Hi Guys .
I found this video. and I was wondering isnt there any use in adding hall sensors to the motors on your copter to give feedback to the controller of how fast each motor is spinning ? this would make fualt finding on underperforming motors very easy , and be good feedback source into the system for automatic corrections and even thrust calculations.
Replies
@Johann,
Spektrum has an electronic speed sensor for brushless motors and I have one. Works great. A Hall sensor is kind of yesterday.
Just a thought.
Regards,
TCIII
A hall effect sensor placed in the way shown should give you pulse resolution equal to the number of magnets in the shell and will give pulses like a quadrature encoder (if you use a latching hall sensor). Using a pair of sensors placed half phase apart will be able to pick up both speed and direction in exactly the same way a quadrature encoder would. This little trick will work nicely on an outrunner because the magnets are on the outer shell - I don't believe this would work on an inrunner - I have done it on inrunner type motors by physically placing magnets onto the shaft and then picking up from the spinning magnets. Its easiest to use latching hall sensors for this need - calculations to determine curve tops can be too intensive to use a non-latching sensor.
@Dusty - I think you're thinking of the more traditional use of a hall sensor directly on the wire that allows it to pick up the current - the wire current can be used to determine the load and roughly from there to determine speed but in the way demonstrated the pulse is being picked up directly off of the spinning magnets so it should absolutely be giving you information from which you can calculate the speed - same as any physical shaft encoder.
Doesn't really tell you how fast it is spinning. Tells you how much current it is drawing, within a tolerance. This tells you how hard the motor is working to spin, not neccessarily how fast the spin is. Can still be useful, but for larger differential rather than small ones. My experience with hall effect is that they are a bit fickle and can be difficult to calibrate to give numbers you can trust. then what do you conclude with the numbers?
I used to work with high powered communications that used DC voltage and some high currents. We used to buy high power .01 ohm resistors that this guy built out of his garage. Then we would just stick them in series with the line voltage and measure the Vdrop. Very accurate. I think they can be hard to find but it has been many years since I did that, so I am not sure.
In theory you don't need the hall sensor, a normal esc kind of senses position already, so that it knows how to power the motor. In theory this can be used to determine the current RPM if you know the number of poles in the motor. The commercial RC ESC on the market today don't do this though, because there is no need on something like a plane.
In reality this would mean using custom ESC firmware to send back the RPM signal to the flight control using some sort of communications channel. That is technically achievable, but on the forums where I have seen this discussed (a number of times) it usually turns into a discussion about what the flight controller should do with this info.
Certainly from my point of view, detecting whether a motor is not turning (because it is broken, damaged, gone, trying to dig a propeller blade into someones leg etc) could be useful so that the controller can cut power to that motor and avoid further damage etc