Design of output inductor for class D amplifier

Status
Not open for further replies.
Jump to Latest
#104
SSassen said:


I guess that was audiophile enough for you then? Shall we try to keep things ontopic from now on? Please? :xeye:

Best regards,

Sander Sassen
http://www.hardwareanalysis.com
Click to expand...

Let's look at this part of a Tripath amp for example. How would changing the output inductor improve the class D amplifier performance. People have previously posted core material parameters and windings. Does it make sense as a better output inductor?
 

Attachments

  • d out.jpg
    d out.jpg
    29.3 KB · Views: 709
#105
I'd put an additional "RF filter" after the main output filter in order to improve on RFI performance.
Remember that a simple LC filter is just a proper 2nd order lowpas IN THEORY.

I don't think that Lars' modulator is very susceptible to RF either. AFAIK it is a self-oscillating topology. Carrier-based amps are more susceptible to RF glitches (and restricted in terms of THD performance due to carrier residual).

And back to topic: Winding schemes that reduce the coil's parasitic capacitance will also improve the RF behaviour. One such scheme has already been mentioned by Lars.

Regards

Charles
 
#106
The Tripath amp (from your posted section of the schematic) appears to use only pre-filter NFB. The output filter is therefore not part of the feedback path, so any problems with core non-linearities may have a measurable and audible effect on the output.
 
#107
Charles,

And back to topic: Winding schemes that reduce the coil's parasitic capacitance will also improve the RF behaviour. One such scheme has already been mentioned by Lars.
Click to expand...

Indeed, one other way is to compact the windings, so there's more windings per mm^2 and hence the inductance of the coil rises, thus we need fewer windings to end up with the same value in mH. Obviously the best way to accomplish this is to use copper foil, rather than copper wire. Hence a copper foil coil will allow you to wind it tighter and hence the coil will be more compact than one wound with copper wire.

As for the coil's capacitance dominating over its inductance at RF frequencies, this cannot be avoided, all we can aim to do is push the frequency at which the coil starts to exhibit this behaviour as far upward in frequency as possible. Hence keeping the parasitic capacitance low is a means to an end, yet as I mentioned we'd best forget about copper wire and use copper foil instead.

Best regards,

Sander Sassen
http://www.hardwareanalysis.com
 
#108
phase_accurate said:


And back to topic: Winding schemes that reduce the coil's parasitic capacitance will also improve the RF behaviour. One such scheme has already been mentioned by Lars.

Regards

Charles
Click to expand...

This is also covered in another thread which brings up the "banked winding" concept that reduces parasitic capacitance even more.

However, the new cored caused reduction in the necessary winding that probably made banked winding unnecessary. Has anyone actually measured this?
 
#109
Ouroboros said:
The Tripath amp (from your posted section of the schematic) appears to use only pre-filter NFB. The output filter is therefore not part of the feedback path, so any problems with core non-linearities may have a measurable and audible effect on the output.
Click to expand...

Tripah engineers said the system would be unstable with post filter feedback. Even if post filter feedback was implemented, measureable and audible effects would still be there.
 
#111
:cop: :cop: :cop: :cop: :cop:

Gentlemen, please calm down. I don't know what it is with switching amplifiers, but the Class D forum currently seems to need as much moderator attention as the rest of the forums put together. Attack one another's arguments by all means, but not personalities. If it is your personal opinion that someone is an idiot who will never learn, then publicly saying so will not change them, and only sours the forum. If things don't calm down, I will be handing out extended Sinbin sentences freely, and banning will follow. You have ALL been warned.

:cop: :cop: :cop: :cop: :cop:
 
#112
SSassen said:
Lars,



I disagree, you're just inserting another passive pole in the loop, in the feedback loop in case of post filter feedback. Either way this is a solution that seems to stem from a misunderstanding of the fact that a properly calculated control loop will behave properly with RF too, provided you have a proper layout and a properly wound coil. I hate to bring it up again as it makes me sound like a fanboy, but Bruno seems to be able to manage this quite well.

Best regards,

Sander Sassen
http://www.hardwareanalysis.com
Click to expand...


This solution was proposed and is used by SONY in their S-Master amplifier.

Best regards

Lars Clausen
 
#115
Lars Clausen said:



This is a common misunderstanding. The pre / post filter scheme has absolutely no bearing on PSRR. However open / closed loop has.

One way to get rid of the RF transmission problem in the coil, is to place a small open aircoil to 'eat' the high slopes before they hit the real filter coil. This way you can wind a normal simple filter coil without considering the RF properties.


2w.gif
Click to expand...

I'm not sure how effective that would be, but it could certainly be simulated in spice if all spice models were available (a problem I constantly run into). Looks quite valid approach at a glance.
 
#120
What size cores are you guys looking for? I've already got some T106 MPP and Sendust on order (check the TI TAS5261 thread for more details), and if they look good when they come in I may order some more. I could even do a little group buy if anyone else is interested. T106 MPP are $3 each and T106 Sendust are $1.50 each from CWS.

On topic:

SSassen said:
Obviously the best way to accomplish this is to use copper foil, rather than copper wire. Hence a copper foil coil will allow you to wind it tighter and hence the coil will be more compact than one wound with copper wire.
Click to expand...

If copper foil is better, why not use copper foil with cores like Lars is using? The only difficulty I see with that is getting the foil connection out from the inside.
 
Status
Not open for further replies.
Top Bottom