I am not a promoter of distortion. I am a promoter of best sound quality, especially from electronics, that is my specialty. I NEVER add distortion, BUT I have found that open loop designs often sound better, and that is the case with this Audible Illusions line amp that I worked around (I did not design it) that uses an open loop triode design. About 30 years ago, I actually attempted to measure the higher order distortion in this preamp. In those days I did not have a spectrum analyzer, but I added a GR wave analyzer to the distortion output of a Sound Technology THD analyzer. There was some distortion, but it was pretty low, probably below normal levels that would be directly audible, but the higher order distortion that I was looking for was virtually not there. I was surprised, but at the same time, impressed. This is a heavily 'massaged' design adjusted to perfection, over the years. Later, I lived with it for a few weeks and compared it to the CTC Blowtorch. It was pretty darn good, maybe better in some ways than the Blowtorch. In fact, the Blowtorch design concept actually followed this design philosophy: No loop feedback, simple (as much as practical) through-path, as linear a transfer function as possible, local feedback (single resistor near input) is OK, probably, and nothing much above 3rd harmonic.
The open loop triode design has the advantage of only having significant 2'nd harmonic (still pretty low), compared to the Blowtorch that has mostly 3'rd harmonic, as 3'rd harmonic is not easily reduced to zero. However, the triode requires a single (very high quality) output coupling cap, and the Blowtorch is direct coupled throughout.
I can never make a jfet as linear as a quality triode, but I can make it balanced more easily, and sometimes this is an advantage.
Now, AI now wants balanced out too, so this is why this new design challenge has come up. AI is trying to include its 'proven' design in this new line amp, but then there has to be an inverting circuit in front to make it work. The solution is not so simple, except for those of you who do not design at this level.
The open loop triode design has the advantage of only having significant 2'nd harmonic (still pretty low), compared to the Blowtorch that has mostly 3'rd harmonic, as 3'rd harmonic is not easily reduced to zero. However, the triode requires a single (very high quality) output coupling cap, and the Blowtorch is direct coupled throughout.
I can never make a jfet as linear as a quality triode, but I can make it balanced more easily, and sometimes this is an advantage.
Now, AI now wants balanced out too, so this is why this new design challenge has come up. AI is trying to include its 'proven' design in this new line amp, but then there has to be an inverting circuit in front to make it work. The solution is not so simple, except for those of you who do not design at this level.
Last edited:
Loudspeakers are electronics too. Speaking of which, how does frequency response at your listening spot look?
In my opinion, frequency response is secondary to audio quality. In fact, I doubt that a really flat frequency response at my ear is possible. A few decades ago, everybody had a 3'd octave equalizer, and so did I. I never used it. Today, I use no added EQ, but I do try to make my components as flat in response as possible. The biggest exception is the Sequerra Met 7 speakers that both Scott Wurcer and I use often, and they are optimized for TIME arrival response, and have relatively lousy frequency flatness, YET they can sound wonderful! Why?
I EQ my OB subs but not as much as "you are supposed to" and I cut a modal peak. I've tried HF boost to compensate for hearing loss but prefer the sound without. I don't measure or EQ my Jordans but listen nearfield. I make no adjustments between recordings (other than volume), but let my ear/brain combo make any necessary modification 🙂
But it is one of those things that can be shared online.
Would you mind sharing the image of your FR?
For you to mock around? Hehe this is the internet where anyone can share information (one who knows and one who doesn't) but unfortunately ANYone also can decide which one is priceless and which one is worthless. Imho, he is sharing priceless info and you are asking for worthless one?
I know. Well, if someone is looking for the latest update and new findings in audio, she won't find it here. Because there's no new findings in audio as far as I know. If there will be one in the future, I'm sure it will be just a 'finishing' of what have been found 20 or 40 years ago.
You can get 0.03 - 0.09 % THD at 8Vpp from a DHT such as an 801 on an open loop common cathode, most of it is H2 while H3 is at ~ -120dB and H4 and higher lower than -140dB.
I think you are looking in the wrong place. JC is stuck in the 1980s solving 1980s problems with 1980s solutions.
But still scoring well in the present market with new products earning high praise in audio magazines. So, my full respect for JC.
I don't think you get it Bill, JC just mentioned of considering 1950 tube technology, probably to ease issues associated with medium quality digital sources such as CD. 🙂
I really could not understand any aversion to older technology since all technology, be it old or new, has its own warts. As an example, in 2010 NASA published a paper on "MOSFET Failures Inside the Advertised SOA" and I quote :
"This same problem was observed at the Jet Propulsion Laboratory (JPL) in 2003 and was labeled as a “Thermal Instability inside the advertised SOA.” The 2003 JPL failure had built a “Protection Circuit” and destroyed the MOSFET every time the circuit was tested. JPL looked into this destruction, talked to the manufacturer, and discovered the auto industry had found the problem in 1997. JPL then reverted to “older parts,” and trusted the manufacturer to advertise the problem; however, this never occurred."
and
"Thermal runaway is a problem affecting a wide range of modern MOSFETs from more than one manufacturer. Older parts also show thermal runaway, but well outside the SOA. Thermal runaway is currently over a larger area of the Vd – Id plane and inside the advertised SOA affecting most modern power MOSFETs."
Hidden side effect of manufacturing process change. And I'm still fascinated with 1920 triode such as the UX210, ~ -70dB THD at 8Vpp open loop single active device.
Last edited:
"Don't eat the yellow snow."
"Don't test for mains power with your tongue!"
"Hold the cold end of the soldering iron."
Don't continue to hold your breath waiting for, oh hold on I have to go do....
Latest stereophile Recommended Components is down to one JC design so hopefully the new stuff will bump his count back up. But the audio magazines are bascially old men pushing old man systems. There is only one stereophile reviewer who actually pushes the envelope in moving beyond basic 2 channel stuff. That is not the world for me. Too much interesting you can do with 21st century technology. But that is me.
- Status
- Not open for further replies.