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.

IGBT overshoot voltage problem

Status
Not open for further replies.

abc_de

Full Member level 5
Joined
Jan 9, 2014
Messages
243
Helped
11
Reputation
22
Reaction score
11
Trophy points
1,298
Location
Ludhiana ਪੰਜਾਬ
Activity points
2,939
Hi
I am working on 3 phase inverter to control 3 ACIM motor.
Motor is running ok. I have check waveform across winding of motor. Then I found voltage overshot. Can someone suggest me how to prevent overshot.
Thanks
 

Hi,

overshot can be suppresed with filters and/or voltage limiters like TVS/Zeners, Varistors....

Just to be sure...
Often the signals on the scope are worse than in reality. This may be caused by wrong (adjusted) probe, GND connection, wiring and others.
Be sure to measure the signals correctely.

Could you send us a picture of a scope screen and your connections.

Maybe you can test your probe adjustment on a "clean" square wave just to check if the probe is OK.

Klaus
 
  • Like
Reactions: abc_de

    abc_de

    Points: 2
    Helpful Answer Positive Rating
Hi
Over shoot is observed at output of inverter 325v dc PWM.
Also I have check the circuit with 24vdc from SMPS then I found clean waveform
 

You will find that it is a problem with your layout. If there is inductance in your layout then any voltage overshoot will be V = L.di/dt where L is your layout inductance and the di/dt is the rate of change of your IGBT going off. This di/dt has to be high to reduce switching losses. If you are seeing this during normal operation and if your inverter ends up in trouble and has to turn off a strong fault current then you will lose your inverter because the voltage overshoot will be too much for the device to survive.

You will need to put some low ESR capacitors across your DC bus as close to the device terminals as possible. This is to absorb the energy held in the parasitic inductances.

In a normal operation and with a good layout, additional snubbing apart from the DC caps is unnecessary.
 
  • Like
Reactions: abc_de and FvM

    FvM

    Points: 2
    Helpful Answer Positive Rating

    abc_de

    Points: 2
    Helpful Answer Positive Rating
Thank you
I have already put de cap across dc bus. Yes may be layout is cause. Shoot is probably under voltage during turn off time. Then how do I can prevent this tiny sharp shoots. May be by adding RC circuit across dc bus or across each igbt
 

Well apart from improving the layout you could slow down the devices with a higher gate resistor value. This will impact your switching losses though so be aware. You do have some gate resistance?
 
  • Like
Reactions: abc_de

    abc_de

    Points: 2
    Helpful Answer Positive Rating
Ok yes I am using 22ohms resistance for turn on and turn off igbt
Now I will try to reduce the value of turn off resistance and rc circuit across dc bus of each Half leg or each igbt and try to change the dead time 1-2usec.
May be these efforts prevent voltage shoot

- - - Updated - - -

HTML:
http://obrazki.elektroda.pl/6414247600_1434466108.jpg
This is picture �� of output waveform

- - - Updated - - -

HTML:
<a title="WP_20150614_17_53_17_Pro.jpg" href="http://obrazki.elektroda.pl/6414247600_1434466108.jpg"><img src="http://obrazki.elektroda.pl/6414247600_1434466108_thumb.jpg" alt="WP_20150614_17_53_17_Pro.jpg" /></a>
 

Attachments

  • WP_20150614_17_53_17_Pro.jpg
    WP_20150614_17_53_17_Pro.jpg
    459.1 KB · Views: 80

There should be a defined signal either positive or negative without the noise around it.....have you looked at the gate waveform? It is easier to look at it with your fundamental set at 0Hz. Difficult to help without actually being there but still trying...
 
  • Like
Reactions: abc_de

    abc_de

    Points: 2
    Helpful Answer Positive Rating
Hi,

if the inverter is a sine modulated PWM, then the waveform isn´t that bad.

If i see Y-axis correct, then it is 50V/div. so max. screen voltage is about +/-200V.
What is your DC rail voltage?
Expect the outpult voltage to be +/- your DC rail voltage.

Then i see 125kSmpl/s. This could be a bit low for the PWM signal. What PWM frequency do you expect?

If you want to see a sine at the output, then you need to add LC filters. But since the motor is an inductive load the filters are not necessary.

Klaus
 
  • Like
Reactions: abc_de

    abc_de

    Points: 2
    Helpful Answer Positive Rating
Hello
PWM frequency is 20khz, dc rail voltage is 325vdc
I have to run 3ACIM which is already running. But my purpose is to make waveform disturbance free clear sharp, I have seen in professional 3 phase inverter for 3ACIM have very clear output.

- - - Updated - - -

May be that is third harmonic distortion &#55357;&#56862;
 

- Show a zoom view of the overshooting pulses you are complaining about.
- Better show both bridge output voltages instead of phase-to-phase voltage.
- If the phase voltages raise doubt about the actual switch control, also show the respective gate waveforms
 

Hello fvm
I will upload zoom veiw of voltage shoot.
Right now I have shown phase to phase voltage then what I have to show half leg output w.r.t negative of dc bus ?
Gate waveform without load or with load ?8-O

- - - Updated - - -

I have reduce igbt gate turn off resistance also reduce the dead time then I found shoot get reduced from last uploaded photo. &#55357;&#56837;
 

Yes, phase voltages against DC- would be fine.

Gate voltage waveform may be interesting if the phase voltages are inplausible and you'll expect false switching.
 

Hello
I checked the problem is associated with PWM of microcontroller, PWM from microcontroller is not clear as expected therefore resultant waveform is distorted.&#55357;&#56866;
I have to improve PWM now PWM carrier frequency is 20khz
My requirement is I have run 1hp motor.
Can someone suggest me best switching frequency range for better results ?

In Vfd mostly 16khz PWM carrier frequency is selected
 

Does your Microcontroller have hardware PWM generator? If you are doing it in software then it will not leave a lot of processing cycles for calculating your PWM ratio. Is this just a dumb V over F algorithm or something more exotic like FOC?
 

Hi,

some hardware PWM periferals have different setups when they update the new pwm value.
Don´t use the "immediate" update setting. Better choose "top", "bottom" or "match", if it exists.

Klaus
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top