How do high-end amplifier designers view Class D technology? Are Class D switching amplifiers really up to the standards of the high end? To answer these and other questions, we brought together three preeminent engineers in the field: Dan D’Agostino, founder of Krell Industries and its lead design engineer; Jeff Rowland, founder of Jeff Rowland Design Group and its principle designer; and Bruno Putzeys, designer of Class D modules and the man behind the Kharma MP150 amplifier reviewed in this issue.
Robert Harley: Class D switching amplifiers—once relegated to powered subwoofers, car stereos, and cheap home-theater-in-a-box units—have moved into the mainstream of high-end audio. Is this push toward Class D driven by sound quality or convenience? Are we trading high fidelity for the functional advantages of Class D, such as efficiency, low heat dissipation, and small form factor?
Jeff Rowland: Through the '90s we were looking at what the next step in amplification would be, and examined all kinds of different topologies. I approached it with a completely open mind. We looked at a number of platforms, including TriPath and other companies who offered their own PWM modules.
I ran into a company by accident at CES called ICEpower, and thought I’d give it a try. I looked at the module as a building block, the way any designer would look at different transistors or discrete components. It’s fundamentally a power-conversion technology. I tried some of the first-generation modules and was initially impressed by their performance.
After some research, I concluded that the technology was getting to be quite mature. I started adding my own value-added components and techniques and found that the overall performance I could achieve was quite astounding. It fulfilled the promise of higher performance, but also had the advantages of smaller size and higher efficiency. It’s also a “green” technology. We achieved increased performance over what could be done in the traditional linear realm.
Dan D’Agostino: We’ve tried all those switching modules because I’ve always wanted to build a very, very large power amplifier for a subwoofer with that technology. Our sonic experience with switching amplifiers is significantly different from what Jeff reports. We can’t get anything in the switching domain to sound anything like what we build in the linear domain, nor have the power, the presence, the staging, or any of those things. We attribute this to the fact that the technology isn’t finished—the devices aren’t fast enough and the filter technology is not ideal for every speaker system that’s out there. We basically have come to a conclusion that switching technology doesn’t even remotely approach what we’re doing in the linear domain.
Bruno Putzeys: In the high-end market, the push is certainly driven by sonics. I see it happening in two phases. The initial vanguard of high-end Class D amps, which includes several amplifiers that are still very popular, weren’t mature and sounded very different from what we’re used to from good linear designs. They caught people’s attention by virtue of sounding different, not necessarily better. Until recently, no switching amplifier had decently low output-impedance or reasonable distortion figures at higher frequencies.
Ironically, this gave them a euphonic edge over amplifiers that behaved correctly, in much the same way that tubed amplifiers continue to make an impression on people. These firstgeneration Class D amplifiers impressed by their artifacts, which, by chance, happened to sound pleasing. Although this helped establish a presence in the high-end market for this technology, I’m happy to say that we’re now at the second stage, where Class D amplifiers can satisfy the truly discerning listener and not just the novelty-seeker.
Top | Previous | Next
Advertisement
All documents notede with file size are in PDF format.
Download Acrobat Reader
