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 7th March 2014, 12:22 AM   #1
diyAudio Member
 
Join Date: Mar 2014
Location: New Jersey
Default Distortion at frequencies above 500Hz???

I have designed a class-d amplifier that will put out 20W into 8 ohms. This is for my own house.

I am lucky enough to have some very good audio test equipment at work namely the Audio Presicion ATS-2 and APx525.

The self-oscillating modulator is my own discrete design with a switching frequency varying from 400kHz to 700kHz. Opamps are OPA1612 and comparator is LT1711.

I am getting around 110dB SNR and around 0.002% distortion from 20 Hz up to 500 Hz.

At 500 Hz the problem starts. The distortion ramps up very quickly such that by 5 kHz the distortion is around 0.15%. This is very stark difference to 0.002%.

Long story short I am starting to suspect that the output filter that I am using is causing the distortion. I have gone over the modulator changing the filter corners and error amplifier / integrator gain. I have also tried multiple other op amps including the LM4562 and AD8620, But no matter what I do the ramp in distortion does not budge.

For this design it is my first time using this type of inductor: http://www.sagami-elec.co.jp/file/7G09B.pdf.
The filter capacitor is Film 150nF 100V: Panasonic / ECWU1154JC9

Output filter picture can be seen as attachment.

Does anybody here have experience with those Sagami inductors? Do they have problems at higher frequencies?

Maybe I am overlooking something else also. All comments are greatly appreciated.
Attached Images
File Type: jpg Output_Filter.jpg (35.2 KB, 162 views)
  Reply With Quote
Old 7th March 2014, 01:27 AM   #2
kevinkr is offline kevinkr  United States
diyAudio Moderator
 
kevinkr's Avatar
 
Join Date: Sep 2004
Location: Boston, Massachusetts
Blog Entries: 6
This appears to be a common mode inductor, do you have both windings phased correctly, the schematic makes it appear they are reversed one relative to the other.. (I am assuming a bridged output.)

It's been more than a decade since I last worked on a class D amp, but the bridged ones I worked with did not use CM chokes, but had an independent diff mode choke on each output of the pair. The output filter is an integrator and I suspect the choke may not have enough differential inductance to do the job, probably a very sizeable amount of HF current in the caps.
__________________
"Extraordinary claims require extraordinary evidence." - Carl Sagan
  Reply With Quote
Old 7th March 2014, 01:35 AM   #3
diyAudio Member
 
Join Date: Mar 2014
Location: New Jersey
Quote:
Originally Posted by kevinkr View Post
This appears to be a common mode inductor, do you have both windings phased correctly, the schematic makes it appear they are reversed one relative to the other.. (I am assuming a bridged output.)

It's been more than a decade since I last worked on a class D amp, but the bridged ones I worked with did not use CM chokes, but had an independent diff mode choke on each output of the pair. The output filter is an integrator and I suspect the choke may not have enough differential inductance to do the job, probably a very sizeable amount of HF current in the caps.
My understanding was that even though it is a single part it is actually two individual inductors (not coupled). I will contact sagami to verify this.
  Reply With Quote
Old 7th March 2014, 02:42 AM   #4
diyAudio Member
 
Join Date: Jun 2009
Depending on the power level you're testing at this might be normal pre-filter feedback behaviour. Since you didn't post a THD versus power sweep for us to look at you might try comparing to, for example, figure 24 of the IRAUDAMP4A datasheet.
  Reply With Quote
Old 7th March 2014, 12:07 PM   #5
diyAudio Member
 
Join Date: Mar 2014
Location: New Jersey
Quote:
Originally Posted by twest820 View Post
Depending on the power level you're testing at this might be normal pre-filter feedback behaviour. Since you didn't post a THD versus power sweep for us to look at you might try comparing to, for example, figure 24 of the IRAUDAMP4A datasheet.
I have both pre and post filter feedback. I will get these plots poster later today.
  Reply With Quote
Old 7th March 2014, 09:38 PM   #6
diyAudio Member
 
Join Date: Mar 2014
Location: New Jersey
I have posted some plots that i did earlier today with the audio analyzer. They should be self explanatory I guess.
Attached Images
File Type: jpg FFT_5kHz_Full_Scale.jpg (209.4 KB, 121 views)
File Type: jpg FFT_500Hz_Full_Scale.jpg (216.1 KB, 116 views)
File Type: jpg FFT_Idling_Zero_Input.jpg (217.3 KB, 114 views)
File Type: jpg FFT_IMD_80hz_&_3kHz.jpg (223.9 KB, 109 views)
File Type: jpg Full_Power_THD+N_VS_FREQ.jpg (196.6 KB, 24 views)
File Type: jpg HALF_Power_THD+N_VS_FREQ.jpg (201.5 KB, 24 views)
File Type: jpg Power_VS_THD+N@5kHz.jpg (188.3 KB, 17 views)
File Type: jpg Power_VS_THD+N@500Hz.jpg (177.3 KB, 14 views)
  Reply With Quote
Old 7th March 2014, 10:31 PM   #7
Khron is offline Khron  Romania
diyAudio Member
 
Join Date: Jul 2005
Location: Finland
Send a message via AIM to Khron Send a message via MSN to Khron Send a message via Yahoo to Khron
On a summary calculation, the cut-off frequency of your filter is about 90kHz - that's VERY high (especially for a self-oscillating amp). More common target frequencies are around half that.

2x 22uH + 330nF would put you around 42kHz.
  Reply With Quote
Old 8th March 2014, 12:25 AM   #8
diyAudio Member
 
nigelwright7557's Avatar
 
Join Date: Apr 2008
Location: Carlisle, England
I use separate t106-2 cores with as many turns as I can get on of 18SWG enamelled copper wire.
__________________
http://www.murtonpikesystems.co.uk PCBCAD50 pcb design software.
  Reply With Quote
Old 8th March 2014, 01:44 AM   #9
diyAudio Member
 
Join Date: Jun 2009
Quote:
Originally Posted by cyaniccypher View Post
I have both pre and post filter feedback. I will get these plots poster later today.
Can you post a schematic too? Kinda hard to reason about this with only a partial schematic that's misleading for the parts of the design it does show.

Last edited by twest820; 8th March 2014 at 01:46 AM.
  Reply With Quote
Old 8th March 2014, 11:27 AM   #10
diyAudio Member
 
Join Date: Mar 2014
Location: New Jersey
Quote:
Originally Posted by twest820 View Post
Can you post a schematic too? Kinda hard to reason about this with only a partial schematic that's misleading for the parts of the design it does show.
Here is the schematic.
Attached Files
File Type: pdf AMP3.pdf (20.6 KB, 27 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
Hammond B3 tech help needed. Distortion only on certain notes/drawbar frequencies. jasonguitar Tubes / Valves 7 2nd June 2011 03:51 PM
midbass 125hz to 500hz kornnylike Subwoofers 55 15th September 2010 11:36 PM
What about using a 18" until 500Hz? hugobors Multi-Way 28 14th September 2010 04:00 PM
Which Driver to use for 100hz-500hz? KCCT82 Multi-Way 27 6th January 2008 03:59 PM
Why don't/can't we use our (sub) woofers from 0 to ~500hz? JinMTVT Multi-Way 11 6th January 2006 04:26 AM


New To Site? Need Help?

All times are GMT. The time now is 01:31 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