Ideas for Designing 3 Phase Power Stage For BLDC/ACIM Motor

Thread Starter

ashkarmalik

Joined Jul 1, 2012
112
Hello People I got the boards now.
So I did some Work on populating the boards to get H-Bridge Running as I intended to use the Board For Multiple Motor Control Unit Brushed/Brushless/ACIM/PMSM.
So For First Stage I am Going to set the Configuration for Brushed Motor on the Cards.

I am Experiencing Negative Shoot through on the H-Bridge.
 

Attachments

cmartinez

Joined Jan 17, 2007
8,220
Hello People I got the boards now.
So I did some Work on populating the boards to get H-Bridge Running as I intended to use the Board For Multiple Motor Control Unit Brushed/Brushless/ACIM/PMSM.
So For First Stage I am Going to set the Configuration for Brushed Motor on the Cards.

I am Experiencing Negative Shoot through on the H-Bridge.
Did you place protective diodes across source and drain on each Fet? ... are you using a tvs across the motor?
 

cmartinez

Joined Jan 17, 2007
8,220
That shoot through is normal when switching inductive loads, and if you don't add a diode between source and drain of each fet, you run the risk of blowing them. A tvs across the motor is unusual and not essential, but I usually like to add it as extra protection.

The IRF540N datasheet says that it's designed to work at up to 100V, and I can see from your scope picture that you're feeding 24V to the motor, right? If that's the case, then I strongly suggest you at least add the didodes I've suggested, or you run the risk of damaging the fets due to the shoot through you've mentioned. Inductive kickback voltage can easily be more than five or six times (even 10 times) that of the circuit's working voltage.
 
Last edited:

Thread Starter

ashkarmalik

Joined Jul 1, 2012
112
That shoot through is normal when switching inductive loads, and if you don't add a diode between source and drain of each fet, you run the risk of blowing them. A tvs across the motor is unusual and not essential, but I usually like to add it as extra protection.

The IRF540N datasheet says that it's designed to work at up to 100V, and I can see from your scope picture that you're feeding 24V to the motor, right? If that's the case, then I strongly suggest you at least add the didodes I've suggested, or you run the risk of damaging the fets due to the shoot through you've mentioned. Inductive kickback voltage can easily be more than five or six times (even 10 times) that of the circuit's working voltage.
Thanks for the clarification.
I want to use Mosfet body Diode for this purpose can this Be feasible if I stick with IRF300P226.
 

Attachments

cmartinez

Joined Jan 17, 2007
8,220
Thanks for the clarification.
I want to use Mosfet body Diode for this purpose can this Be feasible if I stick with IRF300P226.
You could. That fet has better specs than the previous one (up to 300V drain to source). But it's normally good practice to add those diodes. Are you hesitating because you already made the pcb?
 

Thread Starter

ashkarmalik

Joined Jul 1, 2012
112
You could. That fet has better specs than the previous one (up to 300V drain to source). But it's normally good practice to add those diodes. Are you hesitating because you already made the pcb?
I would say you are correct on the PCB Part, I will be revising the PCB but the main reason of not placing external diode is to keep inventory as small as possible, Since External diode will definitely increase the inventory and on top of that I have Put multiple Mosfet for Large current handling capabilities as seen in the Schematic so there would be 12 Diodes in total.
What are your thought on this mater?
 

Attachments

cmartinez

Joined Jan 17, 2007
8,220
What are your thought on this mater?
Use those diodes, you won't regret it. When you mean inventory you're referring to cost? Or to available space? ... if it's the former, it might be more expensive not to have those diodes ... if it's the latter, you can always patch things up a little, or use a small footprint smt diode type.
 

Thread Starter

ashkarmalik

Joined Jul 1, 2012
112
Use those diodes, you won't regret it. When you mean inventory you're referring to cost? Or to available space? ... if it's the former, it might be more expensive not to have those diodes ... if it's the latter, you can always patch things up a little, or use a small footprint smt diode type.
I got the Point you are making and I will Go with your solution.
Could you please clear a bit how to calculate the Diode Ratings for the same!
 

Thread Starter

ashkarmalik

Joined Jul 1, 2012
112
You're right ... I just don't see this component type as too critical ... what's critical is that it must be present in the circuit.
Because of lockdown nothing is getting though.
I do have some 1n4007 With me but that doesn't solve any negative peaks what so ever.
I believe Schottky Diodes are qualified for this project.
 

cmartinez

Joined Jan 17, 2007
8,220
Because of lockdown nothing is getting though.
I do have some 1n4007 With me but that doesn't solve any negative peaks what so ever.
I believe Schottky Diodes are qualified for this project.
I forgot to mention, using dioides between source and gate is also recommended.

Where is this shoot through happening? Between the load's leads? If that's the case, you need a tvs. You can also supplant it by using two zeners connected back to back.
 

Thread Starter

ashkarmalik

Joined Jul 1, 2012
112
I forgot to mention, using dioides between source and gate is also recommended.

Where is this shoot through happening? Between the load's leads? If that's the case, you need a tvs. You can also supplant it by using two zeners connected back to back.
I have Now Started to work on the Testing part of the System now.
So I started to test the System with 1 Amp Load current @ 24VDC
But I see lot of trouble on the Mosfet gate side, I have attached couple of pictures of the Gate Waveform.IMG20200913144916.jpgscope_0.pngscope_1.pngscope_11.png
This Waveform is of High Side Mosfet Gate and Low side Gate With respect to Ground.
 
Last edited:

cmartinez

Joined Jan 17, 2007
8,220
I have Now Started to work on the Testing part of the System now.
So I started to test the System with 1 Amp Load current @ 24VDC
But I see lot of trouble on the Mosfet gate side, I have attached couple of pictures of the Gate Waveform.View attachment 217081View attachment 217082View attachment 217083View attachment 217101
This Waveform is of High Side Mosfet Gate and Low side Gate With respect to Ground.
Did you place a resisitor in series at the mosfet's gates? That would help diminish the ringing you're seeing.
 
Top