BLDC motor from grid directly operation

[Salvador12]

Hello

I have a problem, I have a fan motor with a BLDC motor. It has a circuit board with IC's and a drive IC and it's a typical BLDC board with the rotor position sensors etc. The problem is the board is toast but the motor is fine. The motor is wound as a 3 phase delta motor with 3 wires. Each coil has 67 ohm of resistance as measured.
Now given these types of motors have a permanent magnet rotor. I was wondering do I need a controller board at all, in theory can I just wire the motor to a 3 phase supply and it should run?
It's essentially a synchronous motor right?
Do any of you have any such prior experience and can you give me any advice before I try to do this?

The original board was powered from a single phase source where the IC simply did the switching and created the 3 "pseudo" phases from the rectified mains supply.
The only thing I am not sure is whether these 3 pseudo phases had the same voltage as my normal 3 phase grid has, namely 400v.
Maybe since it's just a fan application I can use single phase with a capacitor , not knowing the motor coil voltage rating that would be safer in theory.
What do you think?

 

[KlausST]

Hi,

i guess it won´t work directly on mains:
because
* frequency needs to ramp up
* voltage needs to ramp up
* and the voltage needs to match to the motor (not sure the 400V will fit to the motor at mains frequency)

Maybe you can get/lend some VFD to make some tests. I´d do tests with a serial resistance at each motor line.

What I´d try:
Run the motor (unloaded. No wings) with constant voltage, the series resistors should avoid motor damage (current limit)
Then step up the frequency and at a certain frequency there should be lowest motor current.

With the frequency and the motor voltage you can do calculations for the BLDC driver setup.

I´ve never done this before. So, no guarantee that this work as I expect...

Klaus

 

[c_mitra]

The original board was powered from a single phase source where the IC simply did the switching and created the 3 "pseudo" phases from the rectified mains supply.
The only thing I am not sure is whether these 3 pseudo phases had the same voltage as my normal 3 phase grid has, namely 400v.

This is a BLDC motor and the stator is fed with a three phase power.

There are real phases, not pseudo phase as you say. First we make a DC from mains, then create three phase sine waves. Your controller board is basically a 3-phase source and the frequency is set by software (or your remote speed controller).

No, you cannot run the fan from the mains directly. You will get lots of smoke but no fresh air.

Your best option is to repair the original board OR get a VFD that can supply 20-40W of power. I guess that is the usual power rating for domestic room ceiling fans. If this is an industrial high power fan, get a more powerful VFD.

Your motor may take around 0.25 to 0.5A current at 50-100V, depending on the design and power rating.

 

[Salvador12]

Ok so I tried powering the motor (3 phase delta) BLDC from a single phase supply using a capacitor. I used a step down transformer so my motor voltage was 100 volts AC.
I tried various capacitances from 4uF up to 16uF but the motor doesn't want to start turning.
My idea for this is that the field from the permanent magnets in the rotor is not strong enough for the motor to start with a steady frequency because the original board that made 3 phase current from rectified DC serves as a frequency converter essentially and I think it starts the motor with some very low AC frequency and then gradually ramps it up to the set frequency that matches the RPM that are commanded from the motor through the control input.

 

[KlausST]

Hi,

but the motor doesn't want to start turning.

As you already mentioned you just don´t follow the requirement (from post#2)

* frequency needs to ramp up

It is a synchronous motor that usually doesn´t start at nominal frequency.
(Unless specially designed to do, like synchrounous clock motors)

Klaus

 

[cupoftea]

yes you need to have a rotor position sensor to drive a BLDC......or use sensorless control, where you assess the rotor position from the back emf voltages.

You need to know the motor coil inductance....and the motor coil current......then basically you commutate from coil to coil, in accordance with the rotor position sensor.

So you bascially set a coil current limit, and PWM the coils one-by-one, and keep a constant current in the coils by PWM....so the current is actually sawtoothing up and down, but you can regard it as "constant".

Since your application is a fan...you can actually do it without rotor position sensing though,
You start commutating from coil to coil at low frequency, and basically entice the rotor to spin.......no jerky changes in commutation frequency, nice and smooth, up to max speed...and the rotor will follow the stator field around........but if you jerk the frequency of commutation, your rotor will go out of synch and youve had it.
...in truth, what i described is difficult, and you are better off sensing the rotor position, and then energising the stator in accordance with that.

You can even just set up the three coils with a current source driving each one....and commutate from one coil to the next like that..just be sure to turn the current source off during the commutation interval otherwise the voltage gets high...but you ccan address that with a comparator sensing the voltage.

But yes, dont put mains voltage onto it...you need to limit the current in the coils.......

 

[Easy peasy]

the control board had 6 mosfets or igbt's to allow each phase to vary in frequency and effective voltage - you appear to have none of that - switching of a 6 switch bridge at the right instants is necessary to make the motor work.

 

[Salvador12]

Yes indeed you guys are right I thought about it. Sure mains is bit too much for this motor , I drove it from less voltage via autotransformer but indeed it cannot start when the frequency is already high it needs to be 'pulled" up slowly.
One alternative would be to use a frequency converter which has a function of starting with low frequency and then ramping up to a set frequency. But I just replaced the motor entirely with an induction motor since the drive circuit had lots of noname chinese SMD parts I figured it's not worth the effort.