Bootstrap Calculation for BLDC Motor

As I understand, the best way to run BLDC motors is to PWM only the High side switches and Low side ones are kept ON or OFF depending upon the commutation logic. Some even suggest using simple commutation of all 6 switches with speed control achieved by a single switch that controls the DC Link voltage, this way minimizing switching losses and getting best possible efficiency.

My question is wouldn't that mean that the the bootstrap cap will have to keep the high side switches on for a really long time without getting a refill till the next time the low side switch is turned on. Also this time will be dependent on motor speed. At high speed, it will recharge after shorter intervals, but at low speed, after quite long intervals.

So does the bootstrap technique work for such long intervals with a large enough bootstrap cap say 100uF.

Motor in question, btw is 120W, 24V, 1500RPM BLDC.

Hi,

My question is wouldn't that mean that the the bootstrap cap will have to keep the high side switches on for a really long time without getting a refill

Click to expand...

As you said before the low side is kept ON, this means the bridge output voltage is low. During this (long) time the bootstrap is charged.
It is discharged when the high side is ON, means the bridge output is high. Therefore you should not use 100% duty cycle. 99% means the output is low for 1%, enough to charge the bootstrap again.

Klaus

Klaus, the first part I understand that low side is on for a LONG time and the bootstrap gets nicely charged.

But duty cycle, according to me won't make a difference here as even when the high side switch turns off during PWM, the switch below it is also off, meaning that bootstrap is not charging.

It will charge only when the low side is turned on, i.e. when the rotor moves to the next section and commutation logic turns the lower switch on.

- - - Updated - - -

www.20sim.com/webhelp/library_signal_block_diagram_non-linear_signallimiter-pwm.php

Just check out the waveforms in this page.

Bootstrap for bipolar, no issues.

But unipolar, should work theoretically, but does it practically without issues?

Hi,

But duty cycle, according to me won't make a difference here as even when the high side switch turns off during PWM, the switch below it is also off, meaning that bootstrap is not charging.

It will charge only when the low side is turned on, i.e. when the rotor moves to the next section and commutation logic turns the lower switch on.

Click to expand...

Does it make sense to switch OFF the low side the whole time during PWM the high side? I think it´s better to switch both sides during PWM. (Take care of dead time)

On the other hand even if the low side is not switch ON, the bootstrp capacitor most probably gets charged by the current through the motor coil.
Check this with the waveform during PWM. If there is (even a a short) time where the bridge_output is near GND, then the bootstrap gets charged.


Klaus