Need some help from you class D experts - 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 29th July 2004, 06:41 PM   #1
diyAudio Member
 
Join Date: Jun 2004
Location: Knoxville
Default Need some help from you class D experts

Part of this circuit that I am making resembles a class D circuit using a Mosfet device to drive a pulse of current into a load.

I am getting an occilation on the output voltage spike, when the device is driving current. I am sure I need a choke or an inductor some place in this circuit but I am not sure where excactly and of what value on account of I do not have much experience with this type of circuit topology. The circuit diagram and waveform sketches may help...
Attached Images
File Type: jpg d-amp.jpg (54.8 KB, 203 views)
  Reply With Quote
Old 29th July 2004, 06:45 PM   #2
diyAudio Member
 
Join Date: Jun 2004
Location: Knoxville
I think that this is where I should place the inductor, but not certain. BTW: The voltage switches from 0V to 50V.




Thanks alot!!


Chris
Attached Images
File Type: jpg d-amp.jpg (55.2 KB, 150 views)
  Reply With Quote
Old 29th July 2004, 08:26 PM   #3
diyAudio Member
 
Join Date: Dec 2003
Location: Munich
...uhps!.... sorry I am no class D specialist...
In fact, I also do not fully understand your circuit and I have no clear indication for this oscillation.
But may be my brainstorming can help....
Do you see a oscillation also at the 50V/0V signal?
On the DC-rail? Do you supply this circuit from
SMPS which operates at 50kHz?
Do you have a current probe and can find the
loop in which the current is oscillating?

But there is one thing, which is critical IMHO.
You are supplying only positive pulses to the speaker.
....means, output signal contains a serious amount of DC.
Do you do this on purpose?

Bye
Markus
  Reply With Quote
Old 29th July 2004, 09:36 PM   #4
subwo1 is offline subwo1  United States
diyAudio Member
 
subwo1's Avatar
 
Join Date: Jan 2002
Location: North American Continent
Possibly, the oscillations are related the confluence of such factors as the MOSFET directly driving a capacitor referenced to ground potential, the MOSFET's gate capacitance and other internal capacitances, and the 1.5k gate resistor. If you lower the gate resistor, does the frequency of the suboscillations increase? At the very least, a pair of output MOSFETs should be included instead of just one and the output fiter inductor should go between MOSFETs and output filter capacitors.

If you have only a single polarity power suppy, you should couple the speaker to the output through a large DC blocking capacitor as was done in audio amps in the old days. A 4700uF 63v one might work for a 4 ohm speaker load.

You could theoretically get by with the 4 ohm resistor replacing a lower MOSFET. But you should still place the 2.2uf capacitor after the inductor, and you should use the large DC blocking capacitor in the case of single rail power suppy use.
__________________
USMPS
http://groups.yahoo.com/group/switchmode/
  Reply With Quote
Old 29th July 2004, 11:46 PM   #5
diyAudio Member
 
Join Date: Jun 2004
Location: Knoxville
Quote:
Originally posted by ChocoHolic
...
But there is one thing, which is critical IMHO.
You are supplying only positive pulses to the speaker.
....means, output signal contains a serious amount of DC.
Do you do this on purpose?

Bye
Markus
Sorry for the misunderstanding, This circuit is a secondary current source for the output transistor and this is only the positive one, the negative side is an equal but opposite circuit. Zout resistor represents the output impedence of the amp.
The output stage is an emitter follower, drivers use +/- 50V reg.

The theory of operation is that I would like to push the outputs a little harder to get the peak and spikes of a full range audio signal and not exceed the breakdown of the "off" transistor, all using just two hi-current output devices. (and 2 Mosfet switches) I am using two common grounded power supplies (60Hz power transformers), one is +/- 36V @2A, and the other is +/- 63V @ 1.5A, with 15,000uF filter caps in order to handle the duty cycle. The circuit I have drawn drives just the NPN output with 50V. Switching diodes isolate the lower V power supply from the higher V one.
In other words, there is 35V on each output, until the signal is +16V or higher, then this circuit switches +50V onto the collector. (I may increase 16V this to 20V though) The PNP device still only has 35V on its collector. When the signal is -16V(or 20V) the negative circuit operates, -50V on the collector of PNP, and there would be only +35V on the collector of NPN. The Mosfet switches are never on at the same time, and this makes sure that 90V is not exceeded ever. I do not want to paralell output devices, just trying to push the envelope and still have plenty of Vce to limit nonlinearities. These devices are linear up to 7.1A. 28V peak would be 100RMS watts on 4Ohms.

When the Fet's switch on, there is NO effect on the audio signal, even at higher freq. But the occilations do affect it slightly.

This is not a common approach by amp designers, but it does seem to work well.

I know that this is kinda complicated, but it is interesting!

Chris
  Reply With Quote
Old 29th July 2004, 11:51 PM   #6
diyAudio Member
 
Join Date: Jun 2004
Location: Knoxville
Subwo1,

If not for the 2.2uF cap to ground, the occilations are much larger even when turned on with no current flowing.

I shall try to lower the value of the gate resistor and see if that changes anything.

any ideas on the inductor?
  Reply With Quote
Old 30th July 2004, 12:38 AM   #7
subwo1 is offline subwo1  United States
diyAudio Member
 
subwo1's Avatar
 
Join Date: Jan 2002
Location: North American Continent
For the inductor, try moving it between the outputs of the MOSFETs and speaker. The capacitor should be in parallel with the speaker.
__________________
USMPS
http://groups.yahoo.com/group/switchmode/
  Reply With Quote
Old 30th July 2004, 12:15 PM   #8
diyAudio Member
 
Circlotron's Avatar
 
Join Date: Jun 2002
Location: Melbourne, Australia
Bottom transistor turns on, pulls mosfet gate low (on), mosfet drain goes high, turns on npn transistor at very left, turns on pnp transistor at top which tries to short out gate drive and turn off mosfet. So, first and last transistor are having a fight. The time taken for the signal to travel around that loop kinda determines the frequency of your oscillation.
__________________
Best-ever T/S parameter spreadsheet.
http://www.diyaudio.com/forums/multi...tml#post353269
  Reply With Quote
Old 30th July 2004, 03:55 PM   #9
diyAudio Member
 
Join Date: Dec 2003
Location: Orange County, CA
Default Circuit

This circuit is horribly ineffficient because the outout can only approach within about 10V or so of the supply. The FET is dropping at least 10V when turned on and is likely getting very hot, wiping out most of the advantages of class-f operation. That is why we use driver ICs like the IR2110 which drive the FET with a bootstrap circuit that supplies drive about 12V above the supply rail.

You must design the circuit so that the FET saturates. Otherwise, it will run very hot. As well, the top of the PWM output will be variable with transistor temperature.

Also, there is no supply rejection at all. Anything that happens on the power supply is passed on directly to the audio. The power supply has to be very well regulated.

Overall, this is a very primitive design. There are better circuits at www.schematicsforfree.mattsoft.net
__________________
Dan Fraser
  Reply With Quote
Old 31st July 2004, 06:16 PM   #10
diyAudio Member
 
Join Date: Jun 2004
Location: Knoxville
Thanks for the info guys, it is very helpful.


Dan, the drain of the FET is the supply rails for the higher voltage current source for the collector, and is referenced to 50V. True that the FET is dissapating lots of heat, but since it is only turned on when the audio signal is above 20V, the duty cycle of this reg. circuit is less than 20%. Besides, I am expecting the higher supply voltage to ripple, this reducing the heat even more, but keeping 50V on the collector all the same, just when the signal is above 20V. This ripple will not affect the rest of my circuit unless it gets below 55V. The negative circuit is equal but opposite. With real full range audio, the spikes and peaks exist usually for a short time span anyway, more like a pulse circuit.
  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
4th BP experts? danielm Subwoofers 4 20th July 2009 04:47 PM
DHT experts, i need your help Cassiel Tubes / Valves 10 25th February 2009 01:42 PM
Any TL experts out there? jmiyake Subwoofers 8 8th September 2006 03:17 AM
advice from the experts notaclue Multi-Way 6 11th July 2006 10:24 PM
Need help from the experts bateman Multi-Way 1 22nd June 2003 02:29 AM


New To Site? Need Help?

All times are GMT. The time now is 06:59 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