"Polyamping" A Look to the Future or Fancy Fad?


In a recent quest for information regarding DIY speaker designs, I was referred to the Linkwitz Orion Project. These speakers employ active crossovers and it is suggested to give each driver its own, separate amplification (actually one for each woofer and one for the tweet/mid - three per speaker). Linkwitz recommends the ATI AT6012, a twelve channel, six zone amp (60W/ch). I am not sure about the merits of the ATI amp but, regardless of amp, does anyone think this will be a "growing" design. I mean I have heard the benefits of biamping and have heard tell of triamping but, in this case, "sextamping"? Octamping would seem to be next. All accounts say that the Orions sound fabulous. Perhaps I am just behind the curve. What so you learned folks think of this direction in audio?
4yanx

Showing 2 responses by sean

Mezmo: While i appreciate the vote of confidence, don't encourage me. I already open my mouth too much as it is. Just ask Brulee....

4Yanx: What is good for the goose is not always good for the gander i.e. personal preference rules. Once you've heard "commanding" bass with impact, definition, authority AND control, you'll know where Greg and others like us come from. You might be amazed at how much power high amplitude reproduction of low frequencies actually takes to do "right". This is especially true if you have low efficiency speakers.

To try and explain this simply, the more excursion that a driver has to make, the greater the distortion and the poorer the transient response. As such, using multiple drivers allows one to move as much air without any individual driver ever having to make much of an excursion. They are all sharing the load rather than having to deal with the physical / electrical demands individually.

This is the approach that Dr Bose took with the original 901's and Bill Duddleston likes to promote in the literature of Legacy products. While such an approach is very valid, you now run into problems with impedances and high levels of reflected EMF*. As far as dealing with low impedances and higher levels of reflected EMF, it takes "muscle" to deliver the power required to control all of the drivers and deal with the reflected EMF.

Bob Carver does a good job of explaining this in the white papers of his Sunfire subs and i encourage those interested in learning about such things to give those a read. He tries to keep things simple yet get the point across. You don't have to like the product to understad that there is a a lot of valid research behind it. As is so often the case, the good intentions and valid research get lost somewhere along the production lines.

To those that that think that "high power" is unnecessary, i'll use the same analogy that i use at work all the time.

If you've been walking all of your life with no other means of travel, a bicycle seems great. When you start driving a Yugo, the Yugo makes the bicycle seem antiquated and slow. When you start driving Lincoln's, Cadillac's, Mercede's, BMW, Porsche, Ferrari, etc... everything else prior to that point seems like a joke. You look back and wonder why it took you so long to understand your current vantage point and realize how much better it could have been and / or how much you've missed along the way.

While we call this "the learning curve", it is all a matter of perspective, experience and personal preference. As such, "good" results to Billy Bob might be "piss poor & pathetic" to Ray-Ray : ) Sean
>

*This is not to mention a slew of other problems that are related with acoustics / room loading, but that is a whole 'nother ball of wax.

PS... i was able to mention Bose, Legacy and Carver all in the same post in a positive manner. That must be a first for Audiogon : )
I think that such an idea is great. Having said that, i would emphatically state that 60 wpc is not enough power to achieve quality reproduction of low frequencies in my opinion based on quite a bit of experience in this area.

With that in mind, i've done something very similar but gone to an extreme in the method that i chose. To try to sum things up briefly, i am actively tri-amping with six stereo amplifiers. The amps are set up to operate in what is effectively a monoblock format even though all 12 channels are in use. While i did so for multiple reasons, my primary goals were to increase dynamic headroom, achieve the ultimate in stereo separation and to spread out what is a very tough i.e very reactive AND very low impedance speaker load over multiple amplifier channels.

With the average impedance of the system hovering around 2 - 3 ohms per amplifer channel, i've got 2400+ wpc ( using the various amplifiers 4 ohm power ratings ) as things are currently arranged. I would realistically rate output capacity as somewhere around 3 KW rms per channel at the actual impedances that the amps see.

Using this approach, none of the individual amplifier channels are "pushed" as hard as they would be if i had them set up in a more conventional manner. On top of this, damping factor is doubled and there is no chance of ever clipping. While the amplification system has worked most excellently once i was able to get things dialed in, i just need to find a way to quit demolishing specific sections of the speakers. I've still got a ways to go on this, but it's pretty much been both a "labour of love" and a learning experience at the same time : ) Sean
>