Skip to content

Latest commit

 

History

History
87 lines (77 loc) · 6.31 KB

README.md

File metadata and controls

87 lines (77 loc) · 6.31 KB
Raw

MMC5983MA Compass Test - Quick-And-Dirty test

For reference see Memsic MMC5983MA Compass Datasheet Rev A, Formal release date: 4/3/2019
Also Nine! videos about this part by Robert.

With the SparkFun Qwiic MMC5983MA board I get poor results; field magnitude measurement problems highlight big offsets in each axis.
The MEMSIC MMC5983-B Prototyping Board gets good results.
Either a problem with the SparkFun board or I'm doing something wrong...

To make a reading, I use the degauss procedure to find the mid-point (the zero-field output value, inapproriately called 'offset' in MEMSIC datasheet). Reminder: The film magnetization procedure uses a SET/RESET function which gives a big pulse and resets the device, clearing any residual magnetization and setting the measurement polarity (these functions might better be called MAGNETIZE_REVERSE_POLARITY and MAGNETIZE_FORWARD_POLARITY). Per the datasheet, I'm making a RESET, measuring, a SET, measuring. Then calculate the center-value (nominal zero-field output which MEMSIC inappropriately calls offset) and measurement given two readings with opposite polarity. All per the instructions in the datasheet.

Rotating the sensor at a spot in 3D, the measured field magnitude should be constant (no hard magnetic offsets on the test boards). I rotate the sensor at a fixed point to find the minimum and maximum sensor output for each axis, which should be identical magnitude if the degauss procedure works and the mid-point is found correctly.

My test program (this GItHub repository) is a C++ 32-bit Windows program using wxWidgets, communicating with the Qwiic MMC5983A boards using an Adafruit MCP2221 USB-to-Qwiic adapter. Included is the MMC5983A C++ device driver I wrote (not yet complete).

MEMSIC MMC5983-B Prototyping Board Good Results

I wired the MEMSIC evaluation board to Qwiic and connected to my PC. Rotating about a point I see reasonably constant field magnitude (95% to 105% of expected value), and the min/max for all axis have consistent magnitude (though the Z axis has slightly higher gain): offset image

SparkFun Qwiic MMC5983MA 19921 Problems

Either I'm doing something wrong (hardly ever happens), or the SparkFun MMC5983MA is not working properly!
I'd be delighted if anyone can report success with the SparkFun MMC5983MA and show me where I've made a mistake!
The SparkFun compass seems to fail a basic sanity test (constant field magnitude reading).

Rotating the MMC5983MA at a fixed point in space, it's measuring different field magnitudes; the magnitude should be constant. Note magnitude is sqrt(X^2+Y^2+Z^2), and reported values range from 131% down to 76% of the expected local field strength for my location (from WMM World Magnetic Model). So in the right ballpark and making measurements, but definitely something is wrong. The file ExampleMagnitudeProblemLog.txt shows details of the erroneous readings. The range of magnitudes measure imply really large gain differences between axis, the auto-zero SET/RESET process does not really auto-zero, and/or some other issue?

Further testing shows that there's a big offset in each axis. The RESET/SET degauss and sense-flip procedure is supposed to take out the offset (ie auto-zero each axis), except it's not working here. There's no hard offset in the test setup that would explain this behavior... Here's an example after rotating the sensor to try find min and max for each axis: offset image

MMC5983MA datasheet problems:

The Memsic MMC5983MA Compass Datasheet Rev A, Formal release date: 4/3/2019 is seriously unclear on a number of points, and sometimes conflicts with MEMSIC's sample code. Unfortunately, I've had great difficulty getting sensible answers from MEMSIC technical support.

  1. Continuous mode vs. one-shot mode aren't well documented.
    • a) Is it necessary to write TM_M to start measurements in continuous mode?
    • b) In continuous mode, is Meas_M_Done ever set?
    • c) Are TM_M and Meas_M_Done only for one-shot measurements?
    • d) Does temperature measurement require continuous pressure mode to be turned off?
      MEMSIC: Temperature sensor does not really work, don't use it.
  2. Can SET and RESET used while continuous mode is enabled?
  3. The use and behavior of auto-SET-RESET is not documented:
    • a) What is the PURPOSE of auto-SET-RESET?
      Unless a large field saturates the sensor, why is this helpful?
      How can this improve performance?
    • b) The Auto_SR_en bit is labeled "Automatic Set-Reset".
      What exactly does the sensor do if this bit is set?
      Does it operate in both continuous and one-shot mode?
    • c) The sample code provided by MEMSIC says "continuous mode with auto set and reset" - is that correct?
      The datasheet page 16 says "The device will perform a SET automatically..."; should that be "performs a periodic RESET/SET sequence"? Which is the correct explanation of what the chip does?
      The order of SET and RESET determines the sensing polarity, so this needs to be clear, right?
  4. Explain saturation detection - what are the steps?
    What is the additional applied saturation test field in Gauss when a saturation-enable bit is set?
    Is it necessary to set the enable bits back to 0, or do they reset automatically like SET/RESET?
  5. The datasheet says SET and RESET operations take 500 nsec.
    MEMSIC sample code waits 500 usec. Which is correct?
  6. The decimation filter is mentioned but not described.
    Given BW and CM settings, What exactly is the sampling and filter?
  7. OTP read controls are discussed, but OTP is never defined. What is it? Why do we care?
    MEMSIC: OTP indicates MMC5983MA's one-time programming information was read successfully by firmware after the last reset. This bit should always be set. It should never be necessary to command the device to re-read OTP.