diyAudio (
-   Class D (
-   -   Cross Conduction (

lumanauw 7th January 2008 04:06 PM

Cross Conduction
2 Attachment(s)
I SIM this simple classD output stage. It's like Behringer output stage, with turnoff transistors.
The driver is idealized Vpulse, +/-10V p-p, rise/fall time=10ns, 2uS period and 0.5uS on-time (to reflect working under load). The load is 1.5ohm.

lumanauw 7th January 2008 04:10 PM

2 Attachment(s)
"Vss#branch" is Vss current.
It shows severe cross conduction between 0->15A (lower mosfet) and -10->-25A (upper mosfet).

What/who makes this cross conduction? Is it caused by dead time?

BWRX 7th January 2008 04:47 PM

It looks like the reverse recovery transient and it corresponds with the off time of about 100ns plus the reverse recovery time of 100ns in the IRF530 datasheet.

lumanauw 7th January 2008 05:01 PM


Thanks for the input :D Is there any trick to eliminate this cross conduction? I tried with IRF540-9540 and they are worse.

Why this cross conduction don't show up if I make on-time 1uS (50% of 2uS), and become worse with smaller dead time?

lumanauw 7th January 2008 05:21 PM

2 Attachment(s)
I'm really curious and make this SIM. Upper mosfet is driven by ideal square wave V1, off=35V, on=45V.
Lower mosfet is driven by V2, off=-45V, on=-35V.
Both T=2uS, Ton=0.5uS, rise/fall time=10ns.

lumanauw 7th January 2008 05:24 PM

2 Attachment(s)
and I still get this :
There are 2 cross conductions now.

How can we advoid this "mosfet inherent recovery time" problem?

BWRX 7th January 2008 05:48 PM

Adding a snubber at the switching node may help some.

BWRX 7th January 2008 07:22 PM

Try adding a 200ohm resistor in series with D2 and an RC snubber (R=10ohms, C=0.22uF) between VDD and SW. That helps get rid of some of the cross conduction but not all of it. I can post the switchercad files of the simulation I cobbled together in a little bit.

Workhorse 8th January 2008 01:35 AM

Untamed Goliath posing problems
Hi David,

R1,R2 should be atleast 1K or even less to speedup turn-off and insert the Rg as said by BWRX to slow turn-On time.
Trr=100nS is enough fast if the SWF=300KHZ even, in Class-D operation


lumanauw 8th January 2008 03:56 AM

There is a good paper on body dioda Qrr, IR AN-967.
It's really a problem in high speed switching. This cross conduction won't show up in idle, but appears in modulated/loaded condition.

I wonder how Fumac get away with this in Mhz modulation, this Trr will be very big compared to the Mhz timing period.

BWRX, Kanwar, I will try to sim your suggestions, thanks.

All times are GMT. The time now is 08:01 AM.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright 1999-2017 diyAudio

Content Relevant URLs by vBSEO 3.3.2