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

Re: IGBT switching speed / resonant switching...



Original poster: "Steve Ward" <steve.ward@xxxxxxxxx>

Rajesh,

It basically comes down to the switching losses.  If your load is
resonant, and you know how the driver behaves you can get a good idea
of the switching phase angle, thus knowing the current being switched.
Look closely at the spec sheet for Eoff and Eon losses (you really
only care about Eon, since you will likely be switching *after* the
current zero crossing, or at least thats what i recommend!).  Eon is
typically in the low mJ range for even giant IGBTs, probably uJ in the
range you are working.  This is just multiplied by the switching
frequency, of course, to determine the average power dissipation.  You
should also try to factor in the conduction losses to get an idea of
the overall power dissipation so that you can suitably heatsink the
devices.

For operation up to 100khz, id probably look for IGBTs with delay
times in the <250nS range.

Steve

On 3/4/07, Tesla list <tesla@xxxxxxxxxx> wrote:
Original poster: "Rajesh Seenivasan" <rajeshkvs@xxxxxxxxxxx>

Dear forum members,

Can anyone tell me how to calculate at the maximum switching speed
for IGBT modules? I understand that the turn on and turn off time
limits the switching speed. and the switching frequency will improve
a lot by utilizing resonant switching or zero current switching.
Consider the following IGBT for example:
BSM75GB120DN2 (or any 1200V, 100A IGBT bricks from Powerex or any
other manufacturer)
Can they be used in a half bridge and switched at frequencies 75KHz
to 100KHz to generate square waves at 600V to 700V and current draw
between 30A to 50A (less than half of the continuous current
specification in datasheet) ?

Thanks and regards,
Raj.

_________________________________________________________________
Latest updates from the world of sports http://content.msn.co.in/Sports/Default