MEMS (sensors)

From theory, we define a Finite State Machine (FSM) as an abstract machine that can be in exactly one of a finite number of states at any given time, and is composed of a START state, a STOP state and a series of states, conditions and transitions. But how is this definition practically implemented in LSM6DSO sensor?For the LSM6DSO sensor, FSM is intended to detect a specific gesture: there are up to 16 independent state machines, both XL and G (and external MAG sensor) data can be used, and a number of interrupts and source registers for the event detection management are available.The activity recognition spans from Free Fall, to Glance, to Flip up and down, passing through Pickup, Shake and Wrist Tilt.If you are interested in a detailed explanation of the LSM6DSO FSM structure, that enables you to develop your own machine, please follow the next posts!

Hi,I want to compensate for the temperature of the LPS22DF sensor.how does one go about programming for it? Is this formula correct?Pcorrected​=Pmeasured​−(TCF×(Tmeasured​−Treference​))If it is, should I use a temperature reference value of 25°C?Also, can you tell me the value of the Temperature Compensation Factor? Can I make a correction formula with TCO(Temperature coefficient offset) with a data sheet and program it?If possible, please let me know the formula that I can use TCO Thank you

Hello,    I am working on a custom board based on STM32F407 CPU with the LSM6DSV16XTR accelerometer+gyroscope and LIS2MDLTR magnetometer.The aim is to get board inclination and compass information with the maximum accuracy.The MotionFX library seems to work fine but the data from magnetometer are not correct since it requires calibration. On CubeMX even if I select the MotionMC software package, the MotionMC library is not added to the output.  I supposed this happen because the Magnetometer calibration functions are already available on the MotionFX library. The MX_MEMS_Process() is called from a FreeRTOS task, SensorReadRequest is set to 1 from a timer as done in the application demo. The problem is that MotionFX_MagCal_getParams(&mag_cal_data_out);does not update the mag_cal_data_out variable. All the hi_bi

Dear forum,We plan to assemble five LPS22DF sensors on the same I3C bus and (off course) to address them individually. According to the datasheet (7.4.1 MIPI I3CSM CCC supported commands) the answer to GETPID can only have two values, depending on SDO state. Beside that answer to GETBCR and GETDCR are fixed. As far as we understand Dynamic Address Assignment of the I3C protocol, there is an issue.We cannot use the I2C static address, because again it can have only two values, depending on SDO state.We cannot use Enter Test Mode (ENTTM) command, because it is not supported by the LPS22DF. (Even if we could, we might have 5 different random values in the Provisioned IDs, but it would not help so much to identify each sensor.)Thank you in advance

hi!I have a problem with the calibration of the IIS2ICLX-Accelerometer. I see there is happening a calibration but because I make a sensor fusion in the next step, I want to have the calibration fixed once its done. In the data-sheet and the forum I couldnt find a solution. Is there a possibility to manually control the background calibration process?Thanks!

I am working on a project using the LPS28DFW pressure sensor to monitor depth in water (~1000mBar / 10m), the sensor has a max operating range of up to 4096 mBar limited by its 24 bit 2s compliment output. As far as I can tell the negative half of the output range is never used unless autozero functionality is used, so I was hoping to use this to offset the pressure by a large value (ideally -4096) to better utilise the output range of the 2s compliment 24 bit output.  I have ran into 2 issues with this; first the REF_P (L and H) registers are read only, so I can't set a fixed offset, second the differential output is only put into the normal output registers and not into the FIFO. Is there any way around this? I see there is also a one time programable offset although ideally we would have control over when to offset for the extra range so the autozero function would be better.

Hello,We want to use LSM6DSV IMU sensor with external 3-axis 18-bit high-precision magnetometer sensor through IMU sensor hub function also with its FIFO function. We refer to following LSM6DSV sensor hub example:https://github.com/STMicroelectronics/STMems_Standard_C_drivers/blob/master/lsm6dsv_STdC/examples/lsm6dsv_sensor_hub.cWe found that we can't get 3-axis 18-bit (3 x 3 = 9 bytes) mag sensor data from above example. It seemed that we can only get 3-axis 16-bit (3 x 2 = 6 bytes) through sensor hub with FIFO function. Because our total sensor data size has 9 bytes more than 6 bytes, we don't know how to use sensor hub to get 9 bytes mag sensor data with LSM6DSV. Could you give us some suggestion to acquire 9 bytes data from sensor hub or workaround method? Thank you

Hi,I'm designing a flight controller pcb using several ST mems sensors, (LSM6DSO, LIS3MDL, LPS22HH) and interfacing them to an STM32H743IIT6 microcontroller over a shared SPI bus. I'm on the second iteration of this board, all sensors worked on the first version except the LIS3MDL, this was consistent across multiple boards, which leads me to believe its either a hardware design or a software issue in how I was trying to communicate with it. The behavior I experience with the LIS3MDL is no response at all when any communication is attempted, this is confirmed by scoping the SPI lines.Below are screenshots from my oscilloscope of an SPI read of register 0x0F. Excuse the poor aliasing on the SCK lines, this was right at the limits of my cheap scope lol. Its not shown here but I also confirmed the CS line goes low before the clock starts and return to high after the read is complete.

Hello ST guys:We want to use LSM6DSV(IMU) with an external magnetometer(LIS2MDL) through Sensor Hub mode to gather data and place acc, gyr and mag readings in the FIFO. We want to make INT2 into an input signal which triggers the sensor hub. And we configure following registers to make START _CONFIG = 1 according to following documentation: sh_cfg_read.slv_add = (LIS2MDL_I2C_ADD & 0xFEU) >> 1; /* 7bit I2C address */ sh_cfg_read.slv_subadd = LIS2MDL_OUTX_L_REG; sh_cfg_read.slv_len = 6; lsm6dsv_sh_slv_cfg_read(&dev_ctx, 0, &sh_cfg_read)

Hello ST Community,Test EnvironmentHW:STEVAL-MKI227KASTEVAL-MKI109V3 SW:Unico-GUI or MEMS-Studio Phenomenon

I am using a LIS2DE12 to detect when a mechanical shock occurs to my device. The datasheet provides only nominal values for the sample rate and the sensitivity.What is the tolerance on the sample frequency, I am using the ODR nominal rate of 5376 Hz?What is the tolerance on the sensitive, I am using the +/-16 g range, which has a typical sensitivity of 187.5 mg/digit.What is the worst case temperature sensitivity?What is the worst case offset?If worst case values are not available is there some characterization data which could provide some guidance?Thank you.

Hi All, From ST website we unable find on the datasheet or more info on LSM6DSLUSTR this part and what is the current status of this part. Besides that, whether both of the part able replace each other.

Hi All I am running Unicon GUI on Windows 10 and using it to communicate with STHS34PF80 sensor plugged onto STEVAL-MKI109V dev board.I can fully interact with the senor via the GUI app. Meaning reading and writing all registers of interest.I however want see what registers are being written and read by the GUI so I attached my logic analyzer to the I2C pins but unable to read any sensible data. So I am assuming that the motherboard is communicating over SPI. I want to be able to sniff the interaction between the GUI app and the  senor over I2C.So what do I have to do to make the GUI change communication protocol.Ola

Hi, I am currently evaluating the LSM6DSV16X IMU and I have a few questions regarding the sensor fusion functionalities, particularly the gyroscope bias.It is not very clear to me when to enable the gyroscope bias, and how to use the data it provides.Here is my reasoning: at rest, the gyroscope data has a little offset from 0. In order to compensate from that offset, one can average the data received, and subtract that from the gyroscope data.Am I correct in thinking that the gyroscope bias generated by the SFLP block is an estimation of this offset ?I intend to record IMU data of a moving object. When the device is moving, does enabling the gyroscope bias functionality automatically correct the output gyroscope data ? Or should I first enable gyroscope bias at rest, save the value it gives me, and then subtract the offset from gyroscope data ?I believe the latter answer to be correct, could you confirm this please ?I am basing my code off of th

Hi all,we have a customer hardware with ISM330IS sensor on .We want to implement a pedometer on it like in the ISM330DHCX. Is it possible ?I followed the repo ism330is-pid/ism330is_reg.c at 413acfb51fd6f81a7449980a1caf286e37c909fe · STMicroelectronics/ism330is-pid (github.com) but it seems to not include it !Instead I see it for ISM330DHCX (STMicroelectronics/ism330dhcx-pid: ism303dhcx platform independent driver based on Standard C language and compliant with MISRA standard (github.com))  Can I use that code too ? 

Hello,The LIS2DUX12 datasheet clearly states that "12.1.3 FSM_START_ADD_L (58h) and FSM_START_ADD_H (59h) FSM start address register (R/W). First available address is 0x35C.".The default address in MEMS-studio is 0x0200. In the FSM examples on GitHub, the address also is the same: 0x0200.Why are these addresses different and which one is correct?

Hi,I created a new flow with the mobile app (Expert Mode) using the acceleration sensor and the RTC time.I adjusted the sample frequency of the acceleration sensor to 960Hz and choosed writing to SD card for output.I used a user defines output file name to verify if the new file is generated correct.After starting the logging with the app I had a look on the SD card.The file was generated properly and also the acceleration data was saved properly, but the sample frequency is always 50Hz (20ms). Anyway what sample frequency I set, the output data is always 50Hz.What do I need to adapt that the correct sample frequency is used?Thank you for help.Christian

I realize there's no footprint-compatible replacement for the LSM330DLC. Is the ISM330DLC a functional replacement? Will our software team have to make changes to our software if we swap from the LSM330DLC to the ISM330DLC? Is there a migration guide for making this change? Is there a different STMicro part that would be a more suitable replacement for the LSM330DLC?

Hi,I'm working on a driver for the ST ILPS28QSW.When I switch from Pressure to Qvar mode (using the CTRL_REG3.AH_QVAR_EN bit), I notice a brief transition problem: The first 3 samples read are invalid.    Please note that those incorrect samples are measured only when the STATUS.P_DA bit is set (therefore new pressure data should be generated).Is there a way to know whe

I have been trying to log data using LSMDSOX + Profimems, but it does not record any data. I have attached pictures of this. I have not updated the firmware, and I am using Ubuntu 22.04. I have exported the data to a file, and it is empty.  

I am using MEMS studio version 1.3. If I use an arbitrary timer when debugging the FSM, then according to the specification the timer value should change with each new incoming data sample. But will the value change after three or four samples? Does anyone have a similar problem? Is the error only during debugging?

Hello,     after experimenting MEMS sensors with the IKS4A1, I developed a custom board based on STM32F407 CPU with the LSM6DSV16XTR accelerometer+gyroscope and LIS2MDLTR magnetometer.The aim is to get board inclination and compass information with the maximum accuracy.On CUBE-MX I select the LIS2MDLTR on the "Board Part Mag" section, the LSM6DSV16XTR on the "AccGyrQVar" section. Then I selected the MotionFX, MotionTL, and MotionEC libraries. Finally I choose "CUSTOM_TiltSensing" on "MEMS1_Application". At this point the yellow warning message imposed me to add the "MOTION_SENSOR" on the "Board Support Custom" section. It is not clear to me the reason of this and what does it means. In any case after I had configured all the parameters the connection with MEMS Studio produces a "Board not identified" message. Does it has

1.Shall ASM330LHB able to detect the sensor(Gyro and Accelerometer) fault in dynamic(not stationary) conditions and send the information to Host?. Self test will give fault information but it is applicable only in stationary conditions.2.If above is not possible ,can you please suggest any algorithms to detect the faults in IMU?

Hi All,I’ve been reading up on accelerometer calibration and noticed that many application notes suggest using an ODR (Output Data Rate) of 100 Hz, gathering 5-10 seconds of raw data, and then applying a least squares method for calibration. However, the IIS3DWB operates at 26 kHz, and storing raw data at this speed quickly depletes the available RAM.Is the method mentioned above also the preferred approach for calibrating the IIS3DWB, given its higher sampling rate? Could you recommend alternative techniques or tips for handling calibration with limited RAM?Any insights would be greatly appreciated.Best regards,Bob

Hi Team,We are working on BHI160B BOSCH IMU sensor using STM32L053R8 board. We can read the Quaternions, calculate RPY. But now we are finding a Yaw drift issue with that IMU. Could an expert from ST help us solve this technical issue?Also, Is there any IMU sensor related to ST that is better than BHI160B and doesn't get the Yaw drift issue? If yes, we will be happy to go with ST IMU sensors that is better than BHI160B.Any technical expert help would be very thankful.Regards,Vamshi.