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 5th May 2004, 03:43 PM   #521
Alcaid is offline Alcaid  Norway
diyAudio Member
 
Alcaid's Avatar
 
Join Date: Nov 2003
Location: Klepp
At this webpage they have a formula to calculate the frequency of the triangle wave generator.

Your values;
Rt: 10k
R1: 33k
R2: 820k

would give a working frequency of 250kHz multiplied with (R2/R1) whick is approx. 621kHz. Is their formula wrong, or have you posted wrong values in the schematic?
  Reply With Quote
Old 6th May 2004, 03:48 AM   #522
sfx is offline sfx  Germany
diyAudio Member
 
Join Date: Feb 2004
Location: Australia
Hi Alcaid,

good question - I checked again, the resistor values and the capacitor are the ones I'm using. They are also still on the breadboard I used for testing. The question is now, if their formula is wrong (unlikely), or if I made a mistake measuring my frequency. On the other hand, I don't really believe that I drive my circuit at 620kHz... I guess I'll have to get the amp out, and measure again. Now I'm curious...

I don't have any experience with MAX wave generators. I found this one on google:
clicky
(that's for the MAX038, which seems to be quite nice)

I won't have time in the near future to build another amp. For the bipolar supply, nothing really changes. I'd leave the modulator with the virtual ground - the main reason for bipolar supplies is the power stage. It saves you bridging, so with the same design, you could drive 2 speakers independently (or you could save one half of the power stage). The speakers are then connected between the output for one half bridge (two fets with filter), and the center-tap of your power supply.
If you added another comparator for a second audio input, and drive the two IR2112s separately, you'd have a stereo amp without any effort.
The disadvantage is that the output amplitude is halfed. So for a 35V amplitude as I have it now, I'd need a -35V / 0 / +35V supply, which also means that the MOSFETs and the diodes have to withstand 70V. So you need less parts, but more expensive ones, since you are operating on higher voltages. Of course this also means more heat.
  Reply With Quote
Old 6th May 2004, 04:09 AM   #523
IVX is offline IVX  Russian Federation
diyAudio Member
 
IVX's Avatar
 
Join Date: Jul 2003
Location: Everywhere (Buddhist's context)
<* incredibly low output impedance/* load-independent>
:-) Really? How much? I've expecting about 25 damping factor for IDEAL power source and switching (3 meter of 1mm wire + 2*Rdson=.08ohm+.08ohm and no any kind of feedback). IMHO, THD on real speaker will be more 1% also. For sub applications 50-100khz switching is quite enough, and better for more efficiency.
__________________
Best regards,
E1.
  Reply With Quote
Old 6th May 2004, 05:01 AM   #524
sfx is offline sfx  Germany
diyAudio Member
 
Join Date: Feb 2004
Location: Australia
Hi IVX,

I probably shouldn't make claims I can't prove...

First, I don't understand your point about the MOSFETs. Yes, they have 50V breakdown voltage, and I drive them from a 35V DC unipolar power supply. The output voltage on the speaker is basically doubled by bridging. Just imagine a +/-17V bipolar supply with nothing connected to the center-tap. I wrote exactly this in my reply to Alcaid's questions.

(remark: oh cool, it is also possible to edit my own posts - should I remove this part now?!?)

For sub applications, a lower switching frequency would be sufficient. I wanted to have 250kHz to get some experience for a full range amp later. Since it works, I didn't change it anymore. Since I will get it out of my sub now anyway, to check some things, I might adjust it to 100kHz. The BCA principle virtually doubles this frequency anyway, so it still should be more than sufficient.

I can't comment on THD and impedance, since I didn't measure it. Just note that you don't have ANY problems or glitches at the zero crossing in BCA amps, and there is also no such thing as "deadtime". BCA amps are cleaner for low signals - less ripple, less distortion, etc. At higher amplitudes, distortion and ripple also grow with the amplitude, which is ok. At zero input, the BCA doesn't have ANY ripple at all, it cancels exactly out.

My outrageous claims regarding load independence came from some general views on Class D amps, partly from other people. It does make sense to me - the speaker is, in contrast to A or AB amps, not in a "variable resistor bridge", but the output voltage is mainly controlled by the duty cycles, and not really influenced by the speaker itself. In Class A amps, the voltage on the speaker depends very much on the resistance of the load (speaker), since the load is part of the "resistor bridge" that determines the voltage.
The overall "on" resistance of the switching output stage is very low, especially much lower than the resistance of the speaker. Also, inductive effects of the speaker/crossover/etc. can "flow off" over the freewheeling diodes. My conclusion is that the output voltage is more or less proportional to the duty cycle, and fairly independent of the load. At Class A Amps, it is not, and also at class AB amps, the load plays an important role.

This is all I can say about it - consider that I'm not an engineer, and I never learned how to calculate all this properly. Oh, and have you ever measured the THD of your speakers, or your room?
  Reply With Quote
Old 6th May 2004, 06:25 AM   #525
IVX is offline IVX  Russian Federation
diyAudio Member
 
IVX's Avatar
 
Join Date: Jul 2003
Location: Everywhere (Buddhist's context)
5 years ago i'd made similar amp but with feedback before filter, and achieved THD0.5%@100hz, damping factor about 200 and huge EMI around my DIYlab, which was partially reduced by many beads using. :-)
You know, just a right now i made measurement for my system+room_ambient THD@about10W... 22hz - 4.8% (vented box resonance) and higher 100hz less then 0.8%.

BTW, how much THD of the microphone i don't know..
__________________
Best regards,
E1.
  Reply With Quote
Old 6th May 2004, 07:55 AM   #526
Alcaid is offline Alcaid  Norway
diyAudio Member
 
Alcaid's Avatar
 
Join Date: Nov 2003
Location: Klepp
I don't have a scope, but would it work if I just adjust the VGnd to the exact value of V+/2 or do I need to check the waveforms and adjust till they are absolutely right?

The buying of a scope would kill the budget for this project
  Reply With Quote
Old 6th May 2004, 11:17 AM   #527
Alcaid is offline Alcaid  Norway
diyAudio Member
 
Alcaid's Avatar
 
Join Date: Nov 2003
Location: Klepp
Nice article if someone wants to learn more about class-D and output filters.

http://www.wictronic.ch/Downloads/CD...erformance.pdf
  Reply With Quote
Old 6th May 2004, 01:09 PM   #528
sfx is offline sfx  Germany
diyAudio Member
 
Join Date: Feb 2004
Location: Australia
Hi IVX,

now, your THD Measurements are impressive, I must say. Especially 4% at 20 Hz are astonishing. Unfortunately my line-in of my PC picks up a lot of crap below 50Hz, so it's difficult for me to measure - at 20Hz I already have 4% directly from line-out back to line-in....
Over the speakers, I got around 1% at 100Hz (Chain: line-out - Amplifier (Marantz), Orbid Sound Supernova Speakers, Behringer ECM8000 Mic, Behringer UB1202 Mixer, Line-In), and between 0.7 and 0.2% from 200Hz upwards. At 10kHz it was 0.02%. While measuring my subwoofer, I noticed that it produces some distortion at very low frequencies, probably because the membrane breaks up, and produces harmonics. I guess I need a new speaker some day - they are just so expensive. The one I have now has ripples in the membrane - crap for subsonic use. Well, for now it does the job, at least until I got 200$ spare...

Alcaid, if you don't have a scope, I guess you can also simply adjust VGnd to V+/2, that's at least what it's supposed to be. My initial idea for that pot was to adjust the waves, such that at zero input the modulator waveforms are exactly balanced. This is important, to make sure there is no DC going through the speaker at zero input. It turned out, that a) it was already quite well balanced, and b) the pot didn't have such a big influence on that. I guess it's not that critical for correct operation.
In general, a scope is very useful for Class D amps, to verify correct modulation before you connect the FETs. Maybe you can find someone who has one, and go there once for tuning. I don't have one myself either, but I can use the ones at university.
  Reply With Quote
Old 6th May 2004, 01:28 PM   #529
Alcaid is offline Alcaid  Norway
diyAudio Member
 
Alcaid's Avatar
 
Join Date: Nov 2003
Location: Klepp
Well, my school also have a lot of scopes, so I'll guess I could calibrate at school.

What if I adjust VGnd while measuring DC offset? That way I could trim for minimum output DC offset. Would that be a way to do it? I know, I'm trying to find shortcuts which doesn't exists....
  Reply With Quote
Old 6th May 2004, 08:58 PM   #530
Ryssen is online now Ryssen  Sweden
diyAudio Member
 
Ryssen's Avatar
 
Join Date: Dec 2003
Location: Sweden--> Here
Quote:
It saves you bridging, so with the same design, you could drive 2 speakers independently (or you could save one half of the power stage). The speakers are then connected between the output for one half bridge (two fets with filter), and the center-tap of your power supply.
So if I dont need all the outputpower (150 w)I can leave out the triangle inverter and use one IR2112 and 2 Buz11,would I get about 75 w/4 ohm then,
is it possible to go as low as 2 ohm´s,what would the output be then?Shoul I connect the speaker to ground or VGnd?
  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 06:07 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