Magnetometer to measure high angular velocity in small object

Question

I have to measure the frequency of a little circle in rotation. You can image this circle flying in air, because this circle can't touch anything that is not in rotation. So I can't use some simple trick to count the number of complete rotation in an amount of time.

So I supposed my only chance was to use an accelerometer, a gyroscope, or a magnetometer. The accelerometer can detect the centripetal acceleration, and the gyroscope and the magnetometer with some calculus directly the frequency.

The problem is the high frequency of this circle (can reach up to 50 Hz). Doing some simple calculus we know we need a gyroscope that can measure big angular velocity: 50*360°/s = 18.000°/s. Also the accelerometer need big range of values (the radius of the circle is only 5 cm): w (angular velocity) = (2 * 3,14) / (1/50) = 314 rad/sec acc_centr = w^2 * R = 98596 * 0,05 ~ 5000m/s^2 ~ 500g

Now I have seen there are some accelerometer or gyroscope for industrial purposes with enough range, but my question is:

How I can understand if a magnetometer can used is this kind of application? Considering there is no magnetic field near the circle, can a magnetometer be used to measure quickly change in inclination? In the datasheet i can read how ofter the sensor can communicate with my arduino, but nothing about how quick the rotation can be.

Is the reason that a magnetometer don't have the limits of a gyroscope or an accelerometer?

Answer 1

Do you need the sensor to be installed on the circle itself? If so, can you move the accelerometer closer to the center so that less force is applied to it?

You might want to look into optical RPM sensors, like this. All you need to do is put a reflective tape on your rotary part so that the sensor can see the reflection once for each cycle.

Answer 2

Im not sure what kind of system you're trying to build but I think an encoder type sensor would work best. You could mount some amount of red leds on the edge of the disk and place a light sensor around the disk. This would let you measure the frequency of the disk without dealing with magnetometers.