My very first Class D pwm (switching) amplifier. - Page 10 - 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 17th June 2003, 01:28 PM   #91
diyAudio Member
 
Join Date: May 2002
Location: Switzerland
You are on the right track.
What speaker impedance didi you design it for ? Just from Looks it seems to be made for a very low impedance or you will have quite a high Q with an 8 Ohms speaker (personally I would double the main inductor and halve the cap).
For the notch part I would also enlarge the inductor and decrease the capacitor. To avoid too sharp a notch I would connect a resistor parallel to the resonant circuit. I would also connect a cap around 10 nF parallel to the speaker output and also a Zobel network (or "things might get interesting" when the output is left open).

Regards

Charles
  Reply With Quote
Old 17th June 2003, 01:41 PM   #92
ssanmor is offline ssanmor  Spain
diyAudio Member
 
Join Date: Jan 2002
Location: Madrid
Well, I have designed it for 4 Ohms, althugh what I am trying to do first is to make sure that I understand the problems, I am not worried by the numbers yet.
Thanks. What is a reasonable rms voltage for the carrier ripple basing on your experience? (I assume that it is not audible however)

Thanks!
  Reply With Quote
Old 17th June 2003, 03:59 PM   #93
diyAudio Member
 
Join Date: May 2002
Location: Switzerland
Hi ssanmor

You are on the right track.
I would however take different values: Iwould double the inductance of the main filter an half the capacitor. The values shown might be good for a very low load impedance or you will have quite a high Q.
I would do the same thing for the notch filter. Additionally I would connect a resistor between 100 and 1000 Ohm in parallel to lower it's Q a little.
I would then add a smaller capacitor (between 1 and 100 nF) at the output to further improve suppression of the higher RF output spectrum.

And finally I would also connect a Zobel network at the output. Otherwise "things might get interesting"* if no load is connected to the output. Just keep in mind that a lowpass without any load is a series resonant circuit !!

If the 44 dB suppression is sufficient depends heavily on your application.

Regards

Charles

*copyright by N.P.
  Reply With Quote
Old 17th June 2003, 04:08 PM   #94
diyAudio Member
 
Join Date: May 2002
Location: Switzerland
Hi ssanmor

Sorry for posting almost the same stuff twice. But I couldn't see my former post when I checked, so I assumed that something went wrong.

As already mentioned the desired carrier suppression is depending on your application.

If your application is a powerded subwoofer with good RF shielding you might get away with very little suppression (you then have to keep efficiency in mind because a speaker isn't a lLOW loss inductor).

The famous TacT MillenniuM amp for instance has a 2nd order output filter with a cuttoff frequency of 60 kHz approx (without any notch!). Try to imagine how much carrier suppression this leaves.

If you want to use long speaker cables you should be concerned about the part above 1MHz. If you want to efficiently get 200 kHz into the air you'd need a fairly large antenna ! Things change however as you get up infrequency.

But your 44 dB is quite a good starting point IMO.

Regards

Charles
  Reply With Quote
Old 18th June 2003, 07:31 AM   #95
ssanmor is offline ssanmor  Spain
diyAudio Member
 
Join Date: Jan 2002
Location: Madrid
Yes, Charles, the TacT Millenium uses a very simple filter which should leave a very high ripple. But, to be fair, we must say that it uses variable supply rails, so at low volumes, the ripple is also reduced, unlike a "conventional" (if this word is compatible with Class D) design, in which you get a constant ripple level, given by the supply rail voltage and the output filter attenuation at the switching frequency. (All this considered, I still don't think it worths 10000 dollars! ).

Returning to our discussion. My primary intend is to build a full range amplifier, not only for subwoofers.

As an important design parameter, I would like you to advice on what could be a reasonable maximum phase shift at the filter that can allow me to perform useful negative feedback if I take it from the speaker output.

Thanks.
  Reply With Quote
Old 19th June 2003, 03:03 PM   #96
koldby is offline koldby  Denmark
diyAudio Member
 
Join Date: Jun 2003
Location: Ruds Vedby
I suppose you all are aware of this :
http://listen.to/audioexperiment

Vrey simple PWM amp, and a Danish Hi-Fi magazine says it sounds godd too.

Enjoy..

Koldby
  Reply With Quote
Old 20th June 2003, 07:34 AM   #97
ssanmor is offline ssanmor  Spain
diyAudio Member
 
Join Date: Jan 2002
Location: Madrid
Yes, Koldby, Thanks.

I saw it a few days ago.

This design has several points to highlight:
First of all, the modulator stage is quite odd: it doesn't use a triangle wave to generate the PWM modulation. On the other hand, it feedbacks the output signal (before filter), substracts it from the input audio and compares to 0. What is not clear for me is how to determine the oscillation frequency.

However, it uses P and N mosfets, what is not very adequate for high power levels. I was involved in a discussion it its corresponding forum about how to implement it with N devices only and we lead to a solution similar to the one used in the crest-LT amplifiers. (using a IR2110/2113 driver).

Referring to the output filter, the author highlights that he has wound his own coil instead of using a commercial one. What about using power inductors as the filter? Personally, I had chosen this: DMT3-35-12.
http://www.coi1craft.com/ds/dmt.pdf

As long as the maximum AC current is not exceeded, it should be OK, shouldn't it?

We will continue discussing.
Best regards to all the DIY'ers.
  Reply With Quote
Old 20th June 2003, 09:14 AM   #98
diyAudio Member
 
Join Date: May 2002
Location: Switzerland
Hi

I finally managed to get my old one scanned. An interesting fact is that the Danish guy used an almost similar approach for the gate drive as I did.

As soon as I have more time I will post some info what I did and why (including what I would do differently today).

Regards

Charles
Attached Files
File Type: zip classd.zip (59.9 KB, 15335 views)
  Reply With Quote
Old 20th June 2003, 09:20 AM   #99
koldby is offline koldby  Denmark
diyAudio Member
 
Join Date: Jun 2003
Location: Ruds Vedby
Ssanmor,

Actually I think that design is very interesting because of the "strange" modulator!
Aparently the switch freq. is determined by time constants (lags) in the feedback loop, but as long as it is high enough, it dosen´t really matter precisly what the freq. is, or what?
If it is sounding as good as the magazine says, it should be possible to improve it very much as the components used in (op-amps - comperator) is only half way decent.
Converting the output to a N only would be nice though.

Koldby
  Reply With Quote
Old 20th June 2003, 09:42 AM   #100
koldby is offline koldby  Denmark
diyAudio Member
 
Join Date: Jun 2003
Location: Ruds Vedby
Hi
A member of the news group related to this "audioexperiment" PWM amp. has just come up with a link, better explaining the theory behind this particually design:
http://www.mueta.com/downloads/AES.PDF

Hope it will be helpfull.

I hope I do not intrrupt the main direction of this thread?

Koldby
  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



New To Site? Need Help?

All times are GMT. The time now is 04:17 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