Re: OT: FET matching

Dave Casey
 

For synchronous turn-on, I would think matching gate capacitance and Vth
should get you pretty close. When turning on, you're spending a small but
still finite amount of time in the linear region before getting to
saturation. For one FET to take the same abuse as the other, you'd want
them to turn on at the same time (matched Cg and Vth) and have as similar
slopes as possible in the linear region on a curve tracer, especially at
higher drain currents.

Dave Casey

On Thu, Mar 23, 2017 at 6:55 PM, mosaicmerc@... [TekScopes] <
TekScopes@...> wrote:



This is probably related to curve tracing but which parameters should be
prioritized to match paralleled MOSFETs in a hard switching application?
I have a little tester that gives me Vth (eg. 4.37V) and Gate capacitance
(in nF) in a couple seconds.

It seems that the MOSFET that switches on first gets a bit beaten up by
the loading and can fail shorted after some months.
I am using IRFP3206 units in a bank of 4 driven by an MCP1407 FET driver
with 5Ω gate resistors and a ferrite bead. Switch currents (700hz, up to
100uS pulses) can hit >800A. The load is slightly inductive at around 5uH
which forces a shark tooth ramping pulse and some avalanche kickback.
Arresting the kickback is not an option, so thermal management is in place
for SOA maintenance.



Any advice please?


[Non-text portions of this message have been removed]



[Non-text portions of this message have been removed]

Join TekScopes@groups.io to automatically receive all group messages.