What does PWM do?

[swbluto]

I know it's used in a SMPS for transforming input voltages to output voltages appropriately(over a given range), but what's its role in a typical e-bike controller? Particularly, what is its purpose during acceleration? Does PWM have a special place in the "brushless world" that it doesn't have in the "brushed world"?

I'm thinking you can adjust the final top-speed by limiting the output voltage(and the corresponding amps) and the corresponding power, so I guess it could be used to "control" the power output for a given throttle input: That's understandable. But during acceleration, if that was PWM's only role, why couldn't the controller just apply the nominal battery voltage directly to the motor?

 

[dirty_d]

if there was no pwm then you would just have motor off and motor on and you need pwm sin some form to limit the current.

 

[swbluto]

Indeed. But, what about during acceleration? During 100% throttle from a stop, it seems rumored that PWM engages(where that "rumbling" comes from) and stops PWMing once it starts coasting. If PWM's only purpose was to limit the current(by reducing output voltage? That still hasn't been made clear.), why not just blast the current full-charge for maximum acceleration? Is it to protect the fets(Not that I know, but does pushing too much current through a FET continuously for too long cause it to die? So, then, PWMing purpose during acceleration would be to prevent that FET damage since the wheel is slowly moving and the FETs wouldn't be switching fast enough otherwise?)?

 

[Link]

PWM is just the modulation of the duty cycle of the (typically) 18kHz square wave signal that is the power supply of the motor (from the controller). It's like flipping a switch really fast and leaving on a certain amount of the time allotted for each flip.

The more throttle you give it, the "wider" the signal looks. At 0%, it's just no power and is a flat line of no voltage. At 50%, it's on half the time and off the other half and looks like a normal square wave. At 100%, it's a flat line with a DC voltage bias (your battery voltage). The higher the duty cycle (%), the higher voltage the motor gets and the faster it spins and the more current it wants to draw.

The PWM itself does NOT limit the current. What happens during current limit is that the controller cuts back on the duty cycle, which is the same thing that happens when you let off the throttle a little. This causes the motor to draw less current. As the motor starts to spin up and its BEMF starts to catch up, the controller increases the duty cycle accordingly, up to what the throttle says it should be.

Also: It's not the PWM that rumbles the motor. It's the phases switching on and off. The PWM makes a high-pitched whine that most people (especially those over 25 or so) can't even hear.

 

[swbluto]

Also: It's not the PWM that rumbles the motor. It's the phases switching on and off. The PWM makes a high-pitched whine that most people (especially those over 25 or so) can't even hear.

Oh, so that's what that whine is! Haha. (I suspected that's what that whine was.)

So, does the duty-cycle ever reach 100% or is the controller usually preventing it? If the duty cycle is 100%, then what would that mean the motor's current isn't being limited(i.e., the motor is naturally drawing less amps than the controller's limit? Like, say, near top speed on a 4099 with an infinity-amp controller on flat-land.) by the controller?

So... during acceleration from a stop... the motor wants to draw much more current than in other modes of usual operation, so PWM(I'll define that as when the Duty Cycle is less than 100%) is almost always engaged in this mode since the current is usually being limited by a finite-amp controller?(usually less-than-possible due to economic reasons? i.e., not enough money for the required batteries, the required controller with higher requirements for the components(thus more expensive), the required wires, etc.)

 

[Link]

So, does the duty-cycle ever reach 100% or is the controller usually preventing it? If the duty cycle is 100%, then what would that mean the motor's current isn't being limited(i.e., the motor is naturally drawing less amps than the controller's limit? Like, say, near top speed on a 4099 with an infinity-amp controller on flat-land.)

Exactly. Once the motor starts drawing less than the controller's limit (when you're at full-throttle), then the duty cycle hits 100% and the FETs are just on all the time, giving full battery voltage to the motor.