|
|
|
|
|
|
|
|
|
Valen
Experienced Roboteer
Joined: 07 Jul 2004
Posts: 4436
Location: Sydney
|
if the PWM is < the time constant of the motor then yeah you will get more torque with high power short pulses. That would likley be around the 3-4Hz level though.
The torque is produced by the current flow in the coil, the more electrons moving the greater the force produced. If your PWM frequency is high enough then the natural low pass filter the coil makes will act to average the current flow. Meaning you wind up with a steady flow of electrons past the magnets. IE constant force.
More volts = more amps = more force.
With the slow PWM mode (where PWM freq < motor time constant) then you don't get that averaging and you effectivley run the motor stalled. IE the motor will pull 100% power.
That was the problem nick was having with the thingaps. Their inductance is so low that the time constant was < the PWM period. Meaning the controller was trying to source 3000A (for a short period of time (ie the throttle position say 10%)) and the motor was doing the same.
By adding his inductors he increased the TC of the motor and got the motors seeing low voltage DC again.
From my understanding its mainly the ChooChoo mob who like the low frequency pulse mode of operation. The problem they have is static friction is greater than dynamic, IE if they slowly ramp up the power going to the motors it will build to the point where it breaks all the sticking force then lurches off at some decent speed. With the low freq pulses they can break the friction and use the mechanical averaging of the train to stop the lurch off the line. For them the excess heating isn't really a worry ;->
In terms of nice to motors you should run the highest PWM freq possible, you will minimize voltage ripple and magnetostriction which can wear through your windings.
In terms of nice to FET's you want the lowest PWM freq possible. Fets heat up with increased switching speed (basically).
So really for an ideal system you need to find the balance point between where your motors blow up and where your ESC blows up, and make them both happen at the same time.
Or you say ~3khz will do most jobs and don't worry about it too much ;->
Smarmy gits with thingaps should know better than to run them off victors which sound like they PWM at 150Hz or so anyway.
In short, the ideal ESC provides a smooth DC supply to the motor.
Torque is a function of power, you won't get more torque without applying more power.
Long PWM is just a way of applying more power, however it uses the mechanical time constant of the motor to average out the pulses of power it receives so it appears that the end result is the same. It will pull a whole bunch more amps doing that job though and as heat lost is I^2*R where I = Amps and R = resistance it will make things hotter.
Amps are bad, use less amps. _________________ Mechanical engineers build weapons, civil engineers build targets
|
Sun May 20, 2007 10:33 am |
|
|
|
|
|
|
|
|