Anyone interested in a digital amplifier project? - Page 54 - 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 1st January 2008, 04:20 PM   #531
Dave Z is offline Dave Z  United States
diyAudio Member
 
Join Date: Dec 2007
Actually, the lower switching frequency has lower distortion and better efficiency but can not handle the higher frequency range of full band audio. That is why you aren't seeing so much really high carrier frequencies. You are starting to see some, but that is due to the consumer electronics companies needing to lower thier costs by using smaller inductors. You will also see more single ended class D amps for the same reason (less inductors, lower cost) but power supply design is much more complicated due to supply pumping and you lose the advantage of the carrier being cancelled out across the load because it's common mode.

The reason for the higher efficiency and lower distortion with lower switching frequencies is this:

A major loss in class D amps is the transistion from low to high or high to low, This is also where a lot of the output stage THD come from. This is where the MOSFETs must travel through thier linear region and have a small amount of cross conduction. With a lower switching frequency, this happens less amount of times per second. The more times per second you are in the linear region the more heat you will be dissipating. Also since the transition is never instant. The rise and fall time of the transition is introduced as THD. Again the more time per second this happens the higher the THD. As far as the rise and fall times, that is why the more cross conduction you have, the lower the THD you will have (Up to a limit). But of course more cross conduction = less efficiency and higher noise levels. These are the trade offs you can make for yourself if you design your own output stage. With the TI chips, you can take the PWM modulator output and put it into your own power stage, and conversely take a TI power stage and create your own analog modulator if you want an analog input class D amp (which I still believe is better than completely digital) but of course you will need to build your own PSRR (feed forward correction) circuit and feedback if you want to use it with a linear supply.

Dave
  Reply With Quote
Old 1st January 2008, 06:33 PM   #532
diyAudio Member
 
John Hope's Avatar
 
Join Date: Feb 2005
Location: nr Oxford UK
Default Performance of switching amps

Yes, all this is true.
What's also true is that at the present time, open loop digitally modulated and even most closed loop Class D amps don't approach the state of the art in terms of THD or S/N.
(John Westlake's closed loop digital amp may be an exception to this, though)

Infuriatingly, the top hand seems to stilll held by analog power stages. One of the interesting developments I've come across here is the National Semiconductor LME49810 analog power amp driver chip which when used with complementary bipolar output transistors leads to a simple power amp of 500W with THD around 0.0006%.

So, one can knock together something like this in a few weeks. Or you can spend 12 years designing a series of the most complex and critical to get right amplifiers known to man and teach yourself a myriad of useful design skills along the way?
__________________
At Spes Infracta
__________________
  Reply With Quote
Old 22nd March 2008, 12:36 PM   #533
diyAudio Member
 
Join Date: Jul 2002
Location: Brisbane
John, any update on your prototype? Have you designed PCBc yet?
__________________
-Tor
  Reply With Quote
Old 22nd March 2008, 01:28 PM   #534
diyAudio Member
 
Join Date: Oct 2003
Location: The Lab
Quote:
(John Westlake's closed loop digital amp may be an exception to this, though)
Is John's amplifier in production by now?
__________________
www.newclassd.com
  Reply With Quote
Old 22nd March 2008, 01:34 PM   #535
diyAudio Member
 
jan.didden's Avatar
 
Join Date: May 2002
Location: Great City of Turnhout, Belgium
Blog Entries: 7
Default Re: Performance of switching amps

Quote:
Originally posted by John Hope
Yes, all this is true.
What's also true is that at the present time, open loop digitally modulated and even most closed loop Class D amps don't approach the state of the art in terms of THD or S/N.
(John Westlake's closed loop digital amp may be an exception to this, though)

Infuriatingly, the top hand seems to stilll held by analog power stages. One of the interesting developments I've come across here is the National Semiconductor LME49810 analog power amp driver chip which when used with complementary bipolar output transistors leads to a simple power amp of 500W with THD around 0.0006%.

So, one can knock together something like this in a few weeks. Or you can spend 12 years designing a series of the most complex and critical to get right amplifiers known to man and teach yourself a myriad of useful design skills along the way?

It's the journey that counts, not the destination

Jan Didden
__________________
If you don't change your beliefs, your life will be like this forever. Is that good news? - W. S. Maugham
Check out Linear Audio!
  Reply With Quote
Old 22nd March 2008, 07:08 PM   #536
diyAudio Member
 
John Hope's Avatar
 
Join Date: Feb 2005
Location: nr Oxford UK
Default PowerDAC3 sings!

I have completed the testing on my PowerDAC3 digital amplifier power module. It checks out really good, with THD around the 0.02 - 0.03% mark 20-20kHz at all powers into 4-8 ohms. I am very happy with the power circuitry now and will not make any further changes to the power pcb.

The control pcb also functions as planned.
For testing and audition I fed it with serial streams of 8-bit 'PWM numbers' derived from the signal processing chain in my PowerDAC 1 digital amplifier. It sounds better than any of the previous PowerDACs I have made and has truly effortless punch.
I'll post some pics sometime soon.

The Actel Fusion AFS600PQ208 FPGA in the control board gave me a lot of trouble. It's a very new device and not properly in production yet. Actel have taken nearly a year to get this device going and they still aren't there. Some of the functionality was lacking in the REV D silicon. Then the s.o.b. failed to properly program its flash after about 30-50 reprogs, so the claimed 500 cycles flash write endurance Actel specifies is still wishful thinking. The other REV D specimen I'd been given had yet other functionality missing, but fortunately not enough to impede my progress towards music!. I've been promised by Actel a fully working REV E device by mid April, but by now I am 'gatvol' of this FPGA and I've decided to redesign the control board of the amp to use a more reassuringly familiar Altera Cyclone II device.

What influenced my decision was the fact that the EMC environment proved benign and the present FPGA logic in the Actel functioned flawlessly on the lower board only a few cm from where up to 10A 60V is being switched in 5ns on the power board above (with no ringing!)

So I thought in the light of this why not go the whole hog and make more than a Power Module that still requires external signal processing - why not offload ALL the signal processing and supervisory functions (that I was going to do in a separate processing module) onto a new drop-in processing board that will take I2S or AES in from source and stack with the preset power board. This would make a really attractive OEM digital amplifier module. So this is what I'm doing.

More on this story later, as they say on the news. . .
__________________
At Spes Infracta
__________________
  Reply With Quote
Old 22nd March 2008, 11:32 PM   #537
jkeny is offline jkeny  Ireland
Banned
 
Join Date: Feb 2007
Location: Dublin
Great work John on this amplifier - it seems like a tremendous amount of work judging by the PCB photos - are these wire-wrap connections between parts? Will the final design be a commercially available product or incorporated into one?

As to PS design, I have done some tests with battery power (12V 7aH SLA) into the output stage of a Panasonic SA-XR57 Equibit amp & I didn't find an improvement in the sound, in fact it lost some punch to the sound compared to the mains PS. http://www.diyaudio.com/forums/showt...55#post1461455

The mains is fed through a step-down isolation transformer (240V to 110V) which I believe, at the moment, is what gives the amp a better sound than the battery! I will be testing further with paralleled batteries to increase the current.

I attach the PS schematic from the SA-XR57 amp for your perusal (the variable control of the PS voltage is circled). Panasonic have changed the this PS circuit from the one that was used in the XR45 & mentioned by Kims further back in the thread. The XR45 PS (considered to be the best sounding) used a DC-DC converter as the control element.

Koon seems to favour a discrete PS vol control over a SMPS one http://koonlab.com/TAS4i.html

Any comments?
Attached Files
File Type: pdf xr57 schematic part1.pdf (75.6 KB, 201 views)
  Reply With Quote
Old 23rd March 2008, 11:55 AM   #538
diyAudio Member
 
John Hope's Avatar
 
Join Date: Feb 2005
Location: nr Oxford UK
Default PowerDAC3 pcbs

Here is a top view of the PowerDAC 3 power pcb. All the pcbs shown are 4 layer and have parts both sides of the board.
Attached Images
File Type: jpg powermod_top.jpg (88.8 KB, 911 views)
__________________
At Spes Infracta
__________________
  Reply With Quote
Old 23rd March 2008, 11:58 AM   #539
diyAudio Member
 
John Hope's Avatar
 
Join Date: Feb 2005
Location: nr Oxford UK
Here is a bottom view of the PowerDAC 3 power pcb. The original output devices shown here (SO-8 packages FDS3672) have been replaced with FDD3684 in the TO-252 SMD package.
Attached Images
File Type: jpg powermod_bot.jpg (97.2 KB, 865 views)
__________________
At Spes Infracta
__________________
  Reply With Quote
Old 23rd March 2008, 11:59 AM   #540
diyAudio Member
 
John Hope's Avatar
 
Join Date: Feb 2005
Location: nr Oxford UK
Here is a top view of the PowerDAC 3 (lower) Control Board. See there is plenty room for other stuff. . .
Attached Images
File Type: jpg ctrlboard_top.jpg (71.0 KB, 824 views)
__________________
At Spes Infracta
__________________
  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
Noise measurement amplifier - interested? jackinnj Solid State 15 27th March 2008 10:36 PM
New digital XO project borges Multi-Way 2 8th January 2008 07:49 PM


New To Site? Need Help?

All times are GMT. The time now is 01:24 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