IRS2092S - ringing at gate drive pins

[Shahin]

Hi all,

This is my first time designing with the IRS2092S and I'm hoping I could get a little guidance. I'm using it along with IRFS4620 FETs. Below is a measurement made with a diff probe on the LO and COM pins.

As you can see there is 65 MHz ringing. Attempts to snub the ringing doesn't affect the ringing frequency which leads me to believe the source is somewhere else.

Feeding the gate of the FETs I have 22R in parallel with 22R + an MMBD1401 diode.

Has anyone else run into this?

 

[vdij]

Put the scheme.

 

[JonSnell Electronic]

A schematic would help but drivers do ring.

 

[nigelwright7557]

I wouldn't worry about it too much if the amp is working fine and not overheating.

You could try a snubber using RC tuned to 65MHz.
1/2 PI R C = F.

 

[tomchr]

Snubber design: http://www.cde.com/catalogs/igbtAPPguide.pdf

~Tom

 

[DUG]

Physical layout of driver, resistors and FET's for the gate drive path can have an effect of the waveform.

How about a nice clear picture?

 

[vdij]

Do not worry - Wire oscilloscope

 

[Shahin]

Here's the schematic and a screen grab from my layout. The amp is working fine. My only concern is that I'll run into EMI problems.

I was wondering if this was induced by the leads of my diff probe. I'll try floating my scope and using a single ended probe with shorter leads to see if things look any different.

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.

http://postimg.org/image/6vlxw8j1f/
http://postimg.org/image/vbfuj2sxh/
http://postimg.org/image/6wbrig9it/

 

[Shahin]

An externally hosted image should be here but it was not working when we last tested it.

screen shot on windows

An externally hosted image should be here but it was not working when we last tested it.

screenshot app

An externally hosted image should be here but it was not working when we last tested it.

program to take screenshots

 

[vdij]

http://www.irf.com/product-info/audio/classdtutorial2.pdf can help

 

[knutselsmurf]

you're probably fooled by your oscilloscope. If the rising and falling edge are the same I think this is the problem.

But to be sure. increase you gate resistor. you have to see a change. (22R is already a significant value)

hope this helps.

 

[DUG]

Lower or raise the gate resistors by a factor of 2 and see if there is a change.

Just noticed post # 11

🙂

 

[ionutzxpo]

Hy Shahin,in my IRS2092 amps, I use heavy snubering 680pF...1nF+27R/2W and I get allmost perfect PWM between the fets.I also use many 1uF/100V X7R across +->GND right next to fets (IRFB5615/IRFB5620) and GND-> - and right at the fets legs to GND 🙂 SWF 400Khz .Great Sound

 

[knutselsmurf]

Hy Shahin,in my IRS2092 amps, I use heavy snubering 680pF...1nF+27R/2W and I get allmost perfect PWM between the fets.I also use many 1uF/100V X7R across +->GND right next to fets (IRFB5615/IRFB5620) and GND-> - and right at the fets legs to GND 🙂 SWF 400Khz .Great Sound

For a halve bridge, snubbering shouldn't be needed. Control you're FET rise and fall time with the gate resistor network. Slow on, fast off this will solve the problem.
Play around with LTspice.

For the good order this is a different problem.
ringing on the gate is mainly caused by Cgs.

and ringing on the leg of the halve bridge is caused by Cds and the inductor.

In a bridge configuration lowing the V/dt solve both problems. except if the Cds is really large. Then use a small snubber. But better find better mosfets. trade Rdson, for a lower Cds.