IGBT Pulsed measurement

Thread Starter

kanoun

Joined Jan 29, 2015
41
I’m trying to make a pulsed IV caracterics (Ic-Vce) of an IGBT ref IKA10N60T. The aim is to have the output characteristics with no self heating. The pulse duration is about 100µs and the period is 50 ms. I’m using this set up. I have two different problems:

The Vce signal is very noisy

I have a 2A delta when compared to the datasheet.

Any hekp please ?
 

Attachments

Rocket.Man

Joined Jan 1, 2012
21
I have been experementing with Power Mosfets if they stay ON too long they make lots of heat. If you can set the duty cycle of your pulse so it is ON 10% or less then OFF 90% or less, much less heat is produced.

 

Thread Starter

kanoun

Joined Jan 29, 2015
41
This is basically the case but still have this problem ... Moreover, I'm measuring the temperature as well and it is not more than 30C
 

#12

Joined Nov 30, 2010
18,224
So many aspects! Is your resistor inductive? Do you have local capacitors on the test jig? Is your current probe defective? Are your grounds generous? Is your IGBT 2 amps better than typical or 2 amps worse than typical? Two amps worse than the minimum guaranteed performance?

Trying to replicate a professional test requires professional skills and professional equipment.
 

Thread Starter

kanoun

Joined Jan 29, 2015
41
Many thanks for this reply


Is your resistor inductive? Yes but what is the problem with this? In fact I did the measurement without the resistance and the results are the same.

The current probe is not defective since I measured the same current on a resistance

My IGBT is 2 A less than the output characteristics from the datasheet (I reproduced the Ic-Vc curve and I figured out that I have a 2 A less)
 

#12

Joined Nov 30, 2010
18,224
Is your resistor inductive? Yes but what is the problem with this? In fact I did the measurement without the resistance and the results are the same.

My IGBT is 2 A less than the output characteristics from the datasheet (I reproduced the Ic-Vc curve and I figured out that I have a 2 A less)
1) an inductive resistance can cause a lot of noise at the speed of your pulses.

2) Two amps less than, "typical" or two amps less than the minimum guaranteed performance?
 

Thread Starter

kanoun

Joined Jan 29, 2015
41
1) an inductive resistance can cause a lot of noise at the speed of your pulses.

2) Two amps less than, "typical" or two amps less than the minimum guaranteed performance?
I'm reproducing the IcVc curves and comparing to the datasheet. 2A less basically
 

#12

Joined Nov 30, 2010
18,224
If that is a "typical" performance curve there might be nothing wrong. If you refuse to look up the minimum guaranteed performance, I also refuse.
 

Thread Starter

kanoun

Joined Jan 29, 2015
41
If that is a "typical" performance curve there might be nothing wrong. If you refuse to look up the minimum guaranteed performance, I also refuse.
I'm not refusing. Is this information available in the datasheet ?
 

Thread Starter

kanoun

Joined Jan 29, 2015
41
I
Maybe a picture of your scope trace and the layout of the circuit would help solve the mystery.
If you are talking about the safe operating area curves they are difficult to produce and need more than your circuit.

http://www.infineon.com/dgdl/Infine...90004&fileId=db3a304412b407950112b428713e3dfa
I will post a graph from the scope ... Is it possible that the fact that I'm not driving the gate with a gate driver but only a pulse generator lessen the current (the actual gate voltage could be the one of plateau of Meyer)
One more question the rise time in my signal is not that steep could it explain this as well ?
I already posted the circuit im using
 
Top