Overshoot/Ringing

[jbec]

Hi,
I have designed and tested a class d amp. I have problems with ringing and spikes, I don't think it is overshoot because I have dead-time. I am designing for an H bridge but I tested with a half-bridge first. The output of the MOSFETs are shown in the image. Same output with the driver outputs

Switching frequency is 30 KHz. Using ir2110 with irfb4019. Bootstrap capacitor as diode are 0.1uF and UF4007.
The driver is powered with 5v and 12 v and MOSFETs with 15v. Can someone please help me out?

 

[KlausST]

Hi,

If you really need good assistance, then it's urgent to provide
* schematic and
* PCB layout with exact locations where you connected the scope (both: GND and signal)

I don't think it is overshoot because I have dead-time.

With dead time you prevent from through shoot / cross conduction.

My first idea is, that most of the overshoot/ringing comes from measurement errors.
But for the multiple ringing pieces during Low phase I have no explanation for now. (Missing informations)

Same output with the driver outputs

Don't know what this means.

In case you want good audio 30kHz is a bit low frequency.

Connect a suitable load to get better measurement results (less than 100 Ohms)

Klaus

 

[Easy peasy]

you need more than 0.1Uf ON THAT HIGH SIDE DRIVER ....!

 

[mrinalmani]

1. Add ceramic decoupling capacitors close to the bridge
2. If the FETs still overshoot, add an RC snubber across each drain-source. A reasonable value to start with would be 1 Ohm in series with 4.7nF.

 

[danadakk]

MOSFET Switching considerations -

https://toshiba.semicon-storage.com/info/docget.jsp?did=59460 Gate drive R calaculations

https://www.infineon.com/dgdl/Infineon-EiceDRIVER-Gate_resistor_for_power_devices-ApplicationNotes-v01_00-EN.pdf?fileId=5546d462518ffd8501523ee694b74f18

Regards, Dana.

 

[jbec]

Hi, thank you for your suggestions. I found out what wa the problem for the overshooting/ringing. It was the deadtime generated from the RC Xor gate network. I changed the circuit to a resistor diode capacitor circuit and it helped a lot. I'm having some more problems, the low side output waveform is measuring a peak to peak of 18v and not 12v. Can anyone assist me with this problem.

 

[danadakk]

The pk-pk includes the undershoot peak you see in your waveform.

Do you have a picture of prototype you are working with, and a schematic ?


Also scope probing methods quite important to getting accurate signal envelopes.

https://download.tek.com/document/02_ABCs-of-Probes-Primer.pdf

Best Practices for High Bandwidth Probing

This webinar highlights best practices and considerations for high bandwidth probing. Innovative probing techniques for various applications including high temperature, memory and high speed serial

www.tek.com

Also capacitor technology used for bypassing very important, eg. ESR vs Frequency

Regards, Dana.

 

[FvM]

As long as the measured gate voltage waveforms isn't exceeding transistor maximum ratings, you don' even need to determine if it's real or a probing artefact. I guess the latter.