# IGBT is broken after a few seconds

Discussion in 'General Electronics Discussion' started by Toby123, Apr 19, 2012.

1. ### Toby123

3
0
Apr 19, 2012
Hello!

The circuit shown in the drawing is working for about 30s, after that the IGBT 1.1 is broken.
The DC/DC converter actually is a 1215, which means 15V Output.
For testing I have connected a 60W lamp. (230V)
I don’t understand why only the IGBT 1.1 gets destroyed, since both IGBT’s are wired equal and do the same job.
Does anybody have an idea what is wrong in that circuit?

I hope you are familiar with the European symbols ;-)

File size:
40.4 KB
Views:
255
File size:
37.6 KB
Views:
334
2. ### Harald KappModeratorModerator

11,816
2,755
Nov 17, 2011
Are L1, L3 and L4 of the same type?

Harald

3. ### Toby123

3
0
Apr 19, 2012

L1 is 4mH
The filter L1 and with C21 (1uF) has a cutoff frequency of 2,5 kHz.
The IGBT´s are switched with 40 kHz.

L1 and L2 is 100mH
It´s two 100uH instead of one 50uH to divide the current.

4. ### (*steve*)¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥdModerator

25,500
2,840
Jan 21, 2010
L3 and L4 will be storing more energy than L1 (I presume you mean uH, not mH -- you seem to be confusing those terms and the diagram says uH (actually uF...))

5. ### Harald KappModeratorModerator

11,816
2,755
Nov 17, 2011
When you apply a switched voltage to an inductor the inductor will generate a counter voltage when the switched voltage changes from on to off. This is due to the inductor's equation U=L*di/dt. And since the currrent can't change instantaneously (due to the energy stored in the inductor) a high voltage with reverse polarity will develop. This explains why the IGBT is destroyed and why only one, not both: the inductances are very different.
Try to put a diode in reverse polarity across the inductor to limit the counter voltage.

Harald

6. ### Toby123

3
0
Apr 19, 2012

The testing circiut does not have the free-wheeling circuit, but it is supposed to be connected.
I will try, if it gets better when it´s connected.

@steve
You are right, the "UF" is confusing, it just was an existing part in EGLE.
And you are right, all the inductors have uH instead of mH. Sorry about that!

Thank´s again!

Last edited: Apr 20, 2012
7. ### Harald KappModeratorModerator

11,816
2,755
Nov 17, 2011
As an alternative to a freewheeling diode you could also place a zener diode from the Collector of the IGBT to GND, thus limiting the voltage spike.

Harald