Accuracy on inclinometer MyRio 1900

[MarcusNilsen]

Long story short, I want to use a MyRio to measure my inclination with the accelerometer.

But I would like to know what the accuracy of the angle I find is.

I have no clue if it's +/- 1 degree og if it's +/- 10...

Can someone help me solve this?

Also I hope that I have posted this to the correct board.

Best regards 

Marcus

[Bob_Schor]

What a great project for an Engineer!

First, calibrate your Accelerometer.  [It is surprising, and shocking to me, to work with some BME students in their senior year, and even some students in the MS Program, who wanted to measure "muscle twitch" movement with an inexpensive triaxial accelerometer that provided X, Y, and Z voltages having a stated Bias and Gain for each channel with an uncertainty of ±10%, and the thought of calibrating their device to determine the true Bias and Gain for each channel never crossed their mind ...].

Now that you have your Accelerometer calibrated so that you know, say, that ±1 V corresponds to ±1 g (9.8 m/sec²) (plug your actual calibration factor in here), you are ready to answer your own question.  Do the following:

• Assume that with the myRIO laying "flat" on a "level surface" (use a bubble level to test this, if you want), call the "upward" direction +Z, the direction pointing to the top (where it says "National Instruments, NI myRIO") the +X direction, and the side where the MXP A and B sockets are the +Y direction.
• Measure the Accelerations in this orientation.  It should be (0, 0, 1), i.e. no acceleration along X and Y, and 1 g along Z (we can argue about signs later).  If you get something wildly different, you forgot to Calibrate.
• Put a coin under +X edge.  Measure how high it has been raised.  Also measure the long dimension of the myRIO.  Compute the ratio Inclination = (coin height)/(myRIO length).  Measure Accelerations, which should be (+small number, 0, 1-smaller number).  I'm making this statement "off the top of my head", so there could be an error in here, but explain why the X value of acceleration * 2 * pi / 360) is very close to the Inclination ratio.  What is the "exact" expression?
• Put other items under the top edge of the myRIO to incline it at other angles.  Measure the angle (which comes down to measuring the height of the top edge and doing a little math).  Calculate the corresponding components of the Acceleration Vector (as seen by myRIO).  Compare, come up with a formula that lets you get one from the other.

Bob Schor

[Henrik_Volkers]

Bobs coin method elaborated is the sine bar

If you find a mech-shop that has one and a set of gauge blocks you can do calibration 🙂

Math involves some matrix trigonometry ..

and if you look at your results closely, you migth find that the axis of your MEMS sensor are not perfectly all  90°...

(usually x,y are but the z-axis (not in the plane with the ligth/etching process) is a little off 🙂

(Do some more measurements and apply a linear fit)

I expect that the limit in resolution is found by looking at the drifts ...  (and applying a low low pass filter 😄 )

(try a one day monitoring, and place it on a sunny (temperature.. sensor onboard ?) place.

Another good question: How to orientate the 3D accelerometer to get the best resolution?