|
OTL Asylum OTL, Output Transformerless Amplifier User Group. |
|
|
|
OTL Asylum OTL, Output Transformerless Amplifier User Group. |
|
In Reply to: Ideal OTL posted by david on March 28, 2002 at 14:24:03:
In your math of Zo, you neglected 1/2 of the output tubes. The measured Zo that we get falls within 7% of the forecast 2.5 Ohms. We assume its due to the actual Rp being a little lower in circuit.Of course in the past we tried the 6C33 (http://www.atma-sphere.com/novacron.htm) and we also made a 6 tube version which is the amp that Lew has. The 6C33 is far less rugged then the 6AS7G for class A operation- all the other amps you mention are not class A! The 6AS7G has a better saturation characteristic with a large grid current window as well. This allows for substantial class A2 operation, something that's pretty hard with the 6C33.
We tried regulated power supplies in the past too. Try the frustration of a regulator failure at 5:30 PM Friday afternoon sometime! We built the amp to be as bulletproof as possible. Interestingly, regulating the output proved to have no sonic benefit at all. It turns out the the output section of the the amp is a lot faster then any regulators we've seen so far. Harvey Rosenburg tried using chokes in the output section and thought is sounded better on high efficiency speakers, but installing any series impedance like that in the output section tends to starve the amp for current should you actually demand some power. Regulating the driver has a nice sonic improvement, but should the regulator ever take a dive the loudspeaker could be toasted in a heartbeat. The amp can easily produce the same kinds of currents that transistor amps can!
We've been doing this a long time (I've been at it about 27 years) and the fact of the matter is that the Atma-Sphere amps are built to a standard that few wish to emulate.
The reason has to do with feedback more than anything else (the other amps you mention employ feedback except for Kevin's project). A long time ago I figured out that feedback is a bad thing. True, it helps in speaker servo-control (BHK measures that servo control and assigns an impedance value to it; we measure the actual internal impedance of the amp) the question is whether that servo control is a good thing. When you compare high feedback amps to those with low feedback (on a proper load) the high feedback amps sound more shrill (Tenor, JE; although this is certainly not the only reason those amps have such a quality) and bass impact is eviserated, gone. A relaxed, easy presentation (like real music, but in your face if the music is that way too) is only going to come from little or no feedback at all.
If the feedback can be made to be real-time, that is, acting at precisely the same instant as the music, it would work. But due to the propagation delays inherent in any amplifier, loop feedback can never be an appropriate solution, despite Fourier and Shannon.
So we build amps with little or no feedback. This limits the speaker choices to a certain degree, but the speakers left that we *can* use will sound better with our amps then speakers we can't use with some-one else's amps. IOW, we painted ourselves into a bit of a corner, but its the best-sounding one and the middle of the room sucks.
Hopefully this sheds some light on the design parameters we've been working with.
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- Re: Ideal OTL - Ralph 08:02:12 03/29/02 (3)
- I have to disagree - Jonathan 10:15:23 04/28/02 (1)
- Re: I have to disagree - Al Sekela 11:12:44 04/28/02 (0)
- Re: Ideal OTL - david 12:23:35 03/29/02 (0)
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
