I have problems understanding bootstrap circuit application note

[Futterama]

I have a motor specific question too. Is the commutation sequence the same for 2-pole and 4-pole rotor motors?

 

[shortbus=]

While this link isn't what you are trying to do, it does explain a lot about the communtation of BLDC motors. It's for the MC33035 IC data sheet.
https://www.google.com/url?sa=t&rct...T1SxTBHJcGNkIrMVJz7A-xQ&bvm=bv.51495398,d.aWc

 

[ronv]

I think there is a portion of each revolution where the boost cap will not be refreshed. I think that is why all the drivers have charge pumps if they drive NFETs on the high side. But it sounds like shortbus knows more about this than I do. Lets see what he has to say.

 

[shortbus=]

Ron, I've never done it with a micro controller. But the link I gave shows another way of doing it. And mostly is the best explanation I've found, of how the wave form to communtate the phases works. Just some backround for the OP to look at, not to takeover the thread.

 

[ronv]

No, no. Just looking for help. The real question is if the bootstrap cap gets refreshed at the pwm rate or at the commutation rate then pwm rate. I think that it is the commutation rate so it could be very seldom as the motor comes up to speed for example. That's why all the chips have charge pumps as well as bootstrap circuits if they use NFETs in the high side?

 

[Futterama]

ronv, look at section 2.2 and figure 2 in AN-6076, it clearly states that the bootstrap capacitor is recharged when the low-side switch is turned ON and the high-side switch is turned OFF.

 

[ronv]

Yes, but if that only happens every 2 or 10 seconds because the motor is just slowly turning the cap must support the power requirements and leakage for that long. Do the commutation sequence and see what it says.

 

[Futterama]

I'm in the process of making the hall sensor board to mount inside the motor.

I have also been examining the motor and I also connected the wires to a power source (through a power resistor) to see the different motor positions. When powering 2 wires, the motor can "jump" to two positions, opposite of each other, which is a property of the 4 poles. When turning the motor by hand, it has 24 positions. I don't get why, shouldn't it only have 12? Is it some inaccuracy in the rotor poles that does this?

Here is the motor data:

Rotor: 4 poles.
Stator: 12 poles.
Winding pattern: WYE.

I have identified the 4 poles (North/South) on the rotor.

I also hooked 2 of the power wires to my scope and turned the motor with a powerdrill, the BEMF looks a bit wierd to me, see attached.

 

[shortbus=]

On a BLDC motor there are 720° electrical for every 360° of rotation. Like I said earlier even though the link for the MC33035 is not a micro controller circuit, it explains how a BLDC works better than the actual books on BLDC motors do. And I have a few books.

The MC33035 will do anything a micro controller will as long as you are using Hall sensors. With very few added components. The chip was developed just for driving BLDC motors. No bootsrap circuit or other mosfet drivers needed. If I remember right they cost less than US$5 on Ebay.

The only advantage I can see with using a micro, is you can do it sensorless. But and it's a big but, a sensorless BLDC/ESC doesn't work well for a "traction" motor. Sommething for drivng a connected load. The sensorless controllers work fine in an airplane(propeller) drive though. When used in a traction situation the controller of the ESC type looses track of the rotor position too easily. Or at least that's the way it used to be, they may have made better circuits now.