Congratulations Atmasphere!


I noticed today that Ralph Karsten (whom regular and even occasional participants in this forum will of course recognize as the designer and proprietor of Atma-Sphere Music Systems, as well as a uniquely valuable contributor to the forum) was granted United States patent number 10,469,042 on November 5, 2019. It covers an audio amplification technique he had indicated here that he has been developing, which in simple terms appears to me to basically be a clever combination of an analog-to-pulse train converter (as used in traditional class D amplifiers for example, among other audio-related applications), with an output stage employing circlotron topology (analogous to the topology used in his OTL power amplifiers, but utilizing solid state devices).

Link to the Patent.

Congratulations Ralph!!

Best regards,
--Al


128x128Ag insider logo xs@2xalmarg
Ralph has a uniquely thorough appreciation and understanding of "what matters to the ear". My mentor Earl Geddes (with Lydia Lee) had a pair of peer-reviewed papers on distortion perception published in the Journal of the Audio Engineering Society. Their findings backed up everything Ralph had been telling me. In fact after writing the papers Earl remarked to me, "now I understand why you and your friends like tube amps".

I think we are extremely fortunate Ralph has chosen to pour himself into designing a switching amplifier that actually works with the characteristics of human hearing instead of against them. We may be about to see switching amps that do not have the traditional drawbacks of switching amps.

Right now I’m really glad I’m a speaker designer instead of an amp designer, because I would hate to compete against what’s coming.

Duke
yeah, I’m also an Atma-Sphere dealer
You guys know that I've been a speaker builder for many years.  Through my struggles in the amp forums here,  the past couple of years, I have built modified and repaired several amps to improve my knowledge.  My studies on circlotron started with Jim Bongiorno and the old Sumo Nine.  I have a highly modified Nine which is very good.  Then looked at how circlotron worked in tubes and how it was modified to work OTL.  
So,  If I understand what Al has written,  Ralph has created a symmetrical bridged class D tube amplifier.... Of course, I'm guessing, but I'd love to hear more. 

Can Ralph or Al expand on this?  Very interesting indeed. 
@timlub 

Hi Tim,

Thanks for the mention, but of course my knowledge of the design is limited to what is said in the patent. And the Abstract at the beginning, supplemented by the figures (especially the first three) seem to me to present a good concise overview.

Regarding your mention of tubes, the figures depict a circlotron output stage that is FET-based, and I suspect that is what Ralph has been developing. I note, though, that the second of the 20 claims broadens the scope of the patent such that it also encompasses tubes as well bipolar transistors when used in a similar configuration.  

Best regards,
-- Al

So, If I understand what Al has written, Ralph has created a symmetrical bridged class D tube amplifier.... Of course, I’m guessing, but I’d love to hear more.
@timlub The tube part isn’t right, but otherwise, yes. Essentially the invention is a Circlotron class D amplifier. Prior to this the prior art was either half-bridge or full-bridge; the former needing two output devices which are arranged in a ’totem-pole’ configuration (and is a fairly simple circuit), the latter being two half bridge circuits driven in opposition. Now there is a third means, the Circlotron, which has more in common with a half bridge circuit except for two important changes: its symmetrical and the output devices are not directly in series with each other. Its the latter bit that is the most interesting, as this essentially reduces shoot-through currents which allows for a circuit with less dead time. In a nutshell its a method of reducing distortion in a class D design.