Small Brushless Motors - can they generate?

The videos..

I used AXI brushless motors for the tests - all the same size, 36mdiameter x 40mm long.

The slave motor was always a 990KV motor

I used a 900KV motor as driver in the 1st video 'Motor-1'

This did not work well, very jerky and intermittent.

I then used a 1200KV AXI motor as the drive motor - this was much better, but still not reliable startup. Motor-2.

Perhaps using a very small motor as the slave, say a 15mm or 20mm diameter motor, of around 800KV, and a larger ( similar to the ones I used) motor, of around 1200 to 2000KV might work? Bottom line is..Maybe, with a lot of fiddling and mucking about - Use two plain, small dc motors, of the old ferrite magnet variety - that should always work!

Joe

EDIT - just to say how I drove the drive motor:

The drive motor was driven in my lathe which has a tacho readout - I varied the spindle speed from 0 to 800 RPM.

Motor-1 Video -  jerked all the way up to 500 RPM and then started to try turning.

Motor-2 Video - jerky up to approx 350RPM and then work turn well up to 450RPM or so, and then jerk again, and then smooth out at 700 RPM, with a few occasional jerks...

Edited By Joseph Noci 1 on 05/09/2018 14:56:50

To answer the question in the header - yes in general if you force the motor to spin faster than the no-load speed, the back EMF will exceed the supply voltage and the motor will generate. This applies to all of the above, including the various brushless motors - as well as"normal" AC induction motors.

There's a lot of ambiguity in motor descriptions but what is usually described as "brushless DC" is driven by 3 trapezoidal ie flat topped phase voltages ("6 step" control). If you drive the motor and observe the waveforms, they will look sort of flat topped. In contrast, an "AC brushless" will generate something approximating to a sinusoid. Both are synchronous machines with permanent magnet rotors and need to be commutated by some form of electronics.

You can commutate brushless motors without absolute position indicators (such as hall sensors, resolvers etc), based on the back EMF but the challenge comes when you wish to start the thing moving. Without movement you have no back EMF, so can't determine where the rotor is. In some applications you can just ramp the speed up and let it follow open loop but for loads with high starting torque or where positioning is critical this doesn't work so well.

Joseph - if you look at the phase voltages between your motors, you'd probably understand why they might have difficulty synchronising. I think you'll find they are a long way from either trapezoidal or sinusoidal.

Werner suggested using a PM motor as a generator but the voltage will vary proportional to speed. If you are driving with a variable speed (eg engine), this may be a problem! With a claw pole alternator ("Lundell" machine) that has a wound rotor, you can change the voltage constant by varying the rotor current, otherwise it behaves in a very similar fashion. That's what the alternator regulator does of course.

Murray

Posted by Joseph Noci 1 on 05/09/2018 14:53:43:

The videos..

I used AXI brushless motors for the tests - all the same size, 36mdiameter x 40mm long.

The slave motor was always a 990KV motor

I used a 900KV motor as driver in the 1st video 'Motor-1'

This did not work well, very jerky and intermittent.

I then used a 1200KV AXI motor as the drive motor - this was much better, but still not reliable startup. Motor-2.

Perhaps using a very small motor as the slave, say a 15mm or 20mm diameter motor, of around 800KV, and a larger ( similar to the ones I used) motor, of around 1200 to 2000KV might work? Bottom line is..Maybe, with a lot of fiddling and mucking about - Use two plain, small dc motors, of the old ferrite magnet variety - that should always work!

Joe

EDIT - just to say how I drove the drive motor:

The drive motor was driven in my lathe which has a tacho readout - I varied the spindle speed from 0 to 800 RPM.

Motor-1 Video - jerked all the way up to 500 RPM and then started to try turning.

Motor-2 Video - jerky up to approx 350RPM and then work turn well up to 450RPM or so, and then jerk again, and then smooth out at 700 RPM, with a few occasional jerks...

Edited By Joseph Noci 1 on 05/09/2018 14:56:50

This speed is too low for reliable operation of a drone type Brushless DC motors. They will not generate enough voltage. Using lower kVP motors will help.

Robert

Edited By Robert Atkinson 2 on 05/09/2018 19:18:24

But I suppose the original question was what motor could be used to generate a voltage to drive a voltmeter as a RPM indicator - A brushless motor will work fine - just choose one with a lowish KV, say 350 or so, add 6 diodes - 1N4001, maybe cost a pound or so, and you will have a nifty RPM indicator. Use a potentiometer to scale the output to suit a low-volt DC voltmeter.

Joe

I think you will find a major issue with any setup that uses a diode rectifier. The diode drop will be about 0.4V so no signal at low speeds (unless you use schottky diodes, and even then it will be enough to result in significant non-linearity).

I would suggest using the unrectified output across two of the motor connections into a moving iron meter. A variable resistor can be used to calibrate the meter and it will give a linear response from 0 to whatever you want.

Neil

If you want to use it as tachometer with an analog or digital read out it is good to use a magnet on a shaft with a reed contact or hall sensor. Then with a NE 555 circuit as triggered monoflop you can convert pulse to DC voltage.

This is a small and cheap electronic circuit . I you want i can provide the plan.

Werner

Not relevant in this case Neil - That's why I said use a low KV motor - it generates a higher voltage at low speeds - and the measured engine surely will not run well below 200rpm or so..A 300KV motor will easily register on the voltmeter at 300rpm through the diodes..

Joe