SmartMotion® MEMS Motion Sensors

Hi,

I just learned that the MPU-9250 is obsolete and that the ICM-20948 is the replacement part. The parts seem to be pin compatible, but there is no documentation further that I could find on my own with more details about any h/w and s/w differences to take into consideration when replacing the MPU-9250 with the ICM-20948 in an existing design. Can someone from TDK InvenSense please provide me with this info? This is urgent as our product manufacturing is on hold due to this.

Regards,
Kishan

hello

i am using an ICM20948 on a custom board. the goal of our application is to use the Orientation data.

we are facing a problem with the YAW angle. this angle is changing if the board is moved from horizontal to Vertical position ( Pitch angle or roll angle changing).

is there any configuration to make to get the YAW angle the same even if the Roll or Pitch changes ?

regards.

Zakaria.

I write a code to set the ICM20948 for accelerometer & gyro reading and it is work fine. Now I want to leave power on all time but place it in sleep when I don't need it and get the current as low as possible. According datasheet if I get into sleep mode it should take about 8ua. I get 1.5ma
I did those steps when I want to get it into sleep mode
1. Disable Acc and Gyro.
2. Place them both in cycle mode which is the low power mode need for using the LP digital mode.
3. Active LP Mode.
4. active sleep mode.

Hi,

I am currently experimenting with the MPU-9250 9-axis motion sensor while waiting for an ICM 20948 to arrive, and I have some questions about the behavior that I've found with the SPI implementation. I haven't yet found detailed enough documentation to answer these myself.

I am initially using the Python spidev package for experimentation, specifically its xfer methods that take a list of SPI register addresses to read from.

Hi
I have a problem with magnetometer. I am following the code of the example eMD-SmartMotion-ICM20948-1.1.0-MP. In the setup I programmed this code:

I apologize if this has already been addressed (I tried a forum search and did not get any hits). It appears that the values for ACCEL_FS_SEL in the ACCEL_CONFIG0 register are given incorrectly in the ICM-42688 Datasheet. The values for 0 through 4 should be 32G down to 2G. The PDF currently states that the maximum range, for a index value of 0, is 16G, when it should be 32G. I have verified this empirically.

I am developing an application based on the ICM20948_eMD_nucleo_1.0, on a nrf52832 board.

The application is working fine but I have an issue. In order to accelerate the automatic calibration of the compass, you have to wave the board in a figure 8 motion for at least 30 sec. From what I understand, if I read the compass biais after calibration and reload them at each start, I won't have to do the whole calibration process.

I build application base on SparkFun software where I made adaption to nRF52 Nordic family.
In some of the setting I need to first read the reg and then make the changes in the require places , then write it back.

Hello,

I am just following the normal calibration process of averaging the samples after power up and subtracting them from the subsequent samples.

1. Now there is something called as bias calculation which includes dividing the averaged offset value by 4 for gyro and 8 for accelerometer. I'm not able to follow-up on that.
2. And also I got a reference where the trim values are added up in the offset values.

Can Kindly anybody explain the need of this ?
Thanks in advance.

Hello

In the datasheet of part ICM20948 ii is stated that there is an interrupt generated by the DMP. But I can't find any reference to any API, or interrupt programing, or similar for the DMP.

Please, could you send us some information of how to communicate with the DMP, or configuring its interrupts or data?

Thankyou in advance, I waif for your feedback.

Bes regards
Eloy