hf ringing after inductor on positive halfwave. - Page 2 - 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 8th October 2012, 12:27 PM   #11
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
yes 50Mhz starts to be ok but it can be made even better, I managed to get >90Mhz with IRFB4227 whose capacitance is not low.

Consider ceramic chip capacitors to create a HF supply loop as short as possible (two from + to - rather than in series, and closer to FET legs) and small electrolytics (again from + to -, lossy) to damp the other usual resonance (around a few Mhz) that appears between the ceramics and medium/big electrolytics (this time ceramic C + all parasitic L closing the loop through the electrolytic).
__________________
I use to feel like the small child in The Emperor's New Clothes tale

Last edited by Eva; 8th October 2012 at 12:33 PM.
  Reply With Quote
Old 8th October 2012, 03:12 PM   #12
diyAudio Member
 
Join Date: Oct 2010
Location: Canada
but eva if you look at the design i am trying to mount my heatsink on the bottom of the board directly to all large transistors that is kind of forcing me to make the leads much longer then normal... its appears to be the biggest down fall i am thinking of bending the leads of the fets right at the die (or package rather) but not sure if that would be such a good idea for the fets either (though if it wont effect the connection to the die it will save A LOT of lead length). any suggestions on how i can get ceramics (1206) near the -+ btw the fets. maybe sandwiching the fets? so it should go in this order from supply to fets : big electrolytics -g+, ceramic -g+, small electrolytics -+, ceramic -+? also i could use a metal film to do the -+ loop but it will add L so i don't think this will help frequencies this high?

btw thanks for the input to all so far!
  Reply With Quote
Old 8th October 2012, 03:31 PM   #13
diyAudio Member
 
Join Date: Oct 2010
Location: Canada
on another note, even with cascoding the levelshift circuit it seems almost impossible to completely match high and low gate fet gate rise/fall times with a discrete driver. i did try playing with RC delays a bunch in simulations but didn't seem to resolve the higher output in the upper gate driver. the fast swing of the totem forces the upper gate driver to turn on harder and faster. i was thinking of trying a discrete comparator/xor with a ir2010(smd version) for the gate drive since i have 20 laying around but fear the propagation delay will be to high. i can't seem to find models for pspice to model any of the IRF gate drivers.
  Reply With Quote
Old 10th October 2012, 03:41 AM   #14
diyAudio Member
 
Join Date: Oct 2010
Location: Canada
so i shortened the mosfet leads to the minimum the to-220's can accept (about 8mm reduction on each lead). the ringing amplitude and frequency remained without any change. i then tried placing metal film caps (.01uf - 1uf) directly to drain of upper and source on lower mosfets, no change. i then attached speaker wires to the board and placed my probe about 1ft away. ringing frequency is still the same 35mhz of course but the amplitude has gone down more then 10x.

kind of stumped on how i can possibly raise the frequency of the ringing .
Attached Images
File Type: jpg IMG_20121008_214719.jpg (277.1 KB, 105 views)
File Type: jpg IMG_20121008_214519.jpg (281.3 KB, 101 views)
File Type: jpg IMG_20121009_210144.jpg (244.4 KB, 99 views)
  Reply With Quote
Old 10th October 2012, 06:51 AM   #15
Pafi is offline Pafi  Hungary
diyAudio Member
 
Join Date: May 2004
Location: Budapest
Measure the ground reference point on the PCB! (Should have been a clean 0 V, but it won't be.)
  Reply With Quote
Old 10th October 2012, 10:48 AM   #16
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
The wide film capacitor does nothing good to the circuit, inductance is too high. The path length of the ceramics matters more. I use to put them directly from FET pad to GND plane, maybe two 100n stacked.

FET lead length, together with gate resistor, control the di/dt of the circuit. Too short leads should be avoided because di/dt may be too high and body diodes may latch at high current and temperature causing both FETs to short, apart from the increased EMI. Bending them where they become narrower is usually a good compromise. Then use the lowest gate drive voltage allowing FET "saturation" for your output current requirements.
__________________
I use to feel like the small child in The Emperor's New Clothes tale
  Reply With Quote
Old 10th October 2012, 11:14 AM   #17
diyAudio Member
 
darkfenriz's Avatar
 
Join Date: Jun 2004
Location: Warsaw
I imagine adding a via directly from between 100nF caps you have to opposite side ground plane can help a lot.
  Reply With Quote
Old 10th October 2012, 11:59 AM   #18
diyAudio Member
 
Join Date: Oct 2010
Location: Canada
i will try adding more 1206 100nf from fets to ground (my current ones also have via's to the ground plane right btw them). So I should put the other FETs back in hmm OK. Maybe its time to put the better ones in.
i took a quick measurement and its not terrible. 0.05%THD+N at about 10w driving a 4ohm speaker.

Last edited by switchleg; 10th October 2012 at 12:24 PM.
  Reply With Quote
Old 10th October 2012, 12:41 PM   #19
diyAudio Member
 
darkfenriz's Avatar
 
Join Date: Jun 2004
Location: Warsaw
Add more vias then. This can do more to loop inductance than the number of capacitors.
You definitely need better FETs, also IRF530N's can be quite different devices depending on manufacturer.
  Reply With Quote
Old 10th October 2012, 12:43 PM   #20
diyAudio Member
 
Join Date: Mar 2008
Switchleg,
this is how I make fairly noise-immune measurements. My scope probes all have BNC probe-tip adapters, and I have made a bunch of tightly twisted leads soldered to BNC sockets. I used Teflon insulated wire rated for 600V, so I dont melt it or arc across anything. And I used a bit of heatshrink & hot melt glue to glue the twisted pair onto the BNC socket - otherwise the wire eventually falls off.

I have a range of different length twisted-pair BNC leads. I do try to use short leads where possible, but it works well enough that I dont have to bother.

I also use a 10x 500 Ohm 1GHz probe (scope = 50R) when I measure across current sense resistors. before I got those, I used DIY 10:1 probes by hacking up a BNC lead (thin coax) and adding a 450R series resistor. that was a bit fragile, but very cheap.

Can you post the rest of your gatedrive schematic? I cant quite see how Q16 & Q18 work, so am obviously missing something.
Attached Images
File Type: jpg measurement_method.jpg (310.2 KB, 91 views)
  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
Tapped HF inductor: where to get? Stabby Parts 0 20th April 2011 11:38 PM
Ringing on class d output inductor. nigelwright7557 Class D 36 25th February 2011 07:49 PM
Separate DACs for each halfwave ? Bernhard Digital Source 94 20th December 2006 06:06 AM
Cure for HF oscillation on positive cycles only RichardJones Pass Labs 4 27th June 2002 12:20 AM


New To Site? Need Help?

All times are GMT. The time now is 02:48 PM.


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