AD Vs BD Modulation in Class-D - Page 5 - diyAudio
 AD Vs BD Modulation in Class-D
 User Name Stay logged in? Password
 Home Forums Rules Articles diyAudio Store Gallery Wiki Blogs Register Donations FAQ Calendar Search Today's Posts Mark Forums Read Search

 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
diyAudio Member

Join Date: May 2010
Quote:
 Originally Posted by AP2 The conversation is interesting but perhaps I get lost in translation. For me they are two separate issues. One is the noise (pink-noise at output), this is much greater than in the BD modulation. second point is the noise (EMI radiation)...
White noise actually. But the EMI is lot more complicated because of mosfet body diode currents, power supply currents etc. Common mode rf stuff. But the white noise can go out onto the speaker line and cause emi as well but this is easy to filter. Big loops in your bridge circuit are a different matter.

diyAudio Member

Join Date: May 2004
Location: Budapest
Quote:
 do you really think, that the residual is of a sinusoidal form?
Of course not. It's the sum of infinite sinusoidal components (fourier series), and the freq of first (and dominant) component is fsw. And this component won't magically disappear if you add the two signal, because they are in phase! Don't mix addition and subtraction! If you subtract them, they cancel each other (this is one of the main goals of BD modulation). If you add them, they dont. And if there is a component at fsw, then the whole signal can't be periodical with 1/(2*fsw).

But at low modulation index both outputs are almost perfectly sinusoidal, in this state you can clearly see (without any mathematics, without spectral composition) that your statement is impossible.

Quote:
 Ok, I will do simulation and I will post result here
Do it, and you will see!

diyAudio Member

Join Date: Jan 2006
Location: Kiel
I simulated both AD and BD modulations, and found, that BD modulation indeed does not doubles the radiated frequency... My bad

But it have proven the Bruno's sentences about cancelling of the radiated ripple in common mode in BTL configuration with AD modulation. Please, see attached images. Red trace is what will be radiated off cables. Each output wire will radiate its own ripple voltage and it will be summed in the air then. And since in BTL configuration with AD modualtion the residual ripple is in opposite in output wires, radiation will be cancelled (but on the speaker it will be doubled).
Attached Images
 AD_BD_test.GIF (12.9 KB, 308 views) AD_modulation.GIF (16.8 KB, 288 views) BD_modulation.GIF (19.7 KB, 281 views)

Last edited by 81bas; 25th May 2010 at 11:23 PM. Reason: added some remarks on images

Banned

Join Date: Dec 2009
Location: Italy
Quote:
 Originally Posted by ericbrooking White noise actually. But the EMI is lot more complicated because of mosfet body diode currents, power supply currents etc. Common mode rf stuff. But the white noise can go out onto the speaker line and cause emi as well but this is easy to filter. Big loops in your bridge circuit are a different matter.
yes white-noise, sorry I said pink only for said "his" noise at output.
this is not possible eliminate with filter,. fb,circuit, and pcb design.
I agree with PAFI,carrier not sum. Remaining (for me) double EMI and linearity of modulation on BD. I tested bd modulator hw, it have IMD on signal output, it tested on identical trw for AD modulation.(300KHz)

Regards

 26th May 2010, 07:50 AM #45 diyAudio Member   Join Date: May 2004 Location: Budapest bas81! Thanks for doing sims, and admitting the result! Yes, balanced AD doesn't have common mode output component, and BD does, that's obvious, but my point was: this is not that important as you may think. I don't know anybody who wants to listen to AM radio near to a big PWM amplifier, but if you want, you can suppress common mode component by 60 dB easily, with about 4 dollars cost.
diyAudio Member

Join Date: May 2004
Location: Budapest
AP2, ericbrookling!

I don't know what noise could you talk about... In a theoretical PWM modulator there is absolutely no audio-band noise. Noise comes from not perfect elements (comparator, etc). There is some disturbance at high modulation index and high modulation frequency, but it's not noise, it's side-band leakage, the same as in FM or PM modulation. Its spectra is not continuous like white noise, but discrete, related to modulating signal.

Quote:
 I tested bd modulator hw, it have IMD on signal output, it tested on identical trw for AD modulation.(300KHz)
What is "trw"?

An ideal 150 kHz BD modulator has a bigger distortion then a 300 kHz AD, that's true, but generally both are much smaller then the distortion of the power stage.

Last edited by Pafi; 26th May 2010 at 08:24 AM.

 26th May 2010, 10:54 AM #47 diyAudio Member   Join Date: May 2002 Location: Switzerland Kanwar asked about the reverse recovery problem. In a full-bridge BD topology the reverse-recovery issues would be the same as with a full-bridge AD or a half-bridge. So no advantage in this respect. Do you still use the topology that you once told me about and that does reduce the reverse recovery problem ? Regards Charles
diyAudio Member

Join Date: Dec 2004
Quote:
 Originally Posted by phase_accurate Kanwar asked about the reverse recovery problem. In a full-bridge BD topology the reverse-recovery issues would be the same as with a full-bridge AD or a half-bridge. So no advantage in this respect. Do you still use the topology that you once told me about and that does reduce the reverse recovery problem ? Regards Charles
Thanxz Charles

Yeah i still have that cascode switching topology in some modules under production and it eliminates the reverse recovery problem completely.

The main issue in Half-Bridge is bus-pumping[with +/-180V the rails pumps to 230V under severe LF signal conditions] which is making me to use bridged versions more likely.

Regards,
Kanwar

Banned

Join Date: Dec 2009
Location: Italy
Quote:
 Originally Posted by Pafi AP2, ericbrookling! What is "trw"? An ideal 150 kHz BD modulator has a bigger distortion then a 300 kHz AD, that's true, but generally both are much smaller then the distortion of the power stage.
Hi,
I was looking for the best way to make a new class D amplifier and I am of the opinion that some innovations on AD are the best you can get.
example I try double comparators (just out of phase) on AD modulator and saw that cancel "lish" and noise. BD did not get anything better.

Regards

diyAudio Member

Join Date: May 2010
Quote:
 Originally Posted by Pafi AP2, ericbrookling! I don't know what noise could you talk about... In a theoretical PWM modulator there is absolutely no audio-band noise. ...
The "noise I'm talking about is out of band. It's just the difference between the PWM output and the intended pure audio. Several papers have modelled this as a uniform noise added to the audio. This is not the normal way to look at PWM, but this "noise" is the carrier with modulated signal and it's harmonics. Forget I mentioned the whole thing, from now on I'll only speak of this in modulation terms. Thanks.

 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 OffTrackbacks are Off Pingbacks are Off Refbacks are Off Forum Rules

 Similar Threads Thread Thread Starter Forum Replies Last Post 454Casull Multi-Way 71 2nd November 2011 11:21 PM xuhaoz Class D 20 5th December 2006 04:15 AM Nirav Desai Class D 6 10th January 2005 09:26 AM pokka Class D 7 15th December 2004 07:33 PM sfx Class D 12 6th October 2004 07:02 AM

 New To Site? Need Help?

All times are GMT. The time now is 05:08 PM.