Are class D amplifier only good for subwoofer speakers ? - Page 6 - 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 22nd February 2014, 08:21 PM   #51
dewardh is offline dewardh  United States
diyAudio Member
 
Join Date: Dec 2005
Location: Berkeley, CA
Quote:
Originally Posted by Pallas View Post
An amp can only be called "flawless" if it maintains a constant frequency response regardless of the driven load. Unlike the Hypex, ICEpower ASX2, and a few others, the CDA designs do not seem to meet this key criterion.
Maybe that's your "key criterion", but it's not mine. Has no significance at all for me, in fact, since any effect is confined to tweeter response, something I always "tune" anyway, and there the CDA's "load sensitivity" almost always works in my favor. Frequency response "adjustment" is an essential part of any system design, and the possibly differing response of pre and post filter feedback designs is not a "flaw" either way. It's completely predictable, and simply becomes another part of the crossover/loudspeaker/system design.

The CDA amps (and the Chinese "reference amp" implementations as well) are "board level" components, for users whose "DIY" ability is enough beyond merely swapping amps and cables that addressing load effects is a trivial exercise.
  Reply With Quote
Old 23rd February 2014, 12:11 PM   #52
diyAudio Member
 
Join Date: Dec 2003
Location: Munich
Quote:
Originally Posted by dewardh View Post
. . if any "splatter" exists it is below the (very low) noise floor of your test system.....
Come on !
By comparing posting #40 & #42 you can see, that the limits of the measurement system are proven clearly better than the unmodified original amp, except k8 & k11.
For reference I am attaching the direct comparison of the unmodified original amp (dead time setting 45ns) vs the test system.

Quote:
Originally Posted by dewardh View Post
What you do show is that using the amplifier as delivered there is no dead time induced "splatter" and any distortion components are more than 100dB down . . .
...well, to me even the spectrum of the unmodified amp looks a little bit like spl... , but most amps do - and IMHO in the region of -100db this is not a catastrophy.
Furtheron if you can easily turn the moderate peaks into skyscrapers simply by increasing the dead time just a few ns then this is a nice visualisation that the mechanism of dead time distortion introduces a pretty unpleasant spectrum of distortion.
I am not saying that classD is bad, I am also not saying the CDA244 would be bad. IMHO the distortion figures of the CDA224 are pretty good.
I am just disagreeing that classD would inherently better than class AB, because of the absence cross over related high order harmonics.
Both technologies can be brought to reasonably low amount of high order harmonics, but both are suffering from unpleasant distortion effects which are related to the use of two power devices, whereoff one is needed to provide positive currents into the load while the other is needed to provide negative currents.
And I am saying that the classD technology is less mature than classAB, which leads to a larger spread of design quality in the market place.


P.S.
From my perspective my sound card is not really great.
The harmonics are OK, but the noise floor is just so so.
Regarding the noise floor: Don't let you get tricked by the graph of spectral density. If in this graph the noise floor is at -134db, this does not mean the S/N ratio would be that low. In order to derive the S/N ratio from the spectral density we have to integrate it over the bandwidth and we will get a pretty frustrating result. Even with the A-weighting, which gives slightly nicer numbers again, it is not great.
Attached Images
File Type: jpg Spectrum_orig_vs_tester.JPG (93.4 KB, 168 views)
File Type: jpg k2_orig_vs_tester.JPG (92.1 KB, 169 views)
File Type: jpg k3_orig_vs_tester.JPG (81.5 KB, 164 views)
File Type: jpg k4_orig_vs_tester.JPG (81.7 KB, 160 views)
File Type: jpg k5_orig_vs_tester.JPG (83.5 KB, 155 views)
File Type: jpg k6_orig_vs_tester.JPG (83.2 KB, 9 views)
File Type: jpg k7_orig_vs_tester.JPG (83.6 KB, 9 views)
File Type: jpg k8_orig_vs_tester.JPG (84.9 KB, 9 views)
File Type: jpg k9_orig_vs_tester.JPG (87.0 KB, 9 views)

Last edited by ChocoHolic; 23rd February 2014 at 12:14 PM. Reason: Edit: corrected a 'not'
  Reply With Quote
Old 23rd February 2014, 03:07 PM   #53
dewardh is offline dewardh  United States
diyAudio Member
 
Join Date: Dec 2005
Location: Berkeley, CA
Quote:
Originally Posted by ChocoHolic View Post
...well, to me even the spectrum of the unmodified amp looks a little bit like spl... , but most amps do - and IMHO in the region of -100db this is not a catastrophy.
It seems that we use "splatter" differently . . . to me splatter is nonharmonically related distortion. But we do agree in that -100dB is not . . . catastrophic . . .

Quote:
Originally Posted by ChocoHolic View Post
IMHO the distortion figures of the CDA224 are pretty good.
Um . . . yes . . . "pretty good". My system is set up so that with the volume at "full" 0dBfs produces 105dB at the listening position. The combined noise floor, microphone to listening room, is at best 25dB (and typically above that). A "distortion component" that is 100dB down would not be heard even completely isolated from the originating signal . . . with a simultaneous 105dB broadband (music) signal it would not be heard, period. I'd call that "pretty good enough" . . .



Quote:
Originally Posted by ChocoHolic View Post
I am just disagreeing that classD would inherently better than class AB, because of the absence cross over related high order harmonics.
Well . . . IMO and to my ear the crossover related splatter from typical Class AB amps is enough worse to be audible, and annoying. There are certainly examples where it is not . . . LM3886 "chip amps", for example, have excellent bias tracking and no audible crossover distortion in normal operation. Adcom produced some Class AB bipolar amps that beat the problem (their MOSFET amps maybe not so much). Doug Self wrote a whole chapter in his amplifier book on how to avoid it. None are better than the lowly Class D amp board you just tested. It's "pretty good enough" . . . and certainly "not just for sub-woofers" . . .
  Reply With Quote
Old 23rd February 2014, 08:42 PM   #54
a.wayne is offline a.wayne  United States
Sin Bin
 
Join Date: May 2008
Location: Front Row Center
Quote:
Originally Posted by Kjeldsen View Post
I have listend a lot to the mk3 version

Lyngdorf Audio - Millennium Mk IV

It's the best amplifier I have heard, and I have been able to AB test with several other amplifiers many different loudspeakers. I love it because it's very revealing. A music producer was also impressed and have never heard such an revealing amplifier, but would never buy it himself, because he could also here the errors in the production or were the cuts were made. Therefore he preffered a more foregiving amplifier.
Is this a class-d amp ? it weighs 25Kilos ......
  Reply With Quote
Old 23rd February 2014, 08:44 PM   #55
a.wayne is offline a.wayne  United States
Sin Bin
 
Join Date: May 2008
Location: Front Row Center
Quote:
Originally Posted by dewardh View Post
It seems that we use "splatter" differently . . . to me splatter is nonharmonically related distortion. But we do agree in that -100dB is not . . . catastrophic . . .


Um . . . yes . . . "pretty good". My system is set up so that with the volume at "full" 0dBfs produces 105dB at the listening position. The combined noise floor, microphone to listening room, is at best 25dB (and typically above that). A "distortion component" that is 100dB down would not be heard even completely isolated from the originating signal . . . with a simultaneous 105dB broadband (music) signal it would not be heard, period. I'd call that "pretty good enough" . . .




Well . . . IMO and to my ear the crossover related splatter from typical Class AB amps is enough worse to be audible, and annoying. There are certainly examples where it is not . . . LM3886 "chip amps", for example, have excellent bias tracking and no audible crossover distortion in normal operation. Adcom produced some Class AB bipolar amps that beat the problem (their MOSFET amps maybe not so much). Doug Self wrote a whole chapter in his amplifier book on how to avoid it. None are better than the lowly Class D amp board you just tested. It's "pretty good enough" . . . and certainly "not just for sub-woofers" . . .
Not even good for subwoofer amps , Imo class-a or a/b amplifiers are better at the freq extremes , hence better bass than class -d ....
  Reply With Quote
Old 23rd February 2014, 08:56 PM   #56
diyAudio Member
 
vacuphile's Avatar
 
Join Date: Apr 2011
Location: Seaside
A.wayne, what makes you think so? It certainly isn't my experience.
  Reply With Quote
Old 23rd February 2014, 09:09 PM   #57
a.wayne is offline a.wayne  United States
Sin Bin
 
Join Date: May 2008
Location: Front Row Center
Listening test where we have compared them (never heard Hypex, we compared BelCanto and Rotel) the class A/B and Class -a were all better in the Bass..

YMMV

Last edited by a.wayne; 23rd February 2014 at 09:11 PM.
  Reply With Quote
Old 23rd February 2014, 09:49 PM   #58
diyAudio Member
 
vacuphile's Avatar
 
Join Date: Apr 2011
Location: Seaside
That sounds like a very controlled environment! C'mon, don't insult our intelligence. How can you make blanket statements about a whole class of amplification based on what some guys in a room think about a couple of different amps.
  Reply With Quote
Old 23rd February 2014, 10:31 PM   #59
a.wayne is offline a.wayne  United States
Sin Bin
 
Join Date: May 2008
Location: Front Row Center
Yes how dare us buy amplifiers and listen to them when we have you to tell us what to think ..


No wonder audio is dead ...
  Reply With Quote
Old 23rd February 2014, 10:39 PM   #60
a.wayne is offline a.wayne  United States
Sin Bin
 
Join Date: May 2008
Location: Front Row Center
Hey Vacuphile , why not postup a 20hz Sq wave from one of your favorite Mr Class -D , a thing of beauty huh, sure beat listening huh ..


  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
DIY Good Class D amplifier Acca Class D 398 27th July 2014 08:31 AM
Are the Dayton HPSA1000 1000W Subwoofer Amplifier any good? harrisni Subwoofers 0 2nd June 2007 04:53 AM
Good Subwoofer Amplifier broka Solid State 0 17th September 2006 06:25 AM
SFX Class D subwoofer amplifier modification sagarverma Class D 67 26th June 2006 04:26 AM


New To Site? Need Help?

All times are GMT. The time now is 04:52 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