[cab]

Hello-

I am wondering if anyones knows if the output filter in these modules is inside or outside of the feedback loop?

Thanks,

[audiojerry]

Is the output filter used in the amp module inside (a la Spectron, Icepower) or outside (Tripath, PS Audio) of the feedback loop?
Thanks....

Why is this significant, Cab?

I believe Kevin mentioned somewhere that he would be unavailable for a time.

[Kevin P]

If you look at the diagram in the LC Audo Cookbook it shows the analog feedback loop from between the two mosfets to the input of the integrator.  The output filter using the large ferrite choke and blackgate capactitors is outside of the feedback loop.

http://www.lightball.dk/download/com/pdf/zp21mateng.pdf

[cab]

Thanks for answering.

The location of the output filter in relation to the feedback loop is important as it has a strong influence on how the amp performs across the frequency spectrum, and how it performs with load impedances. Output filters outside of the feedback loop are optimized for one load impedance, usually 4 or 8 ohms. Variations away from that design point result in less than optimal performance. This is one reason why digital amps like the tripath, which also has its filter outside the feedback loop, have difficulty with low impedance loads, and do not sound their best with "difficult" speakers.

Also, when the output filter is outside the feedback loop, we tend to see rising distortion with frequency, and rising distortion with power level. The output impedance also rises with frequency.

Spectron and Icepower designs place the output filter within the feedback loop, and thus avoid these problems.

[_scotty_]

cab, I think you ascribing a greater importantance to the location of the output filter with regards to the feedback loop than is warranted.  His amplifiers' power supply design,ground return paths, freedom from ringing,stability margins,and behavior under overload conditions also contribute to its' sound.      So far his amplifier has been well received by those who have listened to it.  I think raising doubts about its sonic performance before hearing it is a little premature. His initial offering is also more financially accessible to people than the alternatives you listed in your post.

[cab]

I am talking about design and performance issues only; how it sounds is up to each person to decide for themselves.

Keep in mind though that users that report that they are not satisfied with the high frequency performance may in fact be responding to the design limitations inherent in class d amps of this type when used with loads that vary greatly from the filter design impedence. The same issues have been raised by listeners of the carver and other tripath based amps when used with difficult loads. The PS Audio amp review raised eyebrows as well with its meaured performance, yet there are many who are thrilled with the sound.


In any case, it is a limitation of the design, and people should be aware of that; whether the effects of this are audible in their system or not is something all who are considering this (or any other amp) should check for themselves.  That is the bottom line....

[KevinW]

I am talking about design and performance issues only; how it sounds is up to each person to decide for themselves.
 ...

Precisely... this is why there is a return guarantee on the Fusion amp.  

I agree with Scotty in saying that you are being a little harsh in detracting from this design based one technological aspect.  There is so much more to the sound quality than just the feedback design.

[Kevin P]

I've had Dan Wiggins do some testing on the modules and he seems to be happy with their performance into 2 Ohm loads.  In fact our primary use for a couple of the modules will be into 2 Ohm reactive loads.

We also have had TREMENDOUS positive feedback from electrostatic owners who tend to have difficult loads to drive.

I think the issue described is more likely to do with the low input impedance and the resulting mis-match with a preamp.

[cab]

My apologies if I have come across as "harsh", as that was surely not my intention. I was simply trying to address the facts as they pertain to this amp design.

I agree that there is much more to sound quality than the feedback design, but on the other hand, with these class d amps, the final results are in fact greatly dependent on this particular part of their design. In fact, several class d designers have said that the final sound of a class d amp is heavily dependent on the both the type of output filter (air core is best) and its implementation. It is THE critical component. Have you tried replacing the ferrite core inductors with air core? You may be very pleased with the results....

There is no escaping the fact that distortion with frequency and power output climbs significantly, as does output impedance. One does not see this at these levels in the designs which keep the filter in the loop. It is also a fact that the output filter must be designed to a point load impedance, and this is less than optimal when using a load which is variable. This is all simple science. Is it audible or not? Well, there have indeed been people less than satisfied with the HF response of other amps of similar design, so for some it is clearly an issue. Will it be an issue with this amp? Time will tell.

Since the manufacturer is generously providing a 30 day trial period, the potential users are free to decide for themselves whether or not the above behavior is audibly significant for themselves. I light of the positive reviews many similar class d amps have gained, I am sure this amp will too be enjoyed by many.

Best of luck to Kevin...