Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Question about a Hall effect sensor

  1. Oct 24, 2016 #1
    Been thinking to make a quadcopter, not a huge one, a fairly small and light one to start out with. The problem is I would like it to read the RPM of each engine separately.
    There are several ways I have came up with:
    - One of them is to use the current reading and calibrate it, however using this would result in an estimate and not a continuous update.
    - The second is to use a photo transistor sensor, which honestly would be way easier, but with uncertainty of the movement of the blades it may damage the blades.
    -The third is using a hall effect.
    With the hall effect I know that there are two magnetic sources from which the hall effect might detect. One of them comes straight from the static magnets on the brushed DC motors, and the other the coil, as it is designed to just spin around and wired in such a way that it always has an opposite polarity than the magnets.
    I would need to filter out the source of the magnets using a capacitor, as for all intents and purposes it is a constant voltage source, and the coils move in such a way I would get an oscillation exactly in the frequency of the rotation. The question is, would it be a valid solution to use a hall effect to detect the rotation, and where would be best to place the sensor in relation to the motors?
    Again the motors are brushed, with static magnets and moving coils.
     
    Last edited: Oct 24, 2016
  2. jcsd
  3. Oct 24, 2016 #2
    The field won't be rotating like it would in an AC motor. The commutator keeps the field going in same direction by flipping the current in the coils. The field may "twitch" a bit as the motor spins and the current flips.

    Maybe if you position your sensor where the stator poles split you could pick up a magnetic "twitch" as the armature rolls by. If you have a 2-pole armature then you would get 2 "twitches" per rotation. If it's a 4-pole then you would get 4 "twitches" per rotation.

    I Googled a bit to see if anyone has used the method I recommended. I had no luck so I can't promise it will work. Everyone is making tachs by putting a magnet on the shaft as a target for the sensor.

    About the photo sensor. Another problem you would encounter is that bright sunlight may saturate your sensor and prevent it from working outdoors. You would need to consider the dynamic range of your setup.
     
  4. Oct 28, 2016 #3
    The most common way to measure rpm on the three phase PM motors used for models is to monitor phase voltage. You detect the crossing of voltage from positive to negative which indicates commutation frequency, commonly called electrical rpm. Once known it has to be scaled to mechanical rpm by dividing it with the number of motor pole pairs.

    The problem with using Hall Effect sensing is the imbalance it can cause. Props on multi-copters can spin at quite high rpm. The magnet installed on the motor shaft has to be well balanced and tolerant to high centrifugal forces. It's possible to do it, but it might be a bit hard to find a suitable magnet setup. Model helicopters often use Hall Effect sensing to monitor head speed, but that's at a much lower rpm typically about a tenth of motor rpm. In any case you might be able to adapt a head speed sensor for a model helicopter to your application.

    Optical sensing has it's own set of issues. Typically you need a laser and reflective surface which again can get into balance issues with props. It would be hard to get something like a sticker to stay on there. It would pretty much have to be painted or glued which can cause trouble with prop balance. As already mentioned I don't think using a photo-sensor on its own would be reliable.
     
  5. Oct 30, 2016 #4

    NascentOxygen

    User Avatar

    Staff: Mentor

    A microphone mounted under each rotor and angled upwards would, I'm thinking, pick up the sound of blade rotation, probably one pulse each time a blade passed overhead. Process this signal and include a ÷2 and a frequency counter could then give you a direct readout of rotor revs/sec.

    Easier said than done, though, I'm sure.
     
  6. Oct 30, 2016 #5
    Is there space to mount a ferrous gear on the shaft, say under the blades? This could act as a 'tone ring' similar to automotive ABS speed sensors. This wouldn't throw out the balance and allow a Hall effect sensor to give a square-wave frequency output.
     
  7. Oct 30, 2016 #6

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2016 Award

    Best determined by experiment.
    In a well designed electric moachine the magnetic flux stays inside. I dont know about hobby motors, take a look at one with a search coil.

    You might do as well with a small electret microphone near the blade tips. If the propellers are ducted that'd be handy place to glue them.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted