New self oscillating post filter feedback topology... - Page 3 - 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 30th April 2010, 04:51 PM   #21
diyAudio Member
 
Join Date: Apr 2010
@81bas: It 'll have to work with real world comparator, for I gonna build it. I 've never seen a 2:1 drop with my topology, neither SOCD nor SOCD pole, only a little drop. Over the WE I can't simulate because there is no "home version" of simetrix! But I will get back on this next week if I don't forget to. Unfortunately the pole implementation has increased the frequency slightly over 500k...hmm...
Do you think this is the way a pole loaded circuit should look like / behave?
If I can get the comparator to work with only ~17mV switching residual, and also the power stage, do you think this will be a good amp?
  Reply With Quote
Old 30th April 2010, 05:40 PM   #22
81bas is offline 81bas  Germany
diyAudio Member
 
Join Date: Jan 2006
Location: Kiel
Quote:
Originally Posted by ViennaTom View Post
@81bas: It 'll have to work with real world comparator, for I gonna build it. I 've never seen a 2:1 drop with my topology, neither SOCD nor SOCD pole, only a little drop. Over the WE I can't simulate because there is no "home version" of simetrix! But I will get back on this next week if I don't forget to. Unfortunately the pole implementation has increased the frequency slightly over 500k...hmm...
Do you think this is the way a pole loaded circuit should look like / behave?
If I can get the comparator to work with only ~17mV switching residual, and also the power stage, do you think this will be a good amp?
First of all, I wonder, that you have only a little frequency drop can you please check it on some sinusoidal signal? Seems in case of 4th order output filter, the UCD switching frequency completely depends on the filter's corner frequency...
17mV at comparator inputs should suffice for most types of comparators, I think, but can produce some delay and noise... LT1016 for example needs about 3-4mV overdrive to completely change it's output state. LM319 needs about 1-2mV. (taken from datasheet diagrams) Layout will be critical at such low input voltage too...
I still dislike too low loop gain at high frequencies (26dB you said), but maybe it is not so dramatical at all...
  Reply With Quote
Old 30th April 2010, 06:00 PM   #23
diyAudio Member
 
Join Date: Apr 2010
"First of all, I wonder, that you have only a little frequency drop can you please check it on some sinusoidal signal?"
As I said I can check but next week. Also the frequency is too high now (~650k @zero output) - any idea how the pole implementation could have risen it? need to do some re-adjustment. But using schottky freewheeling diodes (like Vishay V30200 - almost no reverse recovery losses) even 650 kHz could be realistic - don't you think?
"Seems in case of 4th order output filter, the UCD switching frequency completely depends on the filter's corner frequency..."
This is not UCD! I call it "SOCD pole".
"17mV at comparator inputs should suffice for most types of comparators, I think, but can produce some delay and noise... LT1016 for example needs about 3-4mV overdrive to completely change it's output state. LM319 needs about 1-2mV. (taken from datasheet diagrams) Layout will be critical at such low input voltage too..." Yeah LM310 looks quite good, is there an even faster device? - I fear that the devices tend to get slower with less input overdrive....
"I still dislike too low loop gain at high frequencies (26dB you said), but maybe it is not so dramatical at all... " I don't think there is many ways to increase loop gain at the uppermost frequencies...other than megaherz switching...But I don't think too many guys can hear 20kHz, and almost no one can hear slight distortion at that frequency coming from the amp. Especially when the speakers are overblown at a stage event and thousands of drunken people are "raving" or "headbanging"...lool
BTW: I didnt find out how to quote here?
  Reply With Quote
Old 30th April 2010, 06:22 PM   #24
AP2 is offline AP2  Italy
Banned
 
Join Date: Dec 2009
Location: Italy
Hi,

Interesting thoughts on how to solve some problems.
sure,.. if the comp can work in common mode, then some problems would be solved But... this opens other problems.

Last edited by AP2; 30th April 2010 at 06:45 PM.
  Reply With Quote
Old 3rd May 2010, 05:56 PM   #25
diyAudio Member
 
Join Date: Apr 2010
Quote:
Originally Posted by 81bas View Post
First of all, I wonder, that you have only a little frequency drop can you please check it on some sinusoidal signal? Seems in case of 4th order output filter, the UCD switching frequency completely depends on the filter's corner frequency...
17mV at comparator inputs should suffice for most types of comparators, I think, but can produce some delay and noise... LT1016 for example needs about 3-4mV overdrive to completely change it's output state. LM319 needs about 1-2mV. (taken from datasheet diagrams) Layout will be critical at such low input voltage too...
Oh ohh.... the frequency drop issue is very complicated. There is indeed big drop at the very end of the duty cycle. Need to investigate much more. It also depends on the delay time. But why is frequency drop sooo bad? I think Eva doesn't like frequency drop?! I think it can be even good because the pulses will too early become unrealistically short, if the frequency stays constant. E.g. 500kHz 95% duty => 100ns pulses, 250 kHz -> 200 ns
  Reply With Quote
Old 4th May 2010, 10:00 AM   #26
81bas is offline 81bas  Germany
diyAudio Member
 
Join Date: Jan 2006
Location: Kiel
Quote:
Originally Posted by ViennaTom View Post
But why is frequency drop sooo bad? I think Eva doesn't like frequency drop?! I think it can be even good because the pulses will too early become unrealistically short, if the frequency stays constant. E.g. 500kHz 95% duty => 100ns pulses, 250 kHz -> 200 ns
well, maybe it is not "sooo" bad, but it has nothing good for sound quality definitely, I would say It is needed to avoid too early frequency drop somehow, but seems it is inavoidable at the very end of the duty cycle in phase controlled oscillation
  Reply With Quote
Old 4th May 2010, 04:15 PM   #27
diyAudio Member
 
Join Date: Apr 2010
I dont think the drop too bad, because the very small duty cycle will lead to only very small amount of effective ripple voltage at the switching frequency. So, overall the ripple amplitude doesn't really get worse even if the frequency is only 50% @ 95% duty...
  Reply With Quote
Old 16th August 2010, 03:43 PM   #28
diyAudio Member
 
Join Date: Apr 2010
So hello again, some months have passed and I am now putting together the final schematic with layout planned for autumn. Main Comparator will be MAX 961 (need 2 opposite outputs with perfect delay matching for full bridge, need ultra-high speed even at low overdrive), Input prefilter will be 2nd order bandpass 10 Hz - 25khz (OPA134).
As a main choke I want to use Magnetics "58083-A2" or Arnold "HF1570060-2", 2 cores stacked for one full bridge channel, total 18 turns. This gives 52uH (or 2 x 13uH) My goal is to meet or beat LAB fP6400 under all power conditions, so I need to provide 3.2 kW burst sine waves, 8ms, 1kHz at 2 Ohms load (EIAJ burst power test at 2 Ohms is definitely the worst case for the magnetic calculation). This condition needs rms output current of 40A (-> e.g. coil peak current ~60A). So on ~140 oersteads, H-Flux 60 u core will still have ~50% permeability left. And thats by far the worst case design point. Any better idea / opinion?
  Reply With Quote
Old 16th August 2010, 04:20 PM   #29
81bas is offline 81bas  Germany
diyAudio Member
 
Join Date: Jan 2006
Location: Kiel
Which mosfets do you plan to use? Driver?
  Reply With Quote
Old 16th August 2010, 05:35 PM   #30
diyAudio Member
 
Join Date: Apr 2010
Quote:
Originally Posted by 81bas View Post
Which mosfets do you plan to use? Driver?
IRFP4668, series schottky 20V 60A and antiparallel V30200C
Level-shifter: IRS2011, actual driver: MIC4421 (high side driver floating), very low-ohmic gating circuitry required to ***-kick IRFP4668 - this one is quite sluggish...
  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
Post filter AND pre filter feedback..! Lars Clausen Class D 30 4th April 2013 04:03 PM
question :post filter feedback titanchen68 Class D 38 2nd August 2006 02:00 AM
Question- post filter feedback error correction? darkfenriz Class D 11 13th June 2006 10:31 PM
Voltage Mode Pre and Post Filter Feedback Genomerics Class D 30 6th April 2006 07:27 PM


New To Site? Need Help?

All times are GMT. The time now is 12:32 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