Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Class D Switching Power Amplifiers and Power D/A conversion

Body diode bypass, elegant way!
Body diode bypass, elegant way!
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
Thread Tools Search this Thread
Old 2nd December 2010, 11:13 PM   #11
fredos is offline fredos  Canada
diyAudio Member
Join Date: Mar 2005
Location: Quebec
I lost some watts at idle, but got lot better effiency at (equivalent) 0% and 100% modulation, when load is connected to +/- Rail. I know that I short 1 turn to rail (less diode drop), but this seem to force freewhelling current trough the diode and not the body diode. A major advantage at high power. I drop supply to +/- 60V ( 2 turn tap) today and place a modulator (open loop) to check at different duty cycle. It seem that at all duty cycle, less than 5% of freewhelling and recovery current pass trought body diode. More I got close to 0% and 100%, less idle power is lost in the external diode. At idle (50%) I got about 1A flowing trough diode, nothing stressing, about 0.6W of loss (50% duty cycle). When I reach 30A of load current, I got about 15.5A flowing trough external diode (20W dissipation), nothing stressing too. Still have less than 1A trough body diode at all modulation. It's hard to compare with and without this trick, since body diode in the APT5010 are very slow ( 250ns at 25 celcius and 500ns at 100 celcius!). For sure at idle it's worst than any 200V mosfet, but at high volatge and high current it seem very better.

Wave are clean with no overshoot or rigging ( with snubber of course!), turn off behavior of both diode and mosfet seem good to me..
  Reply With Quote
Old 3rd December 2010, 03:37 PM   #12
ViennaTom is offline ViennaTom
diyAudio Member
Join Date: Apr 2010
I agree with darkfenriz - its a short circuit...
  Reply With Quote
Old 4th December 2010, 02:43 PM   #13
darkfenriz is offline darkfenriz  Poland
diyAudio Member
darkfenriz's Avatar
Join Date: Jun 2004
Location: Warsaw
By the way: has anyone tried to store the energy of switching-on event and body diode conduction and recuprate it afterwards?
Recently I came across the so-called "flyback reset ZCS lossless snubber" or something like that, which introduces a one-way inductor to soften the switch-on/body diode conduction period and in the other direction recuperates the energy back into rails in a flyback style.
Any experiences, ideas, thoughts??

  Reply With Quote
Old 6th December 2010, 09:19 AM   #14
phase_accurate is offline phase_accurate
diyAudio Member
Join Date: May 2002
Location: Switzerland
When I tried it in 1988 I also had the impression that it causes a short-circuit (FETs got very hot) but the switching signal looked impressively clean OTHO - although the circuit was built on breadboard.

The Peavey circuit uses a series resistor. I assume it is there to reduce the circulating current.



Last edited by phase_accurate; 6th December 2010 at 09:21 AM. Reason: typo
  Reply With Quote
Old 14th November 2012, 05:50 AM   #15
Workhorse is offline Workhorse
diyAudio Member
Workhorse's Avatar
Join Date: Dec 2004
Well this works, even i checked it.
  Reply With Quote
Old 16th November 2012, 03:27 AM   #16
Sch3mat1c is offline Sch3mat1c  United States
diyAudio Member
Sch3mat1c's Avatar
Join Date: Jan 2003
Location: Milwaukee, WI
Send a message via ICQ to Sch3mat1c Send a message via AIM to Sch3mat1c
You'll want that to be something like 1 turn out of 50 (depends on supply voltage, essentially Vf(diode) / Vsupply), which may be a steep ratio for most coils that don't need that much inductance. Too low a ratio and it doesn't do jack; too high and you short out some fraction of the transistors' saturation region.

Personally, I'd rather shunt the MOSFETs with good diodes. A purpose-made silicon junction diode has lower voltage drop and much greater speed than a MOSFET body diode, so it turns on sooner, carries more current, and both together recover faster as a result. Schottky is also acceptable at *most* audio voltages. ("Super schottkies" are available up to 300V, though they have voltage drops comparable to, or greater than, junction diodes, the advantage being zero recovery and thus higher speed. SiC schottkies have significantly greater voltage drop and internal resistance, and are only useful in high voltage, very high frequency switching applications where recovery cannot be tolerated.)

If you use a core with multiple return paths (E types, or even those four prong things if you can ever find them), you can use a flux unbalancing winding to allow fractional turns around a single limb. The method acts like a vernier on the windings.

Example: suppose you have 10 turns on a core to achieve the correct inductance. Obviously you can't make a 1/50th ratio the normal way. If you loop a separate winding around the two limbs, one turn each, so that the flux in each must be equal, then you can make half-turn windings successfully. But 1/2 out of 10 is only 1:20, and we need 1:50 in this example, so we need 1/5th of a turn. You can wind 4 turns on one limb and 1 turn on the other, to enforce a 4:1 split (1/5 and 4/5 of the total), but this reduces the A_e of the core by almost half (40% -- the 1-turn leg saturates much more easily). If you do a 2:3 split, A_e only drops by 20% (probably within design allowance). Now, if you pass a turn in the positive direction around the "3" leg and negative around the "2" leg, you get (3-2) / (3+2) = 1/5th of a turn, which can be connected in series with the main winding for a 10:0.2 ratio.

This is presented in more detail in an old Unitrode appnote, including winding techniques to minimize leakage inductance.

Seven Transistor Labs, LLC
Projects and Resources / Electronic Design and Consultation

Last edited by Sch3mat1c; 16th November 2012 at 03:31 AM.
  Reply With Quote
Old 16th November 2012, 06:58 AM   #17
Workhorse is offline Workhorse
diyAudio Member
Workhorse's Avatar
Join Date: Dec 2004
Here is another way
Attached Images
File Type: jpg body diode sollution.jpg (149.9 KB, 263 views)
  Reply With Quote
Old 1st February 2017, 07:35 PM   #18
stewin is offline stewin  Kenya
diyAudio Member
Join Date: Jul 2007
ucd 25 watts to 1200 watts using 2 mosfets

ucd 25 watts to 1200 watts using 2 mosfets

ucd 25 watts to 1200 watts using 2 mosfets

more ideas of body diode bypass
  Reply With Quote


Body diode bypass, elegant way!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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
A safe and elegant way for soft-start Bakmeel Power Supplies 5 15th September 2010 07:47 PM
Test for blown body-diode in mosfet? h_a Solid State 2 29th June 2009 07:47 AM
Disabling MOSFET body diode... RiskCord Class D 49 30th May 2008 10:13 AM

New To Site? Need Help?

All times are GMT. The time now is 01:36 AM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio