What's wrong with Class-D? - Page 14 - 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 9th March 2013, 03:20 PM   #131
GoatGuy is offline GoatGuy  United States
diyAudio Member
 
Join Date: Dec 2012
Location: SF Bay Area
I'm real late to this discussion, but for me these are the issues.

[1] Class D "synthesizes" the output waveform by blasting tiny pulses of power at variable width to a strip of high-current low-pass filters ... one hopes that the feedback circuitry is actually accomplishing the task of making a faithful output waveform viz a vis the input.

[2] #1 isn't in itself really all that bad, necessarily. Its like using teeny tiny jackhammers to carve wood. Thousands of them, simultaneously, so I hear. The results can LOOK like carved wood. The low-pass filters is the "sanding and buffing" that renders the carving smooth enough to let anyone think it was actually carved with awls, chisels, knives and manual labor. Sadly, it wasn't. Is this necessarily bad, especially "after sanding"? Dunno. Fear, uncertainty, doubt ... about the 'damage' caused to the waveform, should all that sanding not quite do what the design says it should do.

[3] RF and RFI - are a natural and inescapable consequence of the D method. All those little jackhammer pulses, positive and negative, have rise-times and fall-times on the order of dozens of nanoseconds (else, the D method wouldn't be anywhere near as thermally "efficient" as it generally is!). Hundreds of thousands of ultra-fast slew-rate pulses per second ... 1/P = F ... 1/100ns = 10 MHz ... there's going to be a lot of RF energy in that box.

[4] Because of #3, and because of a series of articles I read some years back where a researcher (Normal Korman) tracked down a not-so-subtle issue with RF and RFI interfering with the sonic purity of his amplifiers ... and on quashing ... replacing no other components than some wiring placement ... was bowled over listening to the same recordings, same equipment, same speakers, etc. RF/RFI was the culprit. And in the home-built and commercial Class D equipment ... is such RF/RFI suppression really all that effective? Ah... FUD perhaps, but the D stands not for digital, but doubt.

So, all I'm saying is - well, listen critically to it. if you're building your own, then make extra, extra special care in shielding all the input wires (coax from jack to board), and make sure the board layout totally isolates things. From having worked on a lot of MHz to GHz designs professionally, I wouldn't even think about implementing Class D without RFI "cages" on the output, to contain the RFI within a Faraday cage and prevent its leaking, unexpectedly, all over the place.

There you are.

GoatGuy
  Reply With Quote
Old 9th March 2013, 04:08 PM   #132
diyAudio Member
 
Kindhornman's Avatar
 
Join Date: Aug 2012
Location: Los Angeles, California
GoatGuy,
You forgot the interaction with a smps power supply and the chance of switching remnants passing through to the amplifier and being mixed with the signal source along with its rfi that can also be injected into the system. I am not saying that all of these issues can't be overcome, just that they do add a layer of complexity to the entire picture besides every other analog issue that still exist like any other analog circuit.
  Reply With Quote
Old 9th March 2013, 04:46 PM   #133
qusp is offline qusp  Australia
diyAudio Member
 
qusp's Avatar
 
Join Date: Oct 2009
Location: Brisbane, Australia
Quote:
Originally Posted by AP2 View Post
Hi,
Just to conclude, on your statements:
capacitance multiplier: it is very far from what we have written, becouse is incapable of transient response. (or not know)
Nice smps: I do not know a single one that I can put in a big class AB and has a fast response to transients.
I agree on the batteries then!

regards
you are still disqualifying your very fast power supplies from being in the SMPS group? or you mean bigger amp than they can supply? its just a little bit of a confusing comment from the designer of the 'SMPS' that has the fastest response of any power supply I have seen with Class AB

Lead acid batteries as quiet power amplifier supplies? really? NOT. they are quietish with low constant loading, thats all, they are neither fast or quiet with high slewing high current demanding needs

some of the new battery tech is much better, but I would and do still post regulate or at least buffer with caps.

Last edited by qusp; 9th March 2013 at 04:49 PM.
  Reply With Quote
Old 9th March 2013, 04:51 PM   #134
diyAudio Member
 
Join Date: Jun 2010
Location: St. Petersburg
Actually low pass in frequency domain means decimation in analog domain literally.
Analog->DSP->analog signal suffers interpolation artifacts in compare to original analog signal.

AFAIK Tektronix interpolation ICs for digital scopes is among some advanced NoHow for example. Wadia decoding computers is another sample of ramification that might be required "to restore" decimated Red Book PCM.
  Reply With Quote
Old 9th March 2013, 05:00 PM   #135
xrk971 is offline xrk971  United States
diyAudio Member
 
xrk971's Avatar
 
Join Date: Oct 2012
Location: Virginia
Quote:
Originally Posted by GoatGuy View Post

So, all I'm saying is - well, listen critically to it. if you're building your own, then make extra, extra special care in shielding all the input wires (coax from jack to board), and make sure the board layout totally isolates things. From having worked on a lot of MHz to GHz designs professionally, I wouldn't even think about implementing Class D without RFI "cages" on the output, to contain the RFI within a Faraday cage and prevent its leaking, unexpectedly, all over the place.

There you are.

GoatGuy
The RFI is a good point, but how is 400 kHz + freq going to affect audible spectrum? Also, if leads from chip to LC filter is smaller 1/4 wave of 400 khz antenna, reduces emission. Shielded wire on input seems like a good idea in general for high impedance high gain inputs. I like your tiny jack hammer analogy, and I would say that given jack hammers the size of 400 grit sandpaper grains, the end result is the same.
  Reply With Quote
Old 9th March 2013, 05:02 PM   #136
diyAudio Member
 
Kindhornman's Avatar
 
Join Date: Aug 2012
Location: Los Angeles, California
Why do we keep coming back to the statement that class D has to be in the digital domain? It can be done completely in the analog domain from what I understand, where is this dsp having to come from keep rearing its ugly head?
  Reply With Quote
Old 9th March 2013, 05:47 PM   #137
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
Oh, still trying to find the magic in something not measurable about audio waveforms that gets lost or dirty in PWM conversions or near a SMPS? Perhaps magic has more to do with expectations and human feelings, what about love? Personally I find more magic in setting up PA and mixing friend DJs and bands playing for friends, these are the origins of music.
__________________
I use to feel like the small child in The Emperor's New Clothes tale
  Reply With Quote
Old 9th March 2013, 05:49 PM   #138
xrk971 is offline xrk971  United States
diyAudio Member
 
xrk971's Avatar
 
Join Date: Oct 2012
Location: Virginia
Quote:
Originally Posted by Kindhornman View Post
Why do we keep coming back to the statement that class D has to be in the digital domain? It can be done completely in the analog domain from what I understand, where is this dsp having to come from keep rearing its ugly head?
Good point. The only thing digital is the fact that FET switches on or off. There is no DSP.
  Reply With Quote
Old 9th March 2013, 05:55 PM   #139
diyAudio Member
 
Join Date: Jun 2010
Location: St. Petersburg
Analog means switching frequency is infinity.
Class D means switching frequency is below infinity usually within 200-500KHz.
  Reply With Quote
Old 9th March 2013, 05:55 PM   #140
qusp is offline qusp  Australia
diyAudio Member
 
qusp's Avatar
 
Join Date: Oct 2009
Location: Brisbane, Australia
more than that, Class D IS in the analogue domain, its an analogue technology, whether the modulator has digital inputs or not is beside the point

why do people continuously misunderstand/misrepresent this even here?
  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
What is wrong? lumanauw Solid State 43 20th January 2013 07:39 PM
Contest: Linear Power Amp in a mint tin (class Aa, class AB, or class B) danielwritesbac Chip Amps 206 19th January 2013 09:52 PM
What am i doing wrong, why wont my JLH operate in class A / get warm? Shamron Solid State 63 12th November 2011 04:27 PM
PASS DIY F4 beaten hands down by my class D amp, what's wrong? mr_push_pull Pass Labs 98 31st May 2011 11:23 AM
Wrong Voltage / current on BoZ... or Wrong Measuring ? gionag Pass Labs 8 14th May 2008 12:24 PM


New To Site? Need Help?

All times are GMT. The time now is 12:45 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