new MosFETs on the block... - Page 4 - diyAudio
Go Back   Home > Forums > Amplifiers > Class D

Class D Switching Power Amplifiers and Power D/A conversion

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 22nd December 2011, 02:24 AM   #31
Eva is offline Eva  Spain
diyAudio Member
 
Eva's Avatar
 
Join Date: Oct 2003
Location: Near the sea
Send a message via MSN to Eva
Sch3mat1c: You are extrapolating data from another type of semiconductors and circuits.

The substantial forward voltage spike already seen on 600V ultrafast diodes when starting conduction is almost gone in body diodes of low voltage MOSFET. Old IR datasheets just said "intrinsic body diode turn on time is negligible" instead of giving a forward recovery time rating.

The IR series of MOSFET we talk about have a Vds rating of 250V or less, and are intended for synchronous rectification among other things. The dies are designed to get both low Rds-on and a decent ultrafast diode, and the cell structure of the MOSFET (like many transistors in parallel sharing a PTC resistive common collector connection) stores more than an order of magnitude less charge than the devices you are used to. Only a too high di/dt and dv/dt and Id combination can make the diode fail, and the failure mode is latching due to parasitic thyristor, not "hot spot" style failure.

In fact, maximum practical MOSFET (forward) turn-off di/dt is not much higher than the recovery di/dt allowed by the diode, 2:1 or less. Industrial switching in my opinion will take years to reach that degree of refinement.
__________________
I use to feel like the small child in The Emperor's New Clothes tale

Last edited by Eva; 22nd December 2011 at 02:31 AM.
  Reply With Quote
Old 22nd December 2011, 07:47 PM   #32
diyAudio Member
 
Join Date: Dec 2003
Location: Munich
Quote:
Originally Posted by Eva View Post
Concerning turn off, I meant reducing loop inductance, not the resistance.
Yup, thinking the same.


Quote:
Originally Posted by Eva View Post
With "light" I meant too high resistance, capacitance is probably ok.
Just looked to the bread board again. It is a parallel connection of 5x
47 Ohms. Ok, it's not 12 Ohms, but 9.4 Ohms. ...for some reason I had in mind that I found the optimum around 12 Ohms.
Anyway, I will have to readjust the snubbers according the PCB in any case.
  Reply With Quote
Old 22nd December 2011, 09:08 PM   #33
diyAudio Member
 
Join Date: Dec 2003
Location: Munich
Worth reading:
http://www.google.com/url?sa=t&rct=j...j2k2NA&cad=rja

I am still convinced that the body diode in my case is acting more or less in soft recovery and the first step in Uds is really an inductive phenomenon.
Coming from a di/dt of slightly above 1000A/us, this 30V step would correspond with loop inductance of less 30nH. These 30nH would reflect the source lead inductance of the upper Mosfet + drain lead inductance of upper MosFet + inductance of upper ceramic cap + inductance of lower ceramic cap + inductance of board connections.
I doubt that this sum is much less than 20nH-30nH and consequently Kirchhoff leaves not much space for a high voltage peak across the junction in my set up.
Up to now I did not manage to kill the IRFB4115. Also all measured values are within allowed limits of the data sheet.

But previously I examined limitations of IRFB4615 at 400kHz and +/-60V.
First I operated with moderate di/dt and got failures already short above 20A load current.
The higher I adjusted di/dt the higher the limits moved and I could shift the failure limit to approx 30A.
I think in case of IRFB4615 I always had the body diode flooded, before
turning ON the other switch- just because of the high Rdson, which cannot bypass very high currents from the diode.
To me all findings indicate that we can assume soft recovery of the body diodes in these types under all conditions, that I tested.
Well - even 'soft recovery' still involves giant di/dt at the end of reverse recovery that triggers resonances of the circuit, if not snubbered properly.
  Reply With Quote
Old 23rd December 2011, 03:14 AM   #34
Eva is offline Eva  Spain
diyAudio Member
 
Eva's Avatar
 
Join Date: Oct 2003
Location: Near the sea
Send a message via MSN to Eva
Aren't your ceramic capacitors from positive rail (upper source) to negative rail (lower drain)? This may result in a slight improvement in inductance (if mated with layout).
__________________
I use to feel like the small child in The Emperor's New Clothes tale
  Reply With Quote
Old 23rd December 2011, 09:32 PM   #35
diyAudio Member
 
Join Date: Dec 2003
Location: Munich
Quote:
Originally Posted by Eva View Post
Aren't your ceramic capacitors from positive rail (upper source) to negative rail (lower drain)? This may result in a slight improvement in inductance (if mated with layout).
Usually I am placing both MosFets side aside, in this case the distance from upper drain to lower source is given by the MosFet geometry and is longer than two SMD caps anyway. I am using multiple parallel strings of two series caps. Center tap has a less relevant connection to GND.
  Reply With Quote
Old 23rd December 2011, 10:55 PM   #36
diyAudio Member
 
Join Date: Dec 2003
Location: Munich
In fact in this particular set up I placed both MosFets with a certain distance inbetween, because I also experimented with the old source inductance di/dt limiter. For these MosFets I came to the conclusion - the faster the better.
Final gate drive wiring was connected directly to the MosFet legs and
connections of the power loop with flat copper plates.
Final PCB will have the MosFets close together....
  Reply With Quote
Old 22nd January 2012, 05:18 PM   #37
ungie is offline ungie  Canada
diyAudio Member
 
Join Date: Dec 2002
Location: Ontario, Canada
Quote:
Originally Posted by ChocoHolic View Post
Anyway, I will have to readjust the snubbers according the PCB in any case.
What should one generally be looking for on Vds waveforms in order to properly optimize these snubbers?
  Reply With Quote
Old 22nd October 2012, 07:31 AM   #38
stewin is offline stewin  Kenya
diyAudio Member
 
Join Date: Jul 2007
hi guys the conclusion guys is ?
what about irfb4229 it is rated at 250vlts and 91amps?
  Reply With Quote
Old 17th November 2012, 03:41 PM   #39
diyAudio Member
 
Workhorse's Avatar
 
Join Date: Dec 2004
Default This one looks promising

http://ixapps.ixys.com/DataSheet/DS9...-69N30P%29.pdf

300V, 69A, Trr=100ns, Qrm=500nC, Rds=0.049ohm
  Reply With Quote
Old 7th October 2014, 04:23 PM   #40
diyAudio Member
 
Join Date: Apr 2010
just found a new one:

IPP120N20NFD, similar to IPP110N20N3
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Hello, new guy on the block DBH Introductions 2 11th June 2010 06:07 PM
Hi from a new Kid on the Block. Valvepower Introductions 1 26th January 2009 06:00 PM
New driver on the block... Septimus Subwoofers 3 22nd December 2006 09:22 PM
n/p Mosfets vs depletion mod mosfets marcello7x Car Audio 6 4th October 2006 06:01 AM
New kid on the block Jrisk Introductions 3 26th February 2006 08:56 PM


New To Site? Need Help?

All times are GMT. The time now is 06:54 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2