Harley: How much influence can the designer have over switching-amplifier sound using the same module?
Rowland: We don’t make our own power transistors; we don’t make our own capacitors. I look at the module as a component I’m buying to which I’m adding my value by applying my experience with input circuitry, power-supply technology, resonance control, and other techniques to add value and performance to an off-the-shelf component. I treat the module as a component to which I’m adding better parts and design techniques to get the finished result of an amplifier that realizes my goal of what an amplifier should be.
Harley: So you’re saying there’s a huge variation in the implementation of the same modules?
Rowland: There seems to be. There are a number of different modules out there, and many amplifiers are simply those modules put in a box. But you can go much further than that. There have been several thousand man-hours of research put into the ICEpower module—this was a research project for a number of years at the Technical University of Denmark. I’m basically just taking advantage of the work that has already been done. It’s got a lot of math that’s way over my head as an electronics designer; I’m just riding on the back of the work that has been done and implementing it for use in high-end
amplifiers.
Harley: Dan, you think that the technology is just not mature enough now. Do you hold out hope in the long term for switching amplifiers?
D’Agostino: I haven’t been holding out a lot of hope Robert, to tell you the truth. We don’t use outsourced designs or modules or anything like that at Krell. We like to do our own designs on everything we do here. Other people may get great results with off-the-shelf modules, but it’s just not our way.
For me a switching amplifier is explained in a simple ramp response of a 1kHz square wave. That tells me everything about that amplifier that I ever want to know. When I can see a square without a lot of fur and dirt on it, then I’ll be interested in it, but as of yet I haven’t seen anything that even remotely resembles a clean square wave. That’s it in a nutshell for me because I guess I’m an old-school guy. I like to see waveforms come out the way they look at the input, and that’s not what we get with switching amps.
Harley: Do all switching amplifiers introduce distortion that’s visible on a 1kHz square wave?
D’Agostino: They’re covered with fur and dirt. As long as switching amps can’t pass a clean waveform, I have a hard time being interested in them. You can say anything you want about how you can model the sound and make ideal filters and tailor the sound, but at the end of the day if you put a signal in, the signal should look the same coming out. I find switching amps unusable for music or anything else that I would be interested in. I’m sure other people have different opinions, and that’s fine. That’s why we live in America.
Harley: Is that inherent in the technology, do you think, or just in the current implementation?
D’Agostino: All the modules on the market that I’ve tested do the same thing. Now maybe you can filter out that fuzz to some degree, but it’s still there. I like the output of my amplifier to look like the signal from the generator. When I put those waveforms in and they come out looking funny, I have to think that when I put music in the music will come out funny, too. I also don’t feel comfortable with putting my name on something that somebody else designs.
Advertisement
All documents notede with file size are in PDF format.
Download Acrobat Reader
