Need a gate driver circuit for asymmetric bridge converter.

Thread Starter

Lomesh

Joined Jan 31, 2018
11
I am designing an asymmetric bridge converter for 3KW 3phase SR motor. Each phase draws a current of 4A. I want to use a gate driver to turn on and off the MOSFETs. The DC link voltage is 400V. Can anybody provide me with the gate driver circuit connections to perform simulations on this type of converter just to start with?

Thanks in advance.
 

shortbus

Joined Sep 30, 2009
10,045
I know nothing about LTSpice but see no path to ground to charge any of your bootstrap capacitors for the high side mosfets to turn on.
 

Thread Starter

Lomesh

Joined Jan 31, 2018
11
I know nothing about LTSpice but see no path to ground to charge any of your bootstrap capacitors for the high side mosfets to turn on.
Thanks for the response. I fixed it now. But the simulations are running very slow.
Apart from that, the rest of the circuit connections values should be alright?
 

crutschow

Joined Mar 14, 2008
34,462
Why are you using a high-side connection from the driver for the low-side MOSFET, Q2?
Isn't Q2's source connected to ground?

Why is the top MOSFET labeled M1 and the bottom one labeled Q2?
 

Thread Starter

Lomesh

Joined Jan 31, 2018
11
Thanks for your quick questions which were useful to solve come some of my mistakes.
I overlooked and didn't connect the source of my lower mosfet to ground.
I replace the second mosfet and now it is labelled as M2.

From your other question, are you suggesting me to connect the low side of the mosfet to the same gate driver but to the low output?
I am trying to switch on both the mosfets at the same time
 

Thread Starter

Lomesh

Joined Jan 31, 2018
11
Why are you using a high-side connection from the driver for the low-side MOSFET, Q2?
Isn't Q2's source connected to ground?

Why is the top MOSFET labeled M1 and the bottom one labeled Q2?
Why are you using a high-side connection from the driver for the low-side MOSFET, Q2?
Isn't Q2's source connected to ground?

Why is the top MOSFET labeled M1 and the bottom one labeled Q2?

Thanks for your quick questions which were useful to solve come some of my mistakes.
I overlooked and didn't connect the source of my lower mosfet to ground.
I replace the second mosfet and now it is labelled as M2.

From your other question, are you suggesting me to connect the low side of the mosfet to the same gate driver but to the low output?
I am trying to switch on both the mosfets at the same time
 

crutschow

Joined Mar 14, 2008
34,462
are you suggesting me to connect the low side of the mosfet to the same gate driver but to the low output?
Yes, the driver IC has two drivers, one with a bootstrap to drive the upper MOSFET, and one without to drive the lower MOSFET (the one connected to ground).

Have you read the driver's data sheet?
 

Thread Starter

Lomesh

Joined Jan 31, 2018
11
Yes, the driver IC has two drivers, one with a bootstrap to drive the upper MOSFET, and one without to drive the lower MOSFET (the one connected to ground).

Have you read the driver's data sheet?
Based upon your suggestion to use IR2110 and not using bootstrap capacitors to low gate driver, I made another gate driver circuit connections. Could you check if my connections and simulation results are good?
 

Attachments

crutschow

Joined Mar 14, 2008
34,462
Why are you using two driver chips?
One chip will do both the high-side (HO) and the low-side (LO) switching.

You circuit is obviously ringing with the inductive load.
What will be the real load characteristics?
 
Last edited:

Thread Starter

Lomesh

Joined Jan 31, 2018
11
I tried with two drivers because the simulation results shows error saying that the time step is too small with one driver IC. If you are using LTspice would you please check the simulation of the model I attached below?

The real load is a 3KW SR motor with a coil inductance of 0.38H.

I wanted to have the simulations of this real motor, but the current in the simulations are very low than what I'm expecting. I assume I need to add few more components in addition to the inductor in the design in order to to see actual motor current. But I've scratching my head sitting days in front of computer doing research on this. I would be very grateful to you if you could you give an advise on it?

upload_2018-2-22_12-47-1.png
 

Attachments

crutschow

Joined Mar 14, 2008
34,462
I assume I need to add few more components in addition to the inductor in the design in order to to see actual motor current.
What series resistance do you have for L1?

Those MOSFETs have a 0.25Ω on-resistance which means they will dissipate 4W when carrying 4A, so will need to be on a heatsink.
 

Thread Starter

Lomesh

Joined Jan 31, 2018
11
A SRM is one of my on going projects. I'm sure you have probably seen all of these links but if you haven't here they are. Almost every chip maker has a paper on the driving of those motors.

https://www.google.com/search?q=swi...r+driver&ie=utf-8&oe=utf-8&client=firefox-b-1
Thanks Short bus. I've been through those links before. At this point of time I have to show the simulations to my manager without actually buying them. I'm using LTspice to create those by my own (gate drivers, controllers and converters). I am trying to create a model and having problems here. As I'm a newbie to this field, the pressure is intense on me with dead lines. Can I kindly ask you to share any models and circuit connections that I can actually run simulations and show to my manager?
 

shortbus

Joined Sep 30, 2009
10,045
I'm more into building the motor it's self. But that said, my way of thinking is how some of them are run, before total digital got involved in this. They will and can be run by only low side switching. By having one leg of all three phases hot all of the time. And individually switching them to ground at the appropriate time. The only thing is with going that direction for commutation is that it needs to use position sensors, either Hall type or optical. It takes the possibility of sensor less of the table, in other words it is "old school". Which for my case is OK, since mine is just for personal use not to be 'marketed'.
 

Thread Starter

Lomesh

Joined Jan 31, 2018
11
What models do you need?
The model I created for SR motor with inductance of 0.38 and series resistance of 52 ohms looks like the image below. Green represents the current through the resistor and blue represents the voltage across the resistor. Could you please tell me if this is the right way to create the equivalent circuit of the motor? If not would you please give me a reference model so that I could make changes changes to my circuit based upon the reference motor?

Thanksupload_2018-2-27_17-16-6.png
 
Top