Re: Disabling the freewheling diode

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Jim Lux" <jimlux-at-earthlink-dot-net> 

Power and current limits are somewhat different for most semiconductor
devices.  As you say, there's a bulk ohmic heating issue, for which duty
cycle is important (but check the datasheet.. small dice have short time
constants, so a 1 second on, 4 seconds off duty cycle isn't going to really
allow you to use 20%).  There's also an instantaneous current limit, usually
determined by the internal bond wires acting like fuses (tiny 1 mil wide
ribbon doesn't have much thermal mass and greatly resembles a fuse).  In
devices that are actually made up of many devices in parallel internally
(i.e. HEXFETs, among others), there's also a "propagation time across the
die" issue to make sure that all the devices have had time to turn on.  Very
high di/dt can result in all the current going through just a few of the
devices, cooking them, even though the overall power is within limits.  For
devices that don't turn on instantaneously (i.e. anything other than a spark
gap or Krytron type device), there's also the power dissipation during turn
on/turn off (which for most FETs is actually the dominant source of power
dissipation), which is highly, highly drive dependent.

Gotta read the datasheets is what it boils down to...
For diodes, there are three ratings: continuous (i.e. average), repetitive
peak (i.e. on every half cycle into a capacitive input filter),
non-repetitive peak (i.e. initial power supply turn on), and they'll have
times/frequencies associated with them.
----- Original Message -----
From: "Tesla list" <tesla-at-pupman-dot-com>
To: <tesla-at-pupman-dot-com>
Sent: Saturday, August 30, 2003 11:42 AM
Subject: Re: Disabling the freewheling diode


 > Original poster: "Gavin Dingley" <gdingley-at-ukf-dot-net>
 >
 > Hi,
 > this brings up a question that I have had in the back of my mind for some
 > time. If a MOSFET has a maximum current rating of say 10A, then this is,
as
 > I have read and understood it, is the maximum rating regardless of the
 > currents duty cycle. So if I switch the MOSFET on and off (across
 > gate-source) with a signal that has a duty cycle of 33%, and the peak
 > current amplitude of the switched current (through drain to source) is
10A,
 > the device will blow even though the average of the main current passing
 > through drain to source is only 3A. I find this strange as the
drain-source
 > is ohmic. Is all this correct, or can you treat the drain-source as a
purely
 > ohmic resistance with regard to destructive heating. Perhaps there is
 > another factor involved that makes a MOSFET blow due to peak amplitudes of
 > current passing through them.
 >
 > thanks in advance,
 >
 > Gavin
 >
 > ----- Original Message -----
 > From: "Tesla list" <tesla-at-pupman-dot-com>
 > To: <tesla-at-pupman-dot-com>
 > Sent: Friday, August 29, 2003 6:54 PM
 > Subject: RE: Disabling the freewheling diode
 >
 >
 >  > Original poster: "Jan Wagner" <jwagner-at-cc.hut.fi>
 >  >
 >  > Hi!
 >  >
 >  >  > In a SSTC H-bridge, the internal freewheeling diode of the
 >  >  > MOSFET is disabled by a clever arrangement of external
 >  >  > diodes. A schottkey diode above the drain keeps the internal
 >  >  > diode from getting forward biased, so that an ultrafast
 >  >  > recovery external diode can provide the path of any reverse
 >  >  > EMF from the load. Any current passing trough the MOSFET must
 >  >  > pass trough the schottkey diode.
 >  >  >
 >  >  > In a typical design, I see MOSFETS capable of handling 20A
 >  >  > continous, with a 3 A schottkey diode on top.
 >  >
 >  > (the diode can be at the source or drain pin, symmetric...)
 >  >
 >  >  > Why is the schottkey diode not dimentioned for a similar 20A?
 >  >
 >  > Because luckily it conducts only for a short period. If the SSTC is
 > in-tune
 >  > and the drive signals have ~50% duty, the diode ideally
 >  > conducts no current at all. When the SSTC is really grossly out of tune
 >  > i.e. pri current draw and the drive voltage are highly out
 >  > of phase, almost no current will be flowing into the primary, and also
the
 >  > diode conducts very little current. When slightly out of
 >  > tune, the diode will conduct the (sine shaped) current close to where
the
 >  > sine waveform crosses zero, so, again, it's not very hard
 >  > on the diodes as the average current is quite low.
 >  >
 >  > Only when you're doing PWM, i.e. changing the duty cycle of the drive
 >  > signals so that the duty t.ex. averages at 25%, the diodes
 >  > will conduct a large current - in the worst case, the peak primary
current
 >  > - and have to be rated the same as the mosfets (or, hmm,
 >  > probably only half the current rating of mosfets... (?))
 >  >
 >  > cheers,
 >  >   - Jan
 >  >
 >  > --
 >  > ********************************************************
 >  >   high voltage at http://www.hut.fi/~jwagner/tesla/
 >  >   Jan OH2GHR - GSM +358-41-4682893
 >  >   jwagner-at-cc.hut.fi - Jan.Wagner-at-cern.ch
 >  >
 >  >
 >
 >