[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [TCML] IGBT or MOSFET in series



Hi Scott,

Yes, your right. If these series active components are simply connected in series, the probability of failure is 99%. Both static voltage balancing across the stack as well as dynamic balancing regarding voltage rise is required. There are a few ideas out there how to do that. For example, RC snubber across C to E, gate charge feedback, etc. But, it's not impossible. The SISG is very different from both of the listed topologies and it works far better in my mind. Think about it, I and others are running 24 IGBT's in series using Terry's SISG circuit, 15,000 volt DC bus.

Finn has also developed a trigger scheme by inserting an SCR to bypass the majority of sidacs in each gate using an inductive trigger voltage. The sidacs standoff the voltage bus which is kept a little lower than the full string would need to fire. The trigger coils pulse and cause the upper sidac in each IGBT circuit to conduct. When the SCRs fire, the voltage through the top sidacs and scrs force the TVS diodes to throw 25V at the IGBT gates (resistive divider is involved) where they all turn on at nearly the same time. A natural benefit with the circuit is that if the bus voltage was for some reason too high (prior to the trigger), the normal sidac string acts as a safety valve or safety gap ensuring the caps do not overvolt, etc. In otherwords, the sidacs in the full string would simply act normally and fire as in a standard SISG fashion for that event.

The SISG is the best means I am aware of at putting IGBTs in series minimizing imbalance issues. The circuit is simple, eloquent, and effective.

Take care,
Bart

BunnyKiller wrote:
Hey Matthew,

theoretically yes... but getting a long string of IGBT's/Mosfets to fire ALL at the SAME time is a near impossibility. Even a few miliseconds off in the firing sequence and you will cause an avalanch destructive "wave" to surge thru the string, destroying each succesive unit next in line. I have seen up to 6 1200V IGBTs in series, but the controller for the unit was very complex. ( way beyond my capacity to reverse engineer it).

Lets see what the real Mosfet gurus have to say.... there has been alot of improvements made in SSTC in the last year...

Scot D


Matthew Boddicker wrote:

I have a particular high voltage application for a solid state Tesla coil design. If I put IGBTs or MOSFETs in series, will their voltage ratings stack? For example, if I put ten 1,200Vces IGBTs in series, will the string be able to handle 12,000volts?



Take care,

Matthew

_________________________________________________________________
Windows Live™: Discover 10 secrets about the new Windows Live. http://windowslive.com/connect/post/jamiethomson.spaces.live.com-Blog-cns!550F681DAD532637!7540.entry?ocid=TXT_TAGLM_WL_t2_ugc_post_022009_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla



_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla

_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla