BLDC motor drive with short circuit protection?

[cupoftea]

Hi,
We are doing a 400W two-phase BLDC motor drive from an inverter Bridge.
The inverter is supplied by a 28VDC rail.
We have a back-to-back FET switch in the connection wire from the inverter to the BLDC motor.
Our contractor put this in.
The idea is that we simply switch this switch off in the event that the motor and inverter both go short circuit.
However, this would obviously be breaking an inductive current so I believe you would agree
that this is a bad idea?
I believe that this back to back FET switch should really be instead placed in the 28VDC rail supply line, just upstream of the 28VDC rail capacitor.
Would you agree?
Also, I believe it should be back-to-back IGBTs (with added diodes) rather than done with FETs ..because FETs are too delicate to break
a short circuit current?

Also, our contractor has implemented the inverter with no current sensing or limiting in the inverter bridge. This surely is
not the way to go? I have never heard of a BLDC inverter with no current sensing and thus no current limitation?

I cannot really give too much details of the application unfortunately. However, the motor will need to turn a handle
at certain times, and keep the handle in place, and quite a bit of torque will be required for this keeping the handle in place.

Everything that our contractor has done is making me concerned because when the motor is holding the handle open, the motor would be static, and so would surely need to implement
that algorithm which allows a motor to give maximum torque when static (not rotating). I dont see any of this algorithm in our contractors work.
(The control algorithm for BLDCs that allows for max torque at zero speed is called Field Oriented Control)
Also, i thought that Field Oriented Control requires a 3 phase inverter, and our 4FET inverter (two phase BLDC drive) will surely not be enough to implement the algorithm for max torque at zero speed?

Also, its only 28VDC input to the inverter, but the hi side FETs dont have protection zeners from gate to source. Surely this is not a good idea?
I mean, the customer has requested 100V FETs so there must be transients etc going on at the 28VDC rail?

What are your thoughts?

 

[dick_freebird]

If the motor and inverter are both shorted, who's
going to cry over a fusible link? Why be fancy when
the rest is fried (per premise)?

If the FET drive is a high-peak amps CMOS / DMOS
type then there's plenty of stiff clamping by the
isolation diodes to the driver rails, and then if the
gate were to be overvoltaged it would have to
be the bootstrap / aux rail run away, and then it's
likely any small size zener clamp would get fried
bond wire and fail to protect anyhow.

As usual my advice is to instrument it up and see
the real answers to your questions.

 

[cupoftea]

If the motor and inverter are both shorted, who's
going to cry over a fusible link? Why be fancy when
the rest is fried (per premise)?

Thanks, its because the 28V rail gets used for other stuff so dont want it getting taken down...and a fusible link may be a bit slow to blow.....but i see your good point anyway.
In fact, i am just wondering why this 2FET switch has been put between inverter and motor.
Whatever, our contractor recomends a Photovoltaic FET driver to turn the 2FET switch ON and OFF....which sounds like bad news.....just too slow and the FETs will fry because in the linear region for too long.

So yes, just re-read your excellent points.....yes you make the great point that the 2FET switch is probably going to get fryed anyhow. Which i would agree with. They are asking for 750V of isolation from the 2FET switch drive pins to everything else, so it sounds like they are expecting a high voltage coming from the broken inductive current.

 

[FvM]

Photovoltaic FET driver to turn the 2FET switch ON and OFF....which sounds like bad news.....just too slow and the FETs will fry because in the linear region for too long.

Look at recent types like VOM1271 with fast turn-off circuit.