Dear Emad Roshandel,

could You show the schematics of discussed circuits?

Regards,

Lukasz Niewiara

Dear Dr. Lukasz Niewiara ,

Thanks a lot for your consideration.

The IGBT has been placed in series with load which has been shown in the attachment. It must be explained that I have no accurate model of the load. In fact, the experimental load is an underground pipe and the available capacitors are resulted from the relationship of the pipe with the ground. The device is used for the cathodic protection of the pipes and works as an interrupter for testing the protective layer of the pipes.

Furthermore, it should be noticed that in the laboratory when a resistance is used as a load the circuit works well.

Best regards,

Emad

Dear Emad Roshandel,

what kind of current flows in the circuit AC or DC ?

Regards,

Lukasz Niewiara

Dear Dr. Lukasz Niewiara

DC current flows through the circuit.

Regards,

Emad

Dear Emad Roshandel,

please, give me some more information about the circuit, i.e. transistor model, gate driver, the value of load current.

Regards,

Lukasz Niewiara

Dear Dr. Lukasz Niewiara ,

Thanks a lot for your follow up. The IGBT is SKM100GB12T4 and its datasheet has been attached here. A TLP250 gate drive IC has been utilized to command the switch. Also, the switching frequency is in the range of 0.05-1 Hz. About the load current I can say that when the gate pulse commands the gate (Vgs=15), the current is 7 amperes. On the other hand, the switch current is 3 amperes when the gate drive does not command the switch (Vgs=0).

Furthermore, the switch voltage is in the range of 2 to 20 volts. It must be noticed that when the switch is off in terms of gate command, the voltage difference between the switch's emitter and its source is 4 Volts.

Best regards,

Emad

Dear Emad Roshandel ,

in my opinion the transistor is incorrectly controled or maybe it is not correct connected into the circuit. Note that the module SKM100GB12T4 consist two transistors with two antiparallel diodes. Please check the connections. On the other hand, it may be a problem with the gate voltage level, try to use negative voltage for transistor turn-off.

Regards,

Lukasz Niewiara

Dear Dr. Lukasz Niewiara,

Thanks a lot for the info. I must explain that the Module connection has been connected correctly to the circuit and it has been tested several times in the laboratory in presence of resistive loads.

I guess I have to provide a specific isolation between the switch and load. What is your opinion in this regard?

Regards,

Emad

Dear Emad Roshandel ,

what do You mean with a specific isolation between the switch and load?

Can You give me a detailed information about the gate driver circuit?

Regards,

Lukasz Niewiara

Dear Dr. Lukasz Niewiara,

About the gate drive circuit, assume that a command signal comes from the processor to TLP250 optocoupler and at its output an isolated power supply provides 15 volts voltage to command the switch gate. A 18 ohm resistor has been used between the output of the TLP250 and the IGBT gate.

I think about a transformer to isolate the switch from the load. However, it is not possible because of the transformer size. Small switching frequency and 100A require a huge transformer to be installed while just a small space is available.

Best regards,

Emad

Dear Emad Roshandel ,

try to use a power supply with negative voltage -15 V for off-state. Do You use a floating ground for the gate driver circuit? Alternatively You can to use a relay instead of a transistor.

Regards,

Lukasz Niewiara

Dear Dr. Lukasz Niewiara,

Thanks a lot for the follow up.

Yes, I use a floating ground for the gate drive. I will use the -15 volts for off-state and check the result based on your suggestion.

Best regards,

Emad

How is the Rgate value?!

Dear Kayhan Ince,

Rgate = 10 ohms.

10Ohm is ok.

Please connect R=10K with 2 zener diodes (15V) parallel to GATE as shown below.

http://tinyurl.com/y3mesbmt

regards

Dear Kayhan Ince,

Thanks a lot for your valuable hint.

Regards,

Emad