HOW TO TEST AND MEASURE AN IGBT 1

[fernandoelec]

Please help me some one that knows how to measure an IGBT and how to test it.

I´ll appreciate your help

 

[dydt]

hi..
is it made up of 3 heavy terminals (A,B,and C) and 4 small gate lead terminals..?

this is standard ..and if this is the case ..do the following test..

Meter Hook Up DVOM (diodetest)
(+) A to (-) B OPEN
(-) B to (+) B 0.4 volts approx.
(+) A to (-) C 0.4 volts
(-) A to (+) C OPEN
(+) B to (-) C 0.8 volts
(-) B to (+) C OPEN

(+)D to (-)E open
(+)G to (-)F open


good luck...

keep in mind..power is OFF!...no wires or circuit connected to it..

dydt

 

[jbartos]

Suggestion: Visit

http://www.infineon.com/cmc_upload/migrated_files/document_files/Application_Notes/anip031e.pdf

for theory first,

http://www.sz.com/public/english/products/products_m3020.htm

http://www.lorlin.com/AboutLorlin.htm

http://www.isea.rwth-aachen.de/Ww/english/texte/tebe.html

http://www.lemsys.com/FicheTP0620-2.html

etc. for testers

 

[fernandoelec]

Thanks to dydt and jbartos for your help, but I have a 1336 plus Allen Bradley AC drive, the dc bus fuse open and it trip by low DC bus level, it working with fuji IGBT, most exactly with the refference 2MBI300N-120, I want to test the IGBTs to determiente if any of them are failed, i tried to find a technique to do it but it so hard. If yuo could tell me the values for those devices i'll appreciate so much. It´s mains terminal are identified by E and C and the gates By G, and it comes with two device package.

Thanks again for your help and prompt answer.

Fernando Sanchez

 

[electricpete]

If you have a fuse blown and you suspect IGBT, then I think you want to check for either IGBT shorted.... either conducting in reverse direction or conducting in forward direction when it shouldn't (no gating applied).

I would think that you would want to examine the resistance between E and C (no gate signal applied). It should be open in both directions, except for possible small leakage current.

How to test for open condition and not get fooled by the normal leakage current? I'm no electronics expert. But I would say that you should apply two different voltages and look for no substantial change in current. i.e. current at 2 volts is approx the same as current at 1 volt. This should apply in both directions between E and C.... no gate signal applied.... intended only to detect a short circuit.

Or else just compare the resistance on your meter to that of a known-good IGBT. Other comments?

 

[cbarn24050]

Hi, my advice is to stop messing with it, you will not be able to fix it. Even if you replace the igbt it is likely to pop as soon as you power it up.

 

[tmahan]

With a blown DC bus fuse in addition to IGBT's the front end is also suspect (the rectificer). An open DC bus fuse is usually the result of near catstrophic failure. Unless this is a very large drive, as CBarn24050 suggest, It probably makes more sense to put the effort being spent trying to repair the current product into finding out what caused the over current situation and just replace the drive.... Please note though there is one senario in particular that should be investigated... If Dynamic breaking resistors are being used with drive have they been correctly hooked up? If not the IGBT's may be ok but the rectifier is almost certainly blown.

 

[jbartos]

Suggestion: To troubleshoot IGBT circuits, it is most helpful to have test points identified with some values of current or voltage stated.

 

[fernandoelec]

Hi friends, thanks a lot for your help (dydt,jbartos,electricpete,cbarn24050,tmahan), i found a test and measure procedure of IGBTS FUJI in the follow address:

http://www.fujisemiconductor.com/pdf/app_notes/reh982.pdf

, in the appnotes section, it correspond to an IGBT instruction manual where it describe the installation procedure and other things, (the test and measure included in the chapter 4, section 4.2) for the series N IGBTS that correspond to my refference devices. I'll to test to determinate which of them are failed.

With respect to the fail is the second time that an inverter fail with that application, it is a pumping system of slurry, i suspect of the cable distance between the inverter and the motor,it is more than 100 ft, 400 ft aprox.
I think that the possible cuase is wave reflected, I'll check the braking resistors and IGBTs fails, and comment to yours the results. Maybe i Have to install an AC reactor line to avoid the problem. Once again thanks for your help.

Best Regards,

Fernando Sánchez

 

[shenyee]

Hello to everybody here! I would like to ask is there anybody here have read about Don Lancaster's "Magic Sinewaves" before? I'm doing some thing about induction motor speed control that may have related to his idea.
Could you please give me a hand. I do appreciate that!

 

[fernandoelec]

shenyee,

What do you exactly need?

I don´t have read yet "Magic Sinewaves" , but tell me your specific question.

Fernando Sánchez

 

[jbartos]

Suggestion: Visit

http://www.tinaja.com/glib/msinprop.pdf

for Magic Sinewaves
Question:
"...induction motor speed control that may have related to his idea." Which idea in particular?

 

[steve1701d]

Allen-Bradley has a "Bulletin 1336 Adjustable Frequency AC Drive Troubleshooting Guide and Renewal Parts" manual that may be helpful. This has some trouble-shooting hints, including how to test their IGBT's. It also has simple block diagram schematic for the 3-30HP and 5-52HP models, and parts lists. I think it may be downloadable on their website.

Reflected waves can also cause high voltage spikes at the motor end, and can break-down the motor wire insulation, especially if the motor isn't wound with spike-resistant wire. With 400 feet of wire, a load-side reactor should certainly be considered. Mount this as close to the drive as is reasonable.

Other causes could be line-side voltage problems, especially single-phasing problems if the drive is big enough. The filter caps on some bigger drives don't smooth out the ripple you get with single-phasing conditions, and if it is trying to run near full load, you get a lot of excess heating on the lower-end.

If you do find a bad IGBT module, and you still want to repair the drive, I suggest replacing all three of the modules, and checking out the caps and bridge as well. I'm just a service tech, but past experience has shown me that you get better results changing all the modules.