Low Side Mosfet Switch Circuit at 250kHz

Thread Starter

Brad55

Joined Mar 31, 2009
11
Hello,

Please see the my circuit and waveform. This was supposed to be a simple 250 kHz square wave circuit to apply about 22 Vpp across the load resistor. The square wave is from a TTL output of a function generator. Why am I getting the 43 Vpp spike 400 ns after the non-inductive 8 ohm resistor is turned off? I never expected this much of a spike with a resistor. Looks like an inductor! I measured the L and phase angle of the resistor at 250kHz on a Wayne Kerr 3260. The inductance was about 100 nH and the phase angle -0.04 degrees (or 0?). My wires are short and neat. The reason for the 15V regulator is to power the gate driver which cannot have more the 20V. Any ideas or suggestions?

Thanks in advance,
Brad
 

Attachments

Thread Starter

Brad55

Joined Mar 31, 2009
11
Crutschow thanks for your reply. I should have run the same thing on LTSpice. I will when I get to work tomorrow. Good point about 100nH at higher freq. But at 250kHz square wave shouldn't the PULSE directive be something like "PULSE(0 11 0 1n 1n 2u 4u 10)? Which if so would even make the transient even worse. If I have time tomorrow I will post my simulation results tomorrow with the IRFP260N. Just so you know the only reason I was using the 8 ohm non-inductive resistor was to prove the circuit to be "clean" square wave prior to putting in an transformer with 1.8 ohms impedance and a phase angle of about 30 degrees at 250kHz. Then I was expecting ringing issues. Thanks
 

Thread Starter

Brad55

Joined Mar 31, 2009
11
Yeap you're right and I realized that after I posted it(and went to bed)! I did run all this on LTSpice with my MOSFET and got the same results. So now I'm trying to clean this up with a MURF860 across the load and with caps on the 15V to the gate driver (.1 and 1.0 uf at the power pin and 47uf x50V at the VR.) Other than that do you have any suggestions?
BTW) Image 1 is of the circuit thus far with the free wheeling diode but without any caps.
Thanks
 

Attachments

crutschow

Joined Mar 14, 2008
34,285
Any rectifier will work to suppress the transient, since all rectifiers are fast turn-on, which is the important parameter for this function.
Fast rectifiers have a fast turn-off which will have little effect on suppressing the transient.
The best place to put the diode is directly from the MOSFET drain (anode) to the plus supply (supply properly decoupled of course) since that also suppresses any transient from wiring inductance..
 

Bordodynov

Joined May 20, 2015
3,177
upload_2017-9-14_14-47-30.png It is necessary to take into account the parasitic inductances of the wires. Weaken the harmful effect by using capacitors. There will be no voltage surges on the resistor.
 
Top