-1
\$\begingroup\$

I have made a PWM controlled Power Mosfet switching circuit as attached. Power Mosfet switching circuit I am switching @ 100 kHz with rise time of ~15ns (my system reuires high speed switching. Hence I have kept gate drive resistor value low). There is >20% ringing in the output pulse. Is there any method to reduce ringing without compromising on the rise/fall time?

\$\endgroup\$
6
  • 1
    \$\begingroup\$ Are you seeing this rining with, or without, your "LC load" connected? And are you really driging the lower IGBT with a "180 deg Phase shifted PWM pulse" or is it just an inverted logic PWM pulse? \$\endgroup\$
    – brhans
    Commented Mar 23, 2017 at 13:55
  • \$\begingroup\$ Actually, I am applying a 20 ns pulse is applied to lower MOSFET to discharge gate capacitance. \$\endgroup\$
    – Jeevan
    Commented Mar 23, 2017 at 14:33
  • \$\begingroup\$ Actually, I am applying a 20 ns pulse to lower MOSFET after 40ns deadtime(40ns after upper mosfet turns off) to discharge gate capacitance. \$\endgroup\$
    – Jeevan
    Commented Mar 23, 2017 at 14:40
  • \$\begingroup\$ Are these MOSFETs or IGBTs? Layout will have a big effect on the ringing and body diode reverse recovery can excite the parasitic L-C as well. If diode recovery turns out to be the issue and you can stand the extra drop you could add a Schottky in series with the drain, and a high voltage ultrafast soft recovery diode from the source of the device to the cathode of the Schottky. \$\endgroup\$
    – John D
    Commented Mar 23, 2017 at 15:39
  • \$\begingroup\$ I have reduced the gate resistor in sink path of lower mosfet. This has reduced reduced ringing by some extent. So I think this may be due to the inbuilt anti parallel diode reverse recovery time. Can I just put a schottky diode across the drain -source terminals of mosfet? What should be the voltage rating of such a diode? My Vdd is 1.5K. \$\endgroup\$
    – Jeevan
    Commented Mar 26, 2017 at 9:10

2 Answers 2

1
\$\begingroup\$

How are you measuring the ringing - is it really there?

A common problem is incorrect grounding of the scope probe? This can result in ringing on a scope that doesn't really exist.

If it does really exist then you have excessive inductance or inadequate decoupling.

What is the frequency of the ringing? Is it the same on rising and falling edges?

Can you show the scope trace and the actual circuit including decoupling capacitors?

Edit

Ringing is going to be caused by excessive inductance somewhere in the circuit. If it is on the rising edges it could be in the grounded side of the upper IBJT. It is important that the loop from where you connect the scope to the active device ground is as short as possible.

\$\endgroup\$
1
  • \$\begingroup\$ It does really exist. I have increased the gate drive resistor value and confirmed that it reduces drastically. I have added a bypass cap (10pF) at output. But it has not improved anything. Where should i add decoupling cap and what value? Freuency of ringing is ~10MHz. i am not observing ringing on falling edge. \$\endgroup\$
    – Jeevan
    Commented Mar 23, 2017 at 14:36
0
\$\begingroup\$

The scope ringing also reduces when you increase the time of the rise of the measured signal by increasing the value of the gate resistor. Also the schematic shows only the ideal situation not the stray components of real system. To really investigate the problem we are not seeing the pcb layout & measurement setup (photos of both, and manufacturer/type of the devices & probes used and how they are connected).

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.