Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Problem with 1kw bldc motor controller

Status
Not open for further replies.

nibitalovexuka90

Newbie level 4
Joined
Feb 20, 2013
Messages
6
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,321
Hi everyone!
I am building an 1kw bldc motor controller.I have a problem with 6 MOSFET IRF3205.All Fet are so hot although motor run without load.I use IR2103 to drive Inverter.Bootstrap capacitors are 10uF.
Futhermore,I tried to run 250W bldc in bicycle electric and MOSFETs are cool.
I use dspic30f4011 to control scheme.High-side is PWM and Low-Side is on in each 60 degree of electrical cycle as Application note AN957 of Microchip
What is going on with my MOSFETs?
Thanks all!My scheme is showed below
1.PNG

2.PNG
 

You didn't mention a switching frequency. Switching speed won't be mindblowing when driving IRF3205 with an unbuffered bootstrap driver IC.

Did you check the actual gate waveforms, particularly the break-before-make deadtime?

I don't think that you should use 1 nF capacitors at the PWM control signals.
 
thank you,FVM.
I measured Vgs of High-side and Low-side .It's ok.
Switching frequency is 15Khz.
Sorry FvM for my English skill is so bad.I don't understand exactly mean of "an unbuffered boostrap driver IC".Can you explain this to me?
 

What's the switching time of the FETs (Vds transitition time high-low and low-high)?

You said that the FETs are getting hot without a load on the motor. How much current is being drawn by the battery in this state? I wonder if shoot-through is a possibility.
 
I meant, that IRF3025 might need a stronger driver than IR2103.
Wow.I didn't hear about this.What driver do I need?What specification do I need to notice?

To Centmo:Transition time low-high:1.5us,and high-low:1us
When i apply 24V to circuit,Peak current exceed 30A.I use ACS712-30A.

I suffer a same problem when I don't use PWM.High and Low are only on state or off state.

Thank everyone!
 

With no load on the motor, it is consuming 30A RMS? 720W...are you sure?
 

No,it's peak current.I think that I have some problem with power commutation board.Something don't run correct.Can anybody help me? :(

- - - Updated - - -

No,it's peak current.I think that I have some problem with power commutation board.Something don't run correct.Can anybody help me? :(
 

Sorry, I misread your comment...30A peak. That's a different story. At high RPM with no load this can be normal since there is a large back-EMF in the motor which must be overcome. 30A is well below the FET current rating (110A at 25) so if the FETs are heating up this much then something is obviously wrong. 1 - 1.5us I would say is a slow switching time. I would want something in the 100-400ns range. It is possible that most of your heat is being dissipated during switching. You could either beef up your gate drive cct or use lower gate capacitance MOSFETs. This may not be the main problem though - there may be a more serious issue with your circuit. Do you have an oscilloscope? I would want to probe Vds of a high-side and low-side FET on the same inverter leg and make sure that they never overlap in the ON position (but you would either need a differential probe to do this or use two regular probes in differential mode for the high-side FET. Remember that regular probes must ALWAYS have their ground clips connected to the same point! If your circuit is battery powered, which it probably is, you could connect the ground clips to the motor-out point and a probe on high-side drain and another on low-side source, and compare them that way keeping in mind the low-side Vds will be inverted). The FETs are rated for 55V, and you are running 24V, so I don't think overvoltage is an issue. There may be something in the gate drive cct which is turning them on when they shouldn't be. Probe those gate signals. Also, timing could be an issue. I assume your cct uses zero-crossing detection? You'd have to make sure those signals are clean. Is the code known to be good, so you are certain this is a H/W issue on this board? Is this a prototype, or do you have one that works?
 
Last edited:
I think that IR2103 don't provide current for Gate MoSfet.Max current of IR2103 is 400mA.Due to Mosfet get hot.Perhaps,I will use IR2110 to replace IR2103.I need to search scheme using IR2110.
Thank u very much!
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top