0
\$\begingroup\$

I have a MC2100ELS-18W motor controller and after inspection found that the IGBT is shorted, the part number is IRGPC40K

I suppose this is a discontinued part and I do not want to buy from China and want to find a replacement. I have no experience when it comes to IGBT and I am not sure about all the important specs to look for. I have been able to specify a replacement and I want some advice if it is suitable or not AOK30B60D1 and AOK40B65H2AL

I was looking for the following: Vce, Vge, Ic, Pd, Input Capacitance, Gate Charge, Turn-On DelayTime, Turn-Off Delay Time

I am really not sure if I should look for something else or not and whether if those specs should be exactly the same or not, some of them are a little bit higher like power ratings and voltages and other are lower like Charges, Capacitance and Delay times

One thing that I am not sure if it will change anything or not that the old part has no diode integrated and the new replacement has integrated one and I suppose that diode specs should also be respected !

This Controller is using IR2127 IGBT Driver and the motor is rated for 130V DC and 2.75KW continues Duty

My Question is is there a problem If some specs are more or less the original part specs? For example, the new replacement (Input Capacitance, Gate Charge, Turn-On DelayTime, Turn-Off Delay Time) are lower than the original part, would that make a problem?

\$\endgroup\$
3
  • \$\begingroup\$ Probably the most important thing is what you didn't ask -- make sure that the IGBT didn't blow because of some other fault in the circuit. Usually the power electronics dying is the most obvious symptom of a fault, but the real fault is elsewhere. So check motor wiring, and everything else in the circuit before you declare yourself done. \$\endgroup\$
    – TimWescott
    Commented Nov 1, 2020 at 21:20
  • 1
    \$\begingroup\$ Yes you are right, but I have already did what you have mentioned so I didn't ask about it in my question, the circuit really does not have a lot of components, it is rather simple. thanks \$\endgroup\$ Commented Nov 1, 2020 at 21:33
  • 1
    \$\begingroup\$ Good! You never know on these internet forums how much the other guy knows -- so I had to point that out. \$\endgroup\$
    – TimWescott
    Commented Nov 1, 2020 at 23:35

1 Answer 1

2
\$\begingroup\$

The internal diode won't matter if the circuit has external diodes.

Never operate a IGBT at more than it's maximum gate-emitter voltage or collector-emitter voltage or it will blow. These two ratings are the most absolute and the easiest to understand.

Collector-emitter Saturation voltage is next important and determines the voltage across the IGBT when conducting which determines heating. Similar to source-drain on resistance for MOSFETs,

Make sure to compare saturation voltage at similar applied gate-emitter voltages. Note the gate-charge vs Vge curve is much better than the number listed in the table.

All specs, especially timings, are subject to the test conditions. So you can't directly compare them between datasheets unless the tests conditions are the same (and they won't be a lot of the time for timing). Test conditions are usually listed in the columns right next to the specification. I basically ignore any specification given in seconds. They're kind of useless.

Gate charge and capacitance determine the timings and are more important than timings given. Note these are also measured under different conditions but these tend to be a lot more comparable than timings. Between charge and capacitance, I ignore capacitance if charge is available as it tends to be more directly comparable.

Higher gate-emitter charge/capacitance leads to longer timings and can lead to overheating while the IGBT is busy switching due to spending too much time in the inefficient (i.e. high loss) region between fully conducting and fully blocking, so try and not stray too high above here.

But lower gate-emitter charge can cause switching to be too fast which is efficient (less heat) but can result in reflections and ringing which can result in voltages that can blow stuff if the design is really border line.

For capacitances other than gate-emitter, higher can allow parasitic currents to flow and cause latching or other problems so smaller tends to be better for these.

All things being equal, you cannot go too low with saturation voltage, too high with gate-emitter or collector-eitter breakdown voltage, or too low with charge/capacitances OTHER than gate-emitter.

\$\endgroup\$
5
  • \$\begingroup\$ So basically you are saying the more the specs are close to the original specs the better specially for charge and timing? \$\endgroup\$ Commented Nov 1, 2020 at 20:09
  • \$\begingroup\$ Also regarding the Vsat according to your comment the lower is the better? \$\endgroup\$ Commented Nov 1, 2020 at 20:09
  • 1
    \$\begingroup\$ Maybe not necessarily better across the board, but less unknowns and coin-toss for charge and timing.. Lower Vsat is better all other things equal. \$\endgroup\$
    – DKNguyen
    Commented Nov 1, 2020 at 20:10
  • \$\begingroup\$ If possible could you take a quick look on the suggested alternatives in my question and give me your quick opinion, I will keep searching according to what I have learned so far.thanks \$\endgroup\$ Commented Nov 1, 2020 at 20:13
  • 1
    \$\begingroup\$ Seems like either would work but AOK40B65H2AL is a bit safer due to AOK30B60D1 having much less gate charge than the original. \$\endgroup\$
    – DKNguyen
    Commented Nov 1, 2020 at 20:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.