EMI QUESTION

[Tinkerer]

Attached is a screen shot showing:

TX Voltage pulse
RX signal
TX current pulse.

On the TX current pulse we can see, at switch on and at switch off, large high frequency noise, caused by the Mosfet switching.

At switch off, the noise frequency is at about 1.2mHz. I would like to reduce the amplitude of the noise because of the EMI emissions.

Would a different type of Mosfet produce a different frequency or lower amplitude noise?

How can I reduce the noise?

Is it possible to control the noise frequency?

Any suggestions?

[homefire]

Looks like the RX is Ringing sorta bad?

Damping change?

[Tinkerer]

Looks like the RX is Ringing sorta bad?

Damping change?

thanks for the feedback.

I need to explain a bit better.

The horizontal setting is 5uS/division.
the vertical setting is 50V/div for the TX pulse and 500mV for the RX signal.

The following picture is the TX voltage pulse and the RX signal only.

The TX pulse has a duration of about 14uS at a peak voltage of about 280V. There is some ringing visible for a duration of about 5uS, after the end of the TX pulse, this is the switching noise of the Mosfet that I want to reduce

The RX signal passes the 0V line, about 15uS after the end of the TX pulse. Then it goes slightly above 0V, to return to 0V about 16uS later. This could be called a slight ringing.

Tinkerer

[taliesin]

hi,
could i ask where is the probe attached on the mosfet noise trace?

[Tinkerer]

hi,
could i ask where is the probe attached on the mosfet noise trace?

Thanks for the response.

For the picture, the probe is attached to the drain of the Mosfet (IRFP460), but, with the voltage at 280, the probe can pick up the signal through the air, capacitively, and it looks the same.

On the current trace the switching noise is much more visible. There you can see, that the switching noise is different for the switching ON transient than from the switching OFF transient.

At the switching moment the current is about 6 Amps.

Tinkerer

[taliesin]

i had the same problem when designing speed contoller for rc car.at 1.2 mhz looks like the capacitance of the mosfet is culprit.in speed controlers often lots of mosfets are used in series as the capacitance is reduced.interesting tho as seems like lots of power is wasted there and as you say EMI .will think more on this.

[Carl-NC]

I'd need to see your circuit, but it could be that the gate drive looks a bit inductive and is reacting with the gate capacitance. Try putting in some series gate resistance.

- Carl

[Tinkerer]

I'd need to see your circuit, but it could be that the gate drive looks a bit inductive and is reacting with the gate capacitance. Try putting in some series gate resistance.

- Carl

Carl, thanks for the feedback.

I had 10 Ohm series gate resistance. Changed it to 100 Ohm and the noise got much lower. The real culprit seems to be the 8" wire between the Mosfet driver that is on the PSU board and the TX board that produces too much stray inductance, just as you say.
I will remember that when I draw the final PCB.

Tinkerer