SmartMotion® MEMS Motion Sensors

Can anyone shed some light on the relative merits of these two chips? We are currently using the ICM-20601 in a prototype app, but are wondering if we should switch to the ICM-20649 for production. Are there any advantages to the ICM-20649 in terms of performance, longevity, costs, etc?

Thanks,
Paul.

Hi, I'm trying to figure out why I'm getting 3.37 mA current measurement when I put MPU9250 to sleep. There seems to be a leakage current of some sort, since I read 6+ mA

My initial attempt was to simply write a 1 to the SLEEP bit of PWR_MGMT_1 register. Subsequently I've tried

* Disabling accel and gyro via PWR_MGMT_2
* Setting magnetometer to "Power-down mode" via Control 1 register
* Powering down the internal voltage regulator (PD_PTAT bit of PWR_MGMT_1)
* Stopping the internal clock (CLKSEL bits of PWR_MGMT_1)
* Disabling I2C Master and FIFO modules

Hi,
I have successfully implemented pedometer in my wearable device application and now I need to wakeup the device to
show time on tap or on a gesture, I tried this by enabling the tap detection feature along with pedometer,
but not interrupts are created on taps, Below given is my code.

Please tell me how to clear FIFO_OVERFLOW_INT in register of int_status (0x3A) when I use MPU9250 for get quaternion from DMP through I2C.

I’m developing an application for nRF52 and MPU9250.
I want to write code that can get quaternion from fifo-buffer through I2C.
But now, int_status is 0x13 every time when reading fifo-buffer at my program.
So int_status is 0x13, FIFO_OVERFLOW_INT is 1.
I think it is abnormal, because fifo-buffer is overflow forever.

Please tell me how to clear FIFO_OVERFLOW_EN in register of int_status (0x3A) when I use MPU9250 for get quaternion from DMP through I2C.

I’m developing an application for nRF52 and MPU9250.
I want to write code that can get quaternion from fifo-buffer through I2C.
But now, int_status is 0x13 every time when reading fifo-buffer at my program.
So int_status is 0x13, FIFO_OVERFLOW_EN is 1.
I think it is abnormal, because fifo-buffer is overflow forever.

Hi all,

When I use the ICM20602's accelerometer with everything on default (with the exception of disabled I2C, I get an output that is shown in ACCImage.jpg

However, when I disable the GYRO (to save some power) with writeICMREG(0x6C, 0x07), the accelerometer output changes to what is shown in ACCImage2.jpg

My setup is just a PCB with the ICM chip that's on my table and I am just knocking on the table to produce these waveforms.

Hi All,
If I put the MPU 9150 on a flat surface and the sensor side with the dot symbol facing towards the sky,

(1) Can I assume that my positive z is towards the sky? In that case, what is my acceleration value in z-axis. Is it +9.8ms^2? Please refer to the attached files.
(2) What is the corresponding register value?

Best Regards
Min Latt

I'm using a SPI interface to access the ICM-20948. I've been unable to communicate with the magnetometer via auxiliary I2C master mode and have a couple of simple questions.

1- Are the transactions with the magnetometer visible on the package AUX_CL and AUX_DA pins?
2- Do the AUX_CL and AUX_DA pins need external pullup resistors even when using the internal magnetometer?

Hey there,

I have a problem with the ICM 20948 driver: The function inv_device_setup of the driver package returns -1.

First the driver seems to flash the content of Invn/Images/icm_img.dmpa.h in register 7D where the bank selection of the DMP occurs with the write in register 7C. There is a validation of the flashed memory in inv_icm_firmware_load (line 63) which returns -1 in my case because the data_cmp does not contain the correct bytes.

Hi, I'm wondering why the limitation of 1.71 - 1.95V on ICM-20948 VDDIO pin, while VDD can go up to 3.6V?

I've tested ICM-20948 with 3.3V on both VDD and VDDIO and it works just fine. Just thinking if I'm driving the chip out of spec and cause damage in the long run, or if the datasheet is wrong.