Dear Research Gate members,
I have question regarding IGBT blasting in traction converter.
What we have observed that IGBT Pair of ARM gets short circuited and IGBT blasted.!!
I would like to understand scenario that: What might would have happened at Inverter that causes Pair of ARM short circuited ?
Might it be a problem of High modulation index or something else ?
Any reference to relevant documentation is highly appreciated...
Hi Lalit G. Patil , check your power supply please. Make sure NOT to use a common ground to your IGBT driver circuit and controller, it is a different voltage value. And also check how much exactly you driver should supply to properly drive your IGBT. And دکتر افشین رشید
you are just spamming, writing "Nice question" is not an actual answer...this is not Facebook please.
Thank you Samuel Lakew for your answer. I would like to add that I am talking purely about Traction converter which have a DC Link of 2800v DC, Motor may Run at 1000Amps and Power may be 2.5MW or around...
Controller is totally separate and it has no common ground with IGBT driver circuit. Both are located at totally different location and IGBT Pulses are fetched with Optical fibre...
Yes, but i will check again for 2 points which you mentioned.
1. No Common Ground between IGBT Driver circuit and Controller
2. Requirement of Supply voltage to Drive IGBT
Dear Lalit G. Patil,
1st: if a pair of IGBTs is destroyed simultaneously, it is highly probable that they were erroneously both in the on-state at the same time. So in addition to the proposals of @Samuel Lakew you might check, if there might be any error in the interlocking of HS and LS-IGBT (or too low time delay between switching off one IGBT and switching on the other). 2nd: Another possibility is that the power supply for the HS drivers is not sufficient in some states. So this IGBT does not swith on properly, experiences too high losses and is thermally destroyed. If then the LS IGBT is swithed on, you have also a hot branch and the LS-IGBT is destroyed in consequence. We had this case one time, when the bootstrap power-supply of the HS driver wasn't active for too long time and thus the supply voltage for the HS-drivers dropped too low. But if this latter case is possible, you can check by considering how your IGBTs are protected against short circuit and how long they can endure one. Maybe this case would also be safely controlled ba your short circuit protection strategy. 3rd: Additionally the control signals might be corrupted on the controller side by high capacitive currents that occur, when the HS-IGBT switches and the IGBT-sided GND-potential of the HS-driver changes very fast. Such a corruption might also lead to an erroneous on-state of an IGBT and thus lead to a hot branch. But as you have optical signal transmission to your IGBT drivers, this reason is not very likely - unless the GND-potential on the controller side is influenced too much by such an effect because there are other capacitive coupling paths between HS Emitter-potential and controller-GND.
Common causes:
(1) Overcurrent: One is short-circuited by load, the other is The logic at the control circuit is disturbed, causing the upper and lower bridge arm components to pass through.
(2) Damage of insulation
(3) Overvoltage: Usually caused by the spike voltage generated by the line stray inductance under the action of extremely high di/dt. The solution is to design a high-performance absorption circuit to reduce the stray inductance of the line.
(4) Overheating: IGBT cannot be fully turned on, component loss in the presence of current
Increase, temperature increase leads to damage.
(5) Communication error rate a. After a period of communication, a sudden error message causes the IGBT to mislead and cause the IGBT to explode. b. The communication board FPGA program is unstable, causing the IGBT to mislead and cause the IGBT to explode.
The considered converter operates at high bus voltage and output current. In such conditions, series-parallel transistor assemblies are usually used.
http://www.divtecs.com/research/high-current-high-voltage-igbt-switch/
For such structures, the conditions for the occurrence and development of an accident are somewhat different from the conditions for single transistors due to the poor distribution of voltages and currents. I accept the causes of the accident for single transistors described earlier. However, I want to expand the set of reasons.
1. It is possible that the Miller effect occurs when, with a sharp increase in voltage across a closed transistor, it spontaneously opens. In this case, an unrestricted current flows through the transistor rack. Usually the transistor rack collapses. Eliminated by using Snubber circuits that limit the rate of voltage rise.
2. The parameters of the transistors in the converter are incorrectly calculated without taking into account the electrothermal operating modes. The traditional mistake is that it is not the temperature of the transistor crystal that is used, but the temperature of its body. Transistor manufacturers usually give the maximum die temperature in the documentation. As the temperature rises, the value of the maximum current of the transistor decreases. Overheating of the crystal also leads to thermal breakdown of the transistor.
The set of reasons can be developed further. We considered the possibility of thermal breakdown in the transducer racks in repeated short-term operating modes. Calculations of electrothermal transient modes of operation have confirmed the presence of thermal shocks on the crystals of transistors in starting modes of operation.
I hope that you will be able to understand the causes of equipment failure and ensure its reliable operation.