Hi,
I have GY-91 board and I am not sure if it is fake or not. I hear some boards has IMU but no Baro.
I receive this error on mission planner :
Config Error: Baro: unable to initialise driver
is there anyway to skip the error and continue to work with existing IMU ?
I even can’t see the the gyro and acc measurements on mission planners when I enable the tunning option
Easy enough to know. Look for a shiny can with a hole in it…
I think this metal can shall eb the Bar.
I could see that on my board
is there anyway to see the gyro and acc measurements on the mission planner and ingore the baro error?
How you did you connect pin by pin the GY91 board to mission planner?
If you have this nice breadboard it seems you are not only playing with ardupilot.
If you also play with arduino you can find simple sketches to test the board
My connection is based on OBAL board.
3V3: 3.3v from PI 4
GND: 0V from PI
SCL: I2C Clock on PI
SDA: I2C Data on PI
If you are working with PI you also will find some GY-91 projects just to check if the board is working correct.
Quick google search shows: GitHub - muzammilpeer/GY-91-PI4J: GY-91 10DOF,pi4j raspberry pi
Thx, I see one address on I2C. I believe this is for the MPU9250.
I wonder why the Baro doesn’t appear now.
this is a fake board or defect Baro ? 
Both is possible, but if it is a fake board than probably the IMU is not a MPU9250 but a 6050 without magnetometer. It has the same address on I2C
to confirm that the MPU is working, I used a python script for that and I can see results.
I would like I display the gyro and acc data from the existing MPU on mission planner, would that be possible?
Under Setup->Basis Hardware ->HW ID you can find the recognized devices
On main screen you can open the status monitor with lots of values including acceleration ax; ay; az and gyro gx; gy; gz. Also you can see the artificial horizon moving
I looked under the “HW ID”. I was surpised that no device ID exists.
I also checked the MAVlink frame and it seems no Gyro or Acc data is sent !
is the SW blocking reading other devices/sensors when the Baro has errors ?
I can’t answer this but I don’t think so. On my ESP32 FC I also have no Baro and I get the data as shown. But I have defined with the board definition that this FC has no baro.
I tested the code in the github you sent me
seems there is no Magno or Baro !!!
it is a fake GY-91 
They sell me MPU6050 with price of GY-91.
Yeah, i got boned like that, none of these GY-91 boards are real i guess, mine had no magnetometer, an MPU6500, but it does have a barometer, at least that i can detect with arduino programs. I’m not sure if a flight controller would recognize it.
I got a hint that if the led on the board is red , then it is fake
if it is green then, high probability it will be genuine.
I started to believe that I shall buy all components discrete and stack them myself to avoid fake purchases …
Why should the fake manufacturer always use red leds, I don’t think this is a clear method to identify fake or not.
From my point is more important where you buy your stuff, from some unknown dealer on aliexpress, ebay etc or from serious local dealers with little higher price and the possibility to return defect modules
Not that on-board magnetometers are useful anyway but if you wanted one buy a board with an MPU9250 or another 9-axis IMU.
The MPU6050 and MPU9250 have been obsolete for so long that it’s hard to believe anyone is still selling a breakout board containing genuine, NOS ICs. I think it’s reasonably safe to assume that anything labeled this way in 2024 is highly likely to be a cheap clone, and as such, unreliable.
There are plenty of supported sensors at cheap price points that aren’t EOL. Suggest sourcing a different part number entirely rather than spinning your wheels on what are likely to be junk components.
The good thing about MPU6050 and MPU9250 is the strong libraries on internet and the support you can get. That makes you confident that it will work with !
The GY-91 is 10 DoF and it is not easy to find such functioanlity on small boards.
I agree with your arguments.
The device is not manufactured anymore and I started to believe that the sensors in the market may be scrapped from old devices or even fake as I experienced.
I am not expert in sensor world so I would ask your experience about what shall be used instead ?
I only know about BNO055/BNO085 (9 DoF) from youtube but no concrete information.
I found GY-912 (bmp388 + icm20948) as an alternative solution but icm20948 is not recommended for new designs as stated on invensense websiste
The chip i got was literally a fake, it had an IC that had MPU-9250 printed on it but every possible I2C scans and probes all confirmed it was an MPU-6500. I would avoid buying those chips completely, and in any case it’s better to have a seperate magnetometer anyway.
