Tripath Output Filter

[serengetiplains]

Seems this topic, which can apply to any Tripath-based module, needs its own thread. Discussion items include:

* What inductor type is best?

* What capacitor types are best?

* Will removing output filter capacitors improve or degrade performance, and under what circumstances?

Regarding the latter question, I'm bi-amping a pair of small speakers and would think any extra RF sputtered into my 5" woofers (don't know about the tweeters) would be inaudible ... unless, of course, the RF mucked up the input signal or psu rail.

[LineSource]

* What inductor type is best?

10 Gauge copper foil with a solid wood center dowel.

Parts express has 12 gauge foil inductors for speakers that you experiment with.

[cotdt]

is aircore inductors a good option?

[serengetiplains]

Cotdt, air cores won't saturate, and saturation causes distortion. But air cores are larger and radiate more EMI. My preference is for air cores. My Charlizes use air cores, and they sound excellent.

[Pano]

Thanks, Tom, for starting the thread.

As for those 12 ga. foil coils over at Parts Express, the smallest value they have is 0.05mH The Tripath filters usually use 0.01 (10uH) or 0.022 (22uH) So you'd have to unwind and throw away 4/5th of your coil. Then you'd have to deal with all the RFI issues that aircore coils have too.

No doubt the big foil coils sound good in a crossover, I have used and like the Goertz coils. A friend of mine is now using 8 gauge coils on his Altec 605s, he claims they improve the midrange of high efficiency speakers. He wants 6 gauge, but no one will make them for him!

Back to the subject. Aircore inductors are nice because there is no ferrite to saturate as the copper is supplying all the inductance. Aircoils are usually very low distortion, a good thing.

But in the Class-T (Class-D) amps, air coils radiate a lot of RFI. If you can contain it, then OK. But it is easy enough to get the RFI into signal lines - does not sound good.

My RF guru, Dennis, led me to some toroid cores that are designed to work in the frequency rage of Tripath switching - 600 kHz to 1MHz (harmonics out to 3MHz). Theses cores work very well at reducing RF -12 dB or better over the stock sonic coils, at the same 10uH value. That rocks! They seem to have very low audio distortion too, as LostCause can testify, he has some in his amp.

In my next post I'll talk a little more about inductors
A buddy of mine will be running distortion tests on several coil types next week. It will be interesting to see if he finds significant differences.

[cotdt]

where can we buy high quality winded toroids? i'm looking for 11mH, right now I'm using a 20 gauge one that I wound myself, but I don't want to go through that again...

[Pano]

I have a batch coming in this week. Contact me for more information. Very nice, low distortion, low RFI, with silver coated teflon insulated wire.

[audio1st]

Are these the same as in the T Amp?http://uk.farnell.com/jsp/endecaSear...=3227169&N=401

[Pano]

Originally posted by audio1st
Are these the same as in the T Amp?

Depends which T-Amp you mean. I don't think the coils used by Sonic are these, I think they may be Toyo on the Sonic.

The Fenice amp uses Panasonic coils like in your link, however the Fenice uses 22uH instead of 10uH.

I think the stock coils in the Sonic sound pretty awful, and the Panasonics on the Fenice are OK. But replacing the Panasonic inductors with good quality toroids reeeealy helps smooth things out. Silky smooth. Aircore coils should have the same effect, with the risk of much higher RFI.