observe significant periodical (around 22 samples cycles) noise on the last acceleration axis (Z). With LIS2MDL stopped there is no noise. More, if I disable the internal pull-up resistors, the noise disappears as well.

Any advice on this?

Many thanks,

Florin

[Miroslav BATEK]

First can you share your schematic?

[Florin Stroiescu]

Here it is, Miroslav.

Many thanks,

Florin

[Miroslav BATEK]

The schematic looks ok, but I have a few more questions:

• Is the capacitor C3 close to the LIS2MDL, the magnetometer generates significant peak current so the capacitor should be as close as possible.
• Which internal pull-up resistor has impact on the noise? If you disable the interrupt are you able to read data from the magnetometer.
• Is the INT2 pin configured as input in MASTER_CONFIG register (START _CONFIG = 1).

[Florin Stroiescu]

Hi Miroslav,

Thank you for your time, and valuable questions!

The answers are as follows:

• C3 is shared between LIS2MDL and LSM6DSL, just near the power pins, and it was not the problem (as detailed below);
• The pull-up resistors that are impacting the noise are the Master I2C internal bus pull-ups - again, this proved to not be the problem too;
• The problem was with INT2 configuration, I would say that the description in datasheet is a bit ambiguous, or at least it confused me:

"START_CONFIG

Sensor Hub trigger signal selection. Default value: 0

(0: Sensor hub signal is the XL/Gyro data-ready;

1: Sensor hub signal external from INT2 pad.)"

After configuring it as per your suggestion (OK, done at the right time in the initialization sequence of the assembly), all works like a charm!

Thank you!

Florin

[Miroslav BATEK · Accepted solution]

Good to hear.

Indeed, the datasheet is not very clear.

You can also refer to application note AN5040 which goes more into details.

https://www.st.com/content/ccc/resource/technical/document/application_note/group0/26/07/3f/bf/12/55/47/62/DM00402563/files/DM00402563.pdf/jcr:content/translations/en.DM00402563.pdf

[Florin Stroiescu]

Thank you Miroslav, indeed, the Application note is more clear, but all is easy after sorting it out :beaming_face_with_smiling_eyes:.

[devilish]

Hi can u please explain me how did u connect lsm6dsl and lis2mdl and what are the things u configured. Like how did u know what to configure. Also one last thing how did u plot.