What matrial for output filter?

lumanauw · 22nd February 2005, 09:11 AM

What material to be used for output filter? I built an experimental self-oscilating classD, making the output LC with ferrites (toroidal) it exhibit strange behavior. In small volume, everything is OK, but in medium and high volume, the power amp suddenly sucks big current, the mosfets becomes hot, and the output inductor seems ringing (can be heard, the ringing follows the bass). When checked in osciloscope, in the "on" pulse, there is a blurry area, like comb (in small volume the square is good, without oscilation). Is this because ferrite don't suitable for output LC? What material to be used?

Alme · 22nd February 2005, 02:38 PM

well, for the best quality they usually use coreless coil but they produce strong external EM field like antenna (you'll need some shield). Ferrite has quite low maximum flux density and high non-linearity, so you will need to choose large size cores and take feedback after LC-filter. I tried to use iron powder toroid cores with mu=80...90 but obtained very high core losses at calculated flux of about 0,5 tesla. Main frequency was 120kHz. Then I switched to special combined materials and got better results. The rule to choose material is simple, keep maximum flux below 0,2T; the higher is your carrier frequency, the smaller mu parameter the core must have. For that 120kHz I found good material with mu=36

Alme · 22nd February 2005, 02:40 PM

sorry, I wanted to add some links where you can learn this stuff and choose proper core. I found mine at
http://www.emicore.co.kr/home-e.html
also good information is here:
http://www.micrometals.com/

Alex

phase_accurate · 22nd February 2005, 02:55 PM

I once used a gapped (by the insertion of insulating material) RM ferrite core like Lars does on his one. A gap on the center leg alone would be better though, but not all cores are available with air gap in small quantities:

http://www.lcaudio.com/zp22-1.jpg

I once posted the coarse formulae for calculating air gaps on this forum.

Regards

Charles

Jaka Racman · 22nd February 2005, 07:32 PM

Hi lumanauw,

Your problem is saturation of the core. Ungapped ferrite has high permeability, so magnetic flux density becomes high with small amount of ampere turns. You neeed low permeability core, either gapped ferrite core that is more linear, or distributed gap core like powdered iron or pemalloy. Ungapped ferrite toroid will exhibit high inductance until it comes into saturation when it effectively becomes an air core. So you are facing several thousand times reduction in inductance (difference between initial ferrite permeability and air permeability) or in your case from several miliHenrys to several hundred nanoHenrys. Typical inductance used in classD amps is usually from 10 to 40uH. You were given a lot of good advice from orher posters, so maybe only two things:

Remember that most of the energy in inductor is stored in the air gap, so larger the gap, more energy storage is available (and more problems with fringing flux).

For experimenting, you can dismantle PC power supply. Usually there is one large toroidal powdered core inside, used as secondary coupled inductor with many turns of various diameter magnet wire on it. I have never measured such a core, but my feeling is that it might be used for medium power amplifier if you rewind it and it it has mu less than 100.

Best regards,

Jaka Racman

subwo1 · 22nd February 2005, 09:12 PM

In addition to the others' ideas, the sound you hear could be an exhibition of magnetorestriction. The physical size of the core actually changes with the magnetic field, causing noise.

A general guideline is that if a DC current is passing through a core, a distributed gap type like iron powder or a gapped ferrite type is required to prevent saturation. Even though the switching frequency may be high in the kilohertz range, the audio component is passing through in a way not totally unlike DC through a power supply filter choke.

lumanauw · 23rd February 2005, 01:01 AM

Thanks for everybody's input. I got a picture now, the core is saturating. I'll look for other core/material that only gives little uH with many turns.

Is toroidal shape is a problem here or the material itself?

I look at Crown BCA, it uses toroidal shape, but it has many-many windings in it (like 100 or so). In my experimental amp with ferrite toroidal, with 12turns only, I've already got 200uH. So I think the saturation could be the cause.

If the core is saturating in my case, why it makes the mosfets hot and draws so much current? Does "Shoot Through" happens with saturated case, or the hot mosfets are caused by something else? I've burned many-many mosfets. (I use IRF640, is it something wrong with using this?)

I take the feedback after LC filter.
Since I experimented with self-oscilating, it's difficult for me to see what happens in the fault condition. In the fault condition, every track seen in the scopes are mess and moving everytime.

From magnetics, there is a material called "Kool Mu". Is this suitable for this LC filter (because ALME said powdered metal cores are not suitable)

lumanauw · 23rd February 2005, 01:58 AM

This is the inductor I use. I stack 2 toroidal ferrites

lumanauw · 23rd February 2005, 02:01 AM

This is the gate voltage when the volume is low. It's clean square.

lumanauw · 23rd February 2005, 02:05 AM

This is the gate voltage with medium level. In the "on" pulses, there are oscilation

22nd February 2005, 09:11 AM	#1
lumanauw diyAudio Moderator Emeritus Join Date: Oct 2002 Location: Bandung	What matrial for output filter? What material to be used for output filter? I built an experimental self-oscilating classD, making the output LC with ferrites (toroidal) it exhibit strange behavior. In small volume, everything is OK, but in medium and high volume, the power amp suddenly sucks big current, the mosfets becomes hot, and the output inductor seems ringing (can be heard, the ringing follows the bass). When checked in osciloscope, in the "on" pulse, there is a blurry area, like comb (in small volume the square is good, without oscilation). Is this because ferrite don't suitable for output LC? What material to be used?

22nd February 2005, 09:12 PM	#6
subwo1 diyAudio Member Join Date: Jan 2002 Location: North American Continent	In addition to the others' ideas, the sound you hear could be an exhibition of magnetorestriction. The physical size of the core actually changes with the magnetic field, causing noise. A general guideline is that if a DC current is passing through a core, a distributed gap type like iron powder or a gapped ferrite type is required to prevent saturation. Even though the switching frequency may be high in the kilohertz range, the audio component is passing through in a way not totally unlike DC through a power supply filter choke. __________________ USMPS http://groups.yahoo.com/group/switchmode/

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Output filter	Skorpio	Class D	5	8th August 2008 06:44 PM
What output filter to choose	Groundloops	Digital Source	2	9th November 2007 01:06 PM
Output filter	m-tech	Class D	11	22nd September 2007 07:04 AM
BTL output filter	ted12	Class D	16	26th April 2005 01:55 PM
output LR filter	Helix	Solid State	7	8th February 2002 12:08 PM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	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

22nd February 2005, 02:38 PM	#2
Alme diyAudio Member Join Date: Jan 2005 Location: in Korea, presently	well, for the best quality they usually use coreless coil but they produce strong external EM field like antenna (you'll need some shield). Ferrite has quite low maximum flux density and high non-linearity, so you will need to choose large size cores and take feedback after LC-filter. I tried to use iron powder toroid cores with mu=80...90 but obtained very high core losses at calculated flux of about 0,5 tesla. Main frequency was 120kHz. Then I switched to special combined materials and got better results. The rule to choose material is simple, keep maximum flux below 0,2T; the higher is your carrier frequency, the smaller mu parameter the core must have. For that 120kHz I found good material with mu=36

22nd February 2005, 02:40 PM	#3
Alme diyAudio Member Join Date: Jan 2005 Location: in Korea, presently	sorry, I wanted to add some links where you can learn this stuff and choose proper core. I found mine at http://www.emicore.co.kr/home-e.html also good information is here: http://www.micrometals.com/ Alex

22nd February 2005, 02:55 PM	#4
phase_accurate diyAudio Member Join Date: May 2002 Location: Switzerland	I once used a gapped (by the insertion of insulating material) RM ferrite core like Lars does on his one. A gap on the center leg alone would be better though, but not all cores are available with air gap in small quantities: http://www.lcaudio.com/zp22-1.jpg I once posted the coarse formulae for calculating air gaps on this forum. Regards Charles

22nd February 2005, 07:32 PM	#5
Jaka Racman diyAudio Member Join Date: Dec 2002 Location: Slovenia	Hi lumanauw, Your problem is saturation of the core. Ungapped ferrite has high permeability, so magnetic flux density becomes high with small amount of ampere turns. You neeed low permeability core, either gapped ferrite core that is more linear, or distributed gap core like powdered iron or pemalloy. Ungapped ferrite toroid will exhibit high inductance until it comes into saturation when it effectively becomes an air core. So you are facing several thousand times reduction in inductance (difference between initial ferrite permeability and air permeability) or in your case from several miliHenrys to several hundred nanoHenrys. Typical inductance used in classD amps is usually from 10 to 40uH. You were given a lot of good advice from orher posters, so maybe only two things: Remember that most of the energy in inductor is stored in the air gap, so larger the gap, more energy storage is available (and more problems with fringing flux). For experimenting, you can dismantle PC power supply. Usually there is one large toroidal powdered core inside, used as secondary coupled inductor with many turns of various diameter magnet wire on it. I have never measured such a core, but my feeling is that it might be used for medium power amplifier if you rewind it and it it has mu less than 100. Best regards, Jaka Racman

23rd February 2005, 01:01 AM	#7
lumanauw diyAudio Moderator Emeritus Join Date: Oct 2002 Location: Bandung	Thanks for everybody's input. I got a picture now, the core is saturating. I'll look for other core/material that only gives little uH with many turns. Is toroidal shape is a problem here or the material itself? I look at Crown BCA, it uses toroidal shape, but it has many-many windings in it (like 100 or so). In my experimental amp with ferrite toroidal, with 12turns only, I've already got 200uH. So I think the saturation could be the cause. If the core is saturating in my case, why it makes the mosfets hot and draws so much current? Does "Shoot Through" happens with saturated case, or the hot mosfets are caused by something else? I've burned many-many mosfets. (I use IRF640, is it something wrong with using this?) I take the feedback after LC filter. Since I experimented with self-oscilating, it's difficult for me to see what happens in the fault condition. In the fault condition, every track seen in the scopes are mess and moving everytime. From magnetics, there is a material called "Kool Mu". Is this suitable for this LC filter (because ALME said powdered metal cores are not suitable)