Why can't you run 2 brushless motors off 1 controller?

[SHARKBITEATTACK]

I just can't seem to wrap my head around this concept. What would happen if you took 2 hub motors and hooked the phases up to each other and then spun one of the wheels? Would both wheels spin at synchronous speeds? 1 acting as a gen and the other as a motor

 

[d8veh]

I just can't seem to wrap my head around this concept. What would happen if you took 2 hub motors and hooked the phases up to each other and then spun one of the wheels? Would both wheels spin at synchronous speeds? 1 acting as a gen and the other as a motor

The pulse timing from the controller has to be matched to the exact motor position. It could only work of the motors were physically constrained to be in exactly the same rotational position all the time, which is virtually impossible. It could work on a chain-drive system id the two motors were connected together by a toothed belt or gears, but you'd have to find some way of getting them into synchronized positions in the first place.

you can use one controller to drive two brushed motors because there's no pulse timing.

It's no problem to use two controllers to drive two motors. You can still have a single battery,throttle, etc. What would be the advantage of a single controller?

BMSBattery do twin controllers for 2WD, but I can't see any advantage over two independent ones:
http://www.bmsbattery.com/controller/607-s06-250w-imitation-torque-square-wave-controller.html

 

[Xrain]

Brushless motors rely on accurate rotation signals coming from the motor. Since the motors are not physically connected they will fall in and out of phase with each other as they rotate, since each wheel naturally travels a slightly different distance every time you go over bumps, turn, or drag from tires.

Since the controller is sending the same signal to both wheels the same poles on each motor are energized. However since the motors are in different positions its likely that while the pole in one motor will be driven in the correct direction, the other wheel will be driven in the incorrect direction.

So in essence, the wheel that is out of phase might drive forwards, backwards, or not at all and will change between these many times a second.

To directly answer your question if you hooked the phases up and just had sensors plugged into one motor and spun that motor, if you are lucky and both happen to be in the right position both will intitally start rotating. However almost immediately the second wheel will start juttering and stop, and make random jerking motions if anything. If the other wheel starts out not in phase it will just either jerk and vibrate randomly or nothing at all.

I can see where your line of reasoning is going now, "well why dont we just hook the sensors together!" This will not work either. If you are working with hall effect sensors they output a voltage (generally 0 to 5V), it outputs voltage spikes, each spike is one pole of the motor passing the sensor, this allows the controller to get rotation information so it can properly drive each phase of the motor. Hooking the sensors together will just make anywhere from the same amount to double the number of spikes that is outputted to the controller. This of-course is incorrect, and bother motors will not work correctly.


It is very possible to make a controller that can drive two, three or a thousand motors. But basically all it will be is two motor controllers buttoned up into one package. There are of course good things about this as there is lots of redundant circuitry between them that can have a single common instead of two sets, and you would better be able to control each motor to work together (E-Differential anyone?) vs. using two separate controllers with the throttle signals tied together.

Edit: It appears I have been ninja'd, that controller that he linked as far as i can tell is just two separate motor controllers with the throttle signal tied together, and a convenient carrying case.

 

[spinningmagnets]

With brushed controllers, the controller doesn't need to know what the exact position is of the rotor magnets in relation to the stator electro-magnets...in order to know when to turn the electro-magnets on and off. At 300-RPM, the timing is very important (also important at 1-RPM). That being said, I'm not entirely sure one brushed controller could manage two motors.

Regardless of the magic black-box voodoo, all you really need to know is that...it's been tried a variety of ways by electronics experts (which I am definitely NOT). There are several two-controller builds around here using one throttle.

Unless you are certain you have figured out a new way to skin that particular cat, and you can afford to fry a few components to prove your theory...I'd recommend copying stuff that is proven to work.

 

[Xrain]

With brushed controllers, the controller doesn't need to know what the exact position is of the rotor magnets in relation to the stator electro-magnets...in order to know when to turn the electro-magnets on and off. At 300-RPM, the timing is very important (also important at 1-RPM). That being said, I'm not entirely sure one brushed controller could manage two motors.

Regardless of the magic black-box voodoo, all you really need to know is that...it's been tried a variety of ways by electronics experts (which I am definitely NOT). There are several two-controller builds around here using one throttle.

Unless you are certain you have figured out a new way to skin that particular cat, and you can afford to fry a few components to prove your theory...I'd recommend copying stuff that is proven to work.

I think you mean brushless, for brushless controllers, as long as the power output is sufficient, the only real difference is that the controller for the two motors would have 6 phase outputs, and 2 sensor inputs (or 0 still if you want sensorless)

If you were running brushed motors you would have no problems at all since brushed motors only really care about if there is power going to their input.

 

[spinningmagnets]

Maybe I was wrong to state it that way. Most hub-motors I've read about use hall-sensors (and a few use optical sensors) to let the controller know the exact position of the rotor in relation to the stator. A sensorless controller uses the back-EMP to try to figure out where the rotor position is at any given time, so...

I believe Crystalyte makes a dual-motor controller, but it's my understanding that it's just two separate controllers in one box. I believe the original poster (OP) was talking about using a single controller that was intended to operate one motor, and using it to run two motors (splitting off the 3 phase wires into 6).

 

[dogman dan]

Wouldn't a 6 phase 10 halls controller technically be 2 controllers? I don't mean just a y connector in the wires, but two separate timings.

I think Thud did run two brushless motors on one controller, by a chain and sprocket linkage of the two motors keeping them in synch. If I understand it right, losing synch fried one of the motors later in the day.

 

[Xrain]

Wouldn't a 6 phase 10 halls controller technically be 2 controllers? I don't mean just a y connector in the wires, but two separate timings.

I think Thud did run two brushless motors on one controller, by a chain and sprocket linkage of the two motors keeping them in synch. If I understand it right, losing synch fried one of the motors later in the day.

Well the initial power conditioning circuitry that feeds the output channels (voltage regulation, smoothing etc.) could be the same, you could also use the same microcontroller so long as it has enough i/o pins and ADC's onboard for the double the inputs and outputs. There would just be 6 phase output channels on the controller and using software you would drive the correct outputs in conjunction with the sensor inputs (or read the right phase lead) to drive the two motors.

The advantage of this is you do get to cut off some of the redundancy you would normally have with two completely separate circuits (reducing size), and the two motors would be driven by the same microcontroller so you could more easily do different control functionality like power balancing, and speed matching between the motors. Without having to make two microcontrollers talk with each other.

I bet Thud's motor burned because chains still have a bit of stretch to them, and the more poles the motor has the smaller the rotation needed to fall out of phase. Give it a bit of gas the motors fall out of phase, something probably shorts in the controller and you get the magic smoke everywhere. The only real feasible way you can run two motors on one controller is if they are both on the same drive-shaft. Only that will have low enough play to not be a risk, you still have to be careful in the initial alignment of the two motors keeping both phases as close as possible, as any misalignment will make you lose power.

Maybe I was wrong to state it that way. Most hub-motors I've read about use hall-sensors (and a few use optical sensors) to let the controller know the exact position of the rotor in relation to the stator. A sensorless controller uses the back-EMP to try to figure out where the rotor position is at any given time, so...

I believe Crystalyte makes a dual-motor controller, but it's my understanding that it's just two separate controllers in one box. I believe the original poster (OP) was talking about using a single controller that was intended to operate one motor, and using it to run two motors (splitting off the 3 phase wires into 6).

I think you started to talk about brushed controllers and switched mid-sentence to talk about brushless controllers so I was a bit confused.

The Back-EMF (EMP is what nuclear bombs make) is used in sensorless motors since only two phases are energized at any time, so the third phase is read by the controller. The motor acts a bit like a generator for the third phase and makes a voltage pulse in the unpowered phase. With a bit of extra maths this gives you the position of the motor.

Only problem with sensorless is since it requires the motor to be turning to figure out the position, they tend to start out rather rough, and if you rotate the motor backwards, cheaper controllers tend to go up in smoke.

Edit:

In short to the (OP) you cant use one controller designed to drive just one motor and parallel the phase wires coming out. Why? Because Fire.

 

[SHARKBITEATTACK]

Ok I was actually theorizing a dual motor offroad rig and would want both motors to stay locked at the same rpm. This could be useful when going through slippery or loose terrain and trying to maintain speed when 1 of the wheels loses traction.

I thought they would work in a similar fashion to synchronus generators on the electrical grid where torque would vary on individual machines but not rpm.

 

[Farfle]

Ripperton has successfully run what I believe is 4 mars motors off of two kelly controllers. Each "pair" of stators are aligned and synchronized on a common shaft, and the kelly only ties into the sensors from one stator.

 

[h0tr0d]

It's possible, this guys do it:
http://www.enstroj.si/images/stories/emrax_twin_2013.jpg
http://www.enstroj.si/Electric-products/emrax-228-motorsgen.html