Ron’s New System

[morricab]

Amps that cancel even order harmonics do not correctly fit the pattern humans have evolved to filter out (our ear makes those harmonics and the brain filters them). Push pull amps cancel even harmonics by design. Fully balanced does the same thing. This was demonstrated by Boyk and Sussmann with modelling a differential circuit.

Keith Howard demonstrated that adding all odd harmonics to a digital file sounded significantly worse than a file adulterated with an exponential decay of even and odd harmonics.

 

[PeterA]

Too long and tedious, sorry, you want to know…you dig.

However, if you look at his latest argument that crossover distortion is not really a thing, all you need to do is peruse Stereophile measurements to see it IS a thing…lots of modern amps have it and they are Class AB even , not Class B or even C as Ralph claims.

Even Douglas Self admits that crossover distortion is a thing and the one thing that can’t be solved other than full Class A or single ended amp design.

Nelson Pass writes about crossover distortion when discussing his Class A designs in Leaving Class A. I know I always preferred the sound of his Class A XA.5 designs to his X designs, but boy were they hot. I now do prefer my SETs on very efficient corner horns. My SS Pass Aleph 3 amp sounded quite decent on my horns too. I would love to try some of FirstWatt SIT amps on the horns.

 

[morricab]

Nelson Pass writes about crossover distortion when discussing his Class A designs in Leaving Class A. I know I always preferred the sound of his Class A XA.5 designs to his X designs, but boy were they hot. I now do prefer my SETs on very efficient corner horns. My SS Pass Aleph 3 amp sounded quite decent on my horns too. I would love to try some of FirstWatt SIT amps on the horns.

Many say his best amps are the Class A single ended SIT amps.

 

[Al M.]

This is an interesting video. It also shows how crossover distortion can be reduced.

 

[PeterA]

Many say his best amps are the Class A single ended SIT amps.

I heard a very large early version that he made for Sony. Huge mono block amplifiers driving four channels of Sony speakers at RMAF 2012 in a huge room. Extremely realistic sounding.

 

[cjfrbw]

I heard a very large early version that he made for Sony. Huge mono block amplifiers driving four channels of Sony speakers at RMAF 2012 in a huge room. Extremely realistic sounding.

Those were push pull with complementary pair VFETs. I heard them with Sony speakers at one of the California Audio Fairs in NorCal years ago.

They sounded great, but I am in possession of one of the scaled down versions NP generously shared on the DIY site. It has single push pull VFET pairs with the same driver circuits. To my ear and with all other distant comparison caveats, I think the smaller one 'sounds' a bit better.

If you are listening to a piece of music at your preferred listening level, and whose voltage envelope only requires, say, 10 watts for full expression, then a 250 watt amp is pretty much acting as a 10 watt amp, just with more stacked devices and current sharing. I seriously doubt that the extra stacked devices and current sharing across the stacked devices make the 10 watts sound better, although there is a common notion (driven at least in part by commercial motives) that the extra stacked watts make better sound.

My own thought about simpler circuits is that our psychoacoustic transforms have a much easier time with simple circuits, even if they nominally have greater distortion, because the distortion is 'predictable and familiar' and thus easier to compensate for in real time listening.

 

[Holmz]

Do you have any real technical arguments? All I'm reading is hurt feelings.

If Ralph is wrong, then you should have no problem proving it, no?

Not saying that SETs can't be great, I just don't see how your line of "argumentation" gets us anywhere.

Without some facts countering Ralph’s input, it does come across as potentially an ad hominem attempt to discredit him on the conflict of interest front.
(The truth could be both, neither, or two other options.)

Subjectively SETs can sound good, at least with horns and other high sensitivity boxes..
And objectively they can have problems.

 

[Holmz]

Well, if we can put volume controls on drivers, and create our own frequency response contour, why does anybody have a problem with my woofer-cooking? Woofer-cooking raises the SPL of the woofer versus the midrange and the tweeter. Attenuating the midrange and the tweeter lowers the SPL of the midrange and the tweeter versus the woofer.

All we're doing is selecting an idiosyncratic SPL output from each driver and thus creating a custom tonal balance. I don't see any conceptual difference here.
...

100%

 

[Al M.]

Subjectively SETs can sound good, at least with horns and other high sensitivity boxes..
And objectively they can have problems.

Objectively *any* amp has its own problems. Even the "low-distortion" ones since, as Ralph and others have pointed out, it's not just the amount of distortion, but the pattern of distortion (higher uneven harmonics etc.).

And many amps can sound subjectively satisfying. Not just one single -- and single-ended -- road leads to Rome.

 

[Ron Resnick]

Many say his best amps are the Class A single ended SIT amps.

SIT 3 was great versus a dozen other amps during Keith's amp survey. (It sounded too SET to Keith, but it was just my cup of tea.)

 

[the sound of Tao]

Amps that cancel even order harmonics do not correctly fit the pattern humans have evolved to filter out (our ear makes those harmonics and the brain filters them). Push pull amps cancel even harmonics by design. Fully balanced does the same thing. This was demonstrated by Boyk and Sussmann with modelling a differential circuit.

Keith Howard demonstrated that adding all odd harmonics to a digital file sounded significantly worse than a file adulterated with an exponential decay of even and odd harmonics.

Having had a range of panel speakers over the last few decades brings home how revealing panels are with amp differences. With Magnepan 20.7 even the smallest changes in system are clear. So I did find it surprising that Ron and three audiophile mates noted primarily just a difference that the mastersounds just sounded a bit more powerful than the Jadis. I’ve never heard medium to larger push pull, solid state or SET sound even remotely alike through any of the ribbon, electrostatic or horn OB panels that I’ve tried them with.

Through Magnepan 20.7s especially which would be closer in context to Ron’s speakers almost any and every small change upstream and even relatively small things like changing out connectors or in hook-up wiring or material damping was relatively clearly easier to identify and characterise than with most other speakers and anything like an amp change was always telling. I do wonder if going through 47 feet of interconnect cabling and then routing up through big woofage boxes in tandem with the ribbons might not be ideal and tends to homogenise the system in ways and obviates differences in the PSE and push pull amps with Ron’s speakers.

 

[KeithR]

SIT 3 was great versus a dozen other amps during Keith's amp survey. (It sounded too SET to Keith, but it was just my cup of tea.)

It was clearly headroom restricted on classical music though. It has a lot better bass than SET tube amps.

 

[Ron Resnick]

Welp, I am disappointed in myself because I used a high pass filter on the input of the Jadis at one point so I knew the technique, but I forgot it. So my bad -- in connection with the Italian edginess.

So . . . Ralph gets full credit for suggesting that I try a high pass filter in-line on the input of the Mastersounds to relieve them of having to reproduce the frequency range below that which is needed to drive the midrange/tweeter ribbons. It worked! The edginess is either substantially gone or totally gone! (I have a friend coming over on Wednesday and a friend coming over on Friday. Each of them has heard the system countless times. So we'll see if they hear any edginess remaining.)

And I still have only about 100 hours on the Italians, so they may become more elastic and smoother still over time.

Jadis JA100s are up for sale.

Mastersound PF100 Litzes have been granted their Green Cards.

 

[the sound of Tao]

Objectively *any* amp has its own problems. Even the "low-distortion" ones since, as Ralph and others have pointed out, it's not just the amount of distortion, but the pattern of distortion (higher uneven harmonics etc.).

And many amps can sound subjectively satisfying. Not just one single -- and single-ended -- road leads to Rome.

I understand the draw for Ron to implement PSE with his Clarysis… in the end one of the reasons I moved over to horn OB from ribbons was the challenge of living with the constraints and compromises when not wanting to go with solid state amps. Glad Ron has found his solution though.

 

[Ron Resnick]

I have been assuming I would get Clarisys Auditoriums someday. But Florián debuted the two-piece, two-way Aria in Munich. It looks pretty cool.

 

[cjfrbw]

SIT 3 was great versus a dozen other amps during Keith's amp survey. (It sounded too SET to Keith, but it was just my cup of tea.)

The SIT 3 is 'technically' a push pull amp, albeit of a different stripe. It is a SIT paired with a carefully matched MOSFET as a complementary push pull pair per channel. The bias on the SIT is provided also by the companion MOSFET as I read it. The voltage amplifier is a transformer.

The SIT 1, SIT 2 and SIT 4 are more 'pure' single ended.

 

[morricab]

This is an interesting video. It also shows how crossover distortion can be reduced.

Yes, but it also shows that it never completely goes away with snything less than full class A.

 

[morricab]

Welp, I am disappointed in myself because I used a high pass filter on the input of the Jadis at one point so I knew the technique, but I forgot it. So my bad -- in connection with the Italian edginess.

So . . . Ralph gets full credit for suggesting that I try a high pass filter in-line on the input of the Mastersounds to relieve them of having to reproduce the frequency range below that which is needed to drive the midrange/tweeter ribbons. It worked! The edginess is either substantially gone or totally gone! (I have a friend coming over on Wednesday and a friend coming over on Friday. Each of them has heard the system countless times. So we'll see if they hear any edginess remaining.)

And I still have only about 100 hours on the Italians, so they may become more elastic and smoother still over time.

Jadis JA100s are up for sale.

Mastersound PF100 Litzes have been granted their Green Cards.

What’s on the bass panels?

 

[morricab]

I have been assuming I would get Clarisys Auditoriums someday. But Florián debuted the two-piece, two-way Aria in Munich. It looks pretty cool.

Sadly was not playing in Munich…it’s somewhat cheaper and smaller than Auditoriums…

 

[Al M.]

Yes, but it also shows that it never completely goes away with snything less than full class A.

At the bias levels and feedback implemented. And who says that this particular, probably pretty old, amplifier is the last word in technical design? You are again talking from the biases of your point of view.

On the contrary, you can also view the video as evidence of how easily the issue can be addressed.