5PCS GY-271 QMC5883L Triple Axis Compass Magnetometer Sensor Module 3.3V 5V for Arduino and Raspberry Pi (5PCS)

At first I was accidentally using the HMC5883L library. That was wrong, this is the QMC5883L. You've probably made this error if the heading hangs at a constant value.I then went and downloaded the QMC5883L library from DFRobot. I found it odd that there was no calibration setting in the example code (you're usually supposed to move the magnetometer around in all orientations to get the x, y, z extrema values, and then input those into the code).It took me a full day to finally realize that the DFRobot library conducts ongoing calibration as you use it. So, if you are going to use this library, the first thing you must do every time you turn on your Arduino is rotate your magnetometer in all different directions. Then you're good to go.Ultimately, I landed on a much more advanced library by mprograms on GitHub (you can find this easily on Google). This library has example code to get your calibration values, and it actually lets you input your calibrations permanently so that you don't need to do it every time you start your Arduino.When you get a good calibration on these things, the data is very clean. The magnetometer seems completed unaffected by any rogue magnetic fields.As far as noise, I'm getting a 6.52 degree standard deviation at a 295 degree heading. That suggests 68% of your values lie within +- 2% of the actual value, and 95% of values within +- 4% (assuming gaussian distribution). Add a little bit of a lowpass to your data and you're golden.

First off, the I2C address for these items show up as 0x0d. A REAL HCM5883L is at 0x1E. I have code running that works with a BN-880 GPS module which contains a REAL HCM5883L. Same code does NOT work with these boards. EVEN when I change the address used by the Adafruit_HCM5883_U library (to match these boards)... I get NO data. I am able to get the Sensor details. That all reads back fine, but I NEVER get any data from the X,Y or Z registers. If someone wants to explain WHY this is the case, I'd be glad to listen. Otherwise, don't bother trying to use these boards.

I like these, but they have the highest DOA ratio of ANYTHING I have EVER bought.

I have had no problems getting this running. I have not yet tested the accuracy of the magnetic measurement, but it seems valid. My particular chip appeared to follow the datasheet from OSOYOO for the qmc5883l. The sensor address is 0x0D, and the data registers start at register 0x00 and go through 0x05.

The chips are labeled as QMC5883L; however, the default address is 0x1E instead of 0x8D (the QMC default), and the registers match the mapping for the HMC (output registers 0x03-0x08), not the QMC(output registers 0x00-0x05). If you get these, use an HMC5883L library or remap the registers as needed. Other than that, though, they work fine.

Work well with custom SlimeVR firmware with MPU6050. Doesn't work with standard SlimeVR firmware currently, must use a forked version.

Works great, very small.

3 of the units in a row all output the same two headings. They constantly output ~42.5 degrees and then occassionally will jump to ~357.4 regardless of orientation. I find it incredibly suspicious that all of the modules that I assembled and tested output this exact same erroneous data. I have not even attempted to get the other 2 modules to work as I'm pretty sure I'd just be wasting my time. Spend a few extra bucks and finding another module.