Quad ESL-63 and low-powered amps; Sun Audio, Atma

that's great Tomic. Kent is the man.

Though I have heard fried dog is underrated.

@montaldo contact made, will most likely send them....kid and dog frying voltages floating around.......ha....

Atmasphere, do you think the differences could be measured someday, but we have not figured it out yet?

We can measure them now, but we don't go about it right.

Essentially what is needed is a weighting system; one that applies more importance to the higher ordered harmonics since the ear is so much more sensitive to them and places less weight on the lower orders since the ear doesn't care about them so much. Then the spec sheet can tell us how the equipment sounds.

Horrors!

The industry is a long ways away from that right now. The way our ears detect sound pressure is sort of an inconvenient truth. Easy enough to prove though and with really simple test equipment...

 here is Kent email if you want to try that: esolution@estatsolution.com

@atmasphere I am hoping to build your kit monoblocks someday, perhaps to run the 63

@montaldo yes i have tried calling twice today, no answer, no voicemail- it is summer...i sent an email....nice website, boards, etc...certainly looks like expertise and a valuable resource for sure....

I know nearly nothing technically, compared to Atmasphere, but my experience certainly confirms what he said. I have owned many amps that I imagine most people would consider well-designed, and they sound so different from one another. I know this is an age-old debate, but I have always assumed that there are measurable differences that we just have not figured out how to.measure, such a s a measurement that would gauge how an amp performs with music vs test signals. 

Atmasphere, do you think the differences could be measured someday, but we have not figured it out yet?

So as long as a well designed amplifier is not driven into clipping it should sound the same as another one not driven into clipping.

This is a pretty big assumption. What is meant by ’well designed amplifier’?

Designers have different goals so their amps are going to sound different. The reason is twofold: output impedance and distortion. All amps make audible distortion- that’s probably the first thing that should be a given, but so many people think that since an amp has 0.005% THD its distortion is inaudible, and that simply isn’t so.

The reason for *that* is that the human ear/brain system uses higher ordered harmonics to sense sound pressure and to this end is more sensitive to these harmonics than most good test equipment. This is complicated by the Fletcher-Munson curve, which shows that our greatest sensitivity is in the upper midrange - birdsong frequencies. Many instruments have fundamentals that are much lower than that, so if harmonics due to distortion show up in this region we’re tuned to hear it.

Most amps that have very low THD also have distortion signatures that are almost entirely composed of higher ordered harmonics. Some of this is on account of the feedback used- while it suppresses distortion, it also adds some of its own (see the writings of Norman Crowhurst; this should be no secret as he was writing about this 50 years ago).

The human ear converts all forms of distortion into some form of tonality- we’ve known this since the 1930s (see the Radiotron Designer’s Handbook, 3rd Edition).

My goal as a designer is to minimize the distortion product to which the ear is most sensitive- IMD and the higher harmonics (5th and above). Others seek to have flat frequency response with overall low distortion, but I’m here to say that doesn’t work out so well. If you don’t pay attention to how the ear perceives sound, you will wind up with colorations- specifically brightness and harshness. That’s a coloration as much as the ever loving 2nd harmonic that so many tube amps have. The ear has tipping points where it will favor distortion over frequency response, so you have to be careful!

Just my opinion of course... The majority of the audio world, which is ’mid fi’ to the high end market, does not agree. In that world, a Pioneer is as good as a Pass for a lot less bucks because the specs don’t lie, right? The problem is that we audiophiles have been lied to for a very long time and we each intrinsically know that we still have to audition the product no matter what the spec sheet says. That is because the spec sheet isn’t based on how the ear/brain system works, its based on how the eye works and how a good story can propagate even when its not true. In a nutshell, most spec sheets are an excellent example of the Emperor’s New Clothes; a lesson which seems to be lost on so many these days...
Bottom line: well designed amps are not going to sound like each other. You have to know the designer’s intention. One might want it to look good on paper, another might want it to sound like real music. That’s a **huge** difference, yet both can be ’well designed’!!

If you decide not to complete the updating yourself I recommend Kent McCollum at Electrostatic Solutions for Quad rebuild. I bought mine from him and had.a great experience.

Have fun with those quads.

been a very very long time since I have mucked about with ESL-63..we sold them alongside KEF, Acoustat, ADS, Vandersteen, Beveridge, Infiniti, etc....
last time I owned a set of Quads was the 57 with a CJ45...lovely....

so i bought a pair at auction here for a song....anyway, they of course need rebuilding...which will actually be great learning fun...

plan to run them with my 1961 MC240

enjoy the music.....

I have used 6336 single ened amp to drive stacked quad57 to my best satisfction.  Stay away from low damp factor tubep such as otl amps. Do not use sub woofer  unless you are not capable of detect  any phase shift of the speakers. 

Funny you mention equalization because my quads and subwoofers did create a 8 or 10 dB bump which I cured very adequately with the use of a Rives Parc equalizer. Like magic. So I did use subs with my quads and totally agree they can be well integrated with esl-63s if you have the right subs and the right crossovers... in my case I used Vandersteen 2wq subs with the MH-5 ( I always get the model number a little wrong but it's the ones used in Vandersteen 5 system). Really amazing.

 I swore off solid state amps (it would be fun to ABX someday and see if solid state could fool me) though, I, I I am a believer that, in general, it is about execution and not about materials or topolology!)... But I had a great experience with subs and esl-63s because I lucked out and found subs that are fast and have pretty brilliant xover approach.
Someday I may go back to quads... They do things few speakers do in the treble region,  in my view. Peter Walker was a genius, and Peter Snell too. I should lobby to have my first grandson named Peter.

Did your listening test with Peter include tube amps, or was that the time period when tube amps were out of Vogue?

Montaldo, thanks for your friendly response. My aim is what Peter Walker once called 'the closest approach to the original sound'. There are many problems with that, and the two biggest ones are the recording and mastering on the one hand, and the speakers and their in-room response on the other. The former set of problems we can do little about other than vote with our money/feet, the latter you and I have solved to the best of what I at least believe is possible, by using Quad electrostats (plus subwoofer support). In between those two ends of the reproduction chain I want to keep things neutral, and with the designer of our speakers I believe that can be achieved without too much trouble. If you believe or fear a perfect match is hard, just try the current incarnation of your speaker designer's own take at the problem of amplifier design, the Quad QSP or now the Artera. I love my more or less identical 606-2.
Of course there are all kinds of ways in which you can make an amplifier that sounds bad, and there are all kinds of ways in which you can make an amplifier that sounds very sweet but that is not neutral or accurate. And yes, you have to measure in real life situations such as a realistic speaker load. But after all that, it is still quite doable to design amplifiers in different ways, and at the end of the day have them all sound equally neutral. Here, listening tests are of course the final arbiter, but they have to be done under controlled conditions, and as I wrote, I did join one of those blind tests, run by Peter Walker himself, and the result was sobering (I had expected to be able to distinguish the different amplifiers).
By the way, if there is one suggestion I may do, it is to use a DSpeaker Antimode 8033 to tame your subwoofers. I did, and it was the best audio investment of many years, particularly with electrostats.

@willemj "There have also been subjective evaluations in the shape of blind tests to see if under controled conditions panels of experienced listeners can distinguish properly designed amplifiers from each other (I participated in one). The answer was that these panels could not distinguish them."

About 10 years ago, I shared a room at CES with a most well-respected and liked (including by me) person in the high-end audio business. If I actually revealed him here some would be floored. One thing I’m not crazy about is he likes to go around challenging people in the most (in my opinion) classless of ways by putting a ridiculous amount of $$$ out there if a person can pass a test he runs to distinguish A versus B. I told him lets not get into the conversation because I don’t play such games, especially with friends or people I like, I just don’t have that kind of ego. Bottom line, the faithful production of music in the home is something I love, not something akin to a male parts measuring contest, though so many make it into that. Anyway, to his absolute astonishment, I (easily!) scored 100%.  He dejectedly / puzzledly / worriedly wondered what he needed to go back to the drawing board to tweak

Willemj, you raise good points. And by the way I am sure you know much more about audio than I do, so I will not be making many technical arguments here. A couple of points/questions:

1. As you suggested may be the case, I think you and I have different objectives. I want a system that most effectively tricks my brain into thinking I am listening to a real musical performance, regardless of what distortions or equalizations or trickery is employed. Separately, I don't seek neutrality to a recording because so many recordings are not representative of the original event. I don't want to hear recording flaws perfectly,fine in all their  glory. These topics, though,  have sprouted arguments for decades so probably best we don't chase them here, eh?

2. Maybe this question is a less well-worn road, and I ask this because I am interested in the answer,  not because I am baiting or trying to set you up in any way for a punch line :) ... Couldn't two amps of differing architectures but that each measure flat and neutral (however you measure that), playing music in exactly the same system, sound different? I have read a number of times that playing a dynamic music signal is an entirely different game than is playing test signals. Test signal performance may not be entirely predictive of what happens with a dynamic signal. Is this true? And further, could not the interactions with other components cause the otherwise "measured-equal" amps to sound different because they may operate differently into the real-world speaker load, with real music?

You made me curious, of course, so I did a quick online test (not the equivalent of a more serious test, of course): I could easily hear 13 kHz, which is apparently pretty good for a male in his early sixties, although it reminds you of your physical decline. It does put the audiophile golden ear into perspective, of course.
When I did that blind comparative listening test with Peter Walker's setup I was some 25 years old.

How old are you?

@willemj , somewhere, you posted you could not hear hiss coming from your amplifier ( through the loudspeakers ), and that hiss was not normal. However, your young daughter heard the hiss. Need I say more.

Poor Ocham to be so abused.

willemj

... the job of an amplifier is an easy one ... absolute perfection in this is unnecessary ... technology has progressed to such an extent that ever since the seventies of the last century it has been possible to produce amplifiers with imperfections that are below the threshold of human hearing acuity ... 'straight wire with gain' has been achieved, and at increasingly lower cost ... yet there are continuing accounts of audiophiles who claim they can hear significant differences. How could this be?

Simple. Occam's razor.

Montaldo, I don't think we need to be omniscient for this - we will never be. Let us not forget that the job of an amplifier is an easy one: it just has to amplify an electrical signal without otherwise changing it. Fortunately absolute perfection in this is unnecessary: there are experimental data on the critical levels of imperfections: how non-flat can frequency response be before we notice, how much distortion and what kind can we have before we notice etc? And fortunately technology has progressed to such an extent that ever since the seventies of the last century it has been possible to produce amplifiers with imperfections that are below the threshold of human hearing acuity. Peter Walker's 'straight wire with gain' has been achieved, and at increasingly lower cost. There have also been subjective evaluations in the shape of blind tests to see if under controled conditions panels of experienced listeners can distinguish properly designed amplifiers from each other (I participated in one). The answer was that these panels could not distinguish them.
And yet there are continuing accounts of audiophiles who claim they can hear significant differences. How could this be? Are the differences they hear real? And do these audiophiles actually aim for neutrality or do they want a more euphonic version of the original (i.e. do we have the same objective and are we, therefore, talking about the same thing)?
When you look at measurements of audiophile amplifiers in Stereophile you can see that many do not measure particularly well, and are likely to colour the sound in one way or another. In short, they will sound differently, since they are not playing the same 'straight wire with gain' game. I presume this is intentional, and if that is what consumers want, it is up to them. I have tried to describe what makes such amplifiers different. My contention is simply that if you want a straight wire with gain, that has become an easy and fortunately also quite affordable target.
Anyway, to end this, I want to post a link to an interview with my audio hero, the late Peter Walker: http://www.meridian-audio.info/public/interview%5B4446%5D.pdf

Douglas, thanks for an elegant response.

Willemj, I also find it hard to swallow that all perceived differences in amplifier sound are due to issues like clipping. My biggest problem with statements like that is the same problem I have with the objectivist camp in general: it assumes we understand why amplifiers sound the way they do to our ear/brain interface. I don't think we know even a small fraction of that. So how can anyone assert that this or that is the sole cause of sonic differences in amplifiers? It presumes perfect knowledge no one has.

willemj, not intending to misrepresent your position. I appreciate the clarification. Aside from the issue of clipping, I see then that we will not agree on the question of amps sounding different from each other. So be it.  :) 

The 63’s have 4ohm load from 6khz-12khz and then an even lower (epdr) of 5ohms combind with -40 degree phase at 14khz.

This indicates an amp that needs to be able to drive 4ohms as it’s best feature, not as a secondary feature. Which means (for tube if you can get it) an output transformer that’s specifically wound for 4ohms use, not 8ohms with a 4 ohm tap.

https://www.stereophile.com/images/66Quadfig01.jpg
" Dropping below 6 ohms below 50Hz and between 5kHz and 18kHz, with minima of 3.5 ohms at 10Hz and 13.85kHz, this represents a reasonably demanding load for an amplifier to handle. Tube amplifiers would be best used from their 4-ohm tap (if they have one), though any solid-state amplifier that more or less doubles its power output into 4 ohms compared with 8 ohms would have no problems driving the Quad."

Cheers George

I had great success using 60-70 watt ARC tube amps when I ran Quad 63s (mine were Crosby modified). I went from the 57, which was augmented with a sub and ribbon tweeters and found the 63 (without the need to supplement) to be the better all around speaker, but it lacked the "magic" of the original 57. Guess which speaker I kept, had restored and still use (in my second system)?

No, that is a misrepresenttaion of my position. What I said earlier was that they should be judged within their design parameters. So as long as a well designed amplifier is not driven into clipping it should sound the same as another one not driven into clipping. However, elsewhere I have also argued that you need a lot of amplifier power to avoid such clipping of dynamic passages. And that is precisely what I think was happening with my poor little 303. So again, I don't doubt people hear differences, but it is important to investigate what it is that they are hearing. Beyond insufficient output power, there may also be issues with input clipping due to a gain mismatch at the input stage, or a load dependent frequency response. But those are different issues that I already discussed earlier. So my position is a bit more complex than you may have thought.
As for my experience, well I am not a novice, not at all.

willemj, is it just me, or are you changing your opinion in regard to amps as you gain experience. If my memory serves me correctly you were stating in the past that all amps sound the same at matched level. However, here you are discussing different amps and sonic changes; I presume without level matching. So, are you forming new opinions about amps in systems? 

BTW, stats are great! I have the Kingsound King III electrostatic and find that power cords make a substantial difference in the performance of the speaker. Don't scoff; try two or three different ones with various geometries and AWG. You'll thank me later.  :) 

Funny to see this old thread reactivated! I started it back in 2015. Lots of equipment under the bridge since then and I thought I would share my experiences.

Roberjerman is right... Single ended amps just didn't work with esl-63s. Usually dead on top. Other amps didn't work, such as Shindo, because they are made for 16 ohm loads and anything under 8 ohms is a real problem. The Quads literally blew amp resistors. However the sound was as good as I have ever had... stunningly beautiful, textured, real. Made most recordings sound like music.  It was a 30-watt Montrachet. 
Regarding power, I found even 30 watts can make great music, though be aware that I use Vandersteen 2wq subs so the bass dynamics were taken care of. Still, depending on your Sonic priorities, big power is not necessary. I also used VTL MB-185s on the Quads and they sounded absolutely amazing. The power does bring the macrodynamics to life in a very noticeable way. If your priority is mid-range musicality and timbre and texture, power is less important. If you need great dynamics, then more power. But if macrodynamics are and/or bass are really that critical to you, maybe Quads are not the speaker for you. Even with big power esl-63s are no match for many dynamic speakers, in terms of macrodynamics.
I never thought I would sell my Quads but I did. I compared them head to head with a fully refurbished set of Snell Type A and preferred the Snell's. Now I run a VAC Renaissance 30/30 with the snells and Vandy 2wq's and the sound is the best I have had. But nothing does graceful treble like the esl-63s... I miss that!
 

+1 I own their later incarnation, the 2805, after having owned the original ELS57 for many years. The 2805 (and 63) has a pretty low sensitivity (lower than the ELS57). Originaly I drove my 2805’s with my old 2x45 watt Quad 303 power amp, but that was not enough (we had also moved to a bigger house). So I bought a refurbished 2x140 watt Quad 606-2 (the almost identical equivalent of the more recent Quad 909, QSP and Artera amps). That was clearly better for more dynamic music. Even so, I can still hear limitations, and consultation with Quad UK’s Rob Flain suggested that in our large room the 2x260 watt monoblocks would be an improvement. This is just to give you an idea of the ballpark you should be considering. And of course, Quad’s own current dumping amplifiers like my 606-2 or its later incarnations work well with the speakers - and for a quite modest outlay. It was not for nothing that around the time of the introduction of the ELS 63 Quad came out with their 2x100 watt 405 current dumping amplifier.

No single-end tube amp will work well with the 63's because: sensitivity is low and the impedance curve varies from high to low (30 ohms to 2 ohms). All single-end amps have rather high output impedances (2 to 8 ohms). They will NOT have a flat frequency response into a speaker without a flat impedance (Ohm's Law!).

Correction the assemblage is single ended and 8 watts. Looking at so many amps I got them confused.

I don't think you can go wrong with CJ and Quads. It's more a question of whether you like the older CJ sound that you're getting with the MV-60 or the newer, more neutral and detailed CJ sound. I ran my 63's with several different amps including Music Reference RM-10's both in stereo and then mono blocks, and CJ Premier 12's. CJ combined with the Quads was always a step above everything else. Also, don't necessarily agree that you need at least 50-60 watts with the 63's. My Rm-10's in stereo at about 25 watts did just fine in a 16'x 24' room at all but very loud volumes with all types of music but some Classical.

Montaldo, you might want to look at this post:

http://forum.audiogon.com/cgi-bin/fr.pl?aamps&1420484322&openflup&51&4#51

and this link:

http://www.atma-sphere.com/Resources/Paradigms_in_Amplifier_Design.php

Unless you have a very small room, you will not have enough power with a 300b-based SET. They only have 7 watts. You need more like 50-60 as a minimum to really make the speakers sing in the average sized American room (17' by 20' or so).

Quads go lower than 50Hz, unless you are cutting them off with an electronic crossover. Actually they can make quite convincing bass!

Also, I didn't want to lead with this, but my sonic priorities lean more toward midrange magic and musicality. I am not one who seeks ultimate detail, "neutrality" in the audiophile sense, etc. I play at pretty moderate volumes, and never turn it up to extreme volume. So I am very curious about 300b and/or SET sound. But I am keeping my quads, so does this leave out SET entirely?
Another possible amp is an assemblage 300b push pull that might be fun to try. At 18 watts (I think) it should have enough oomph since it doesn't need to do much below 50hz?

Thanks to everyone for the responses. Sorry for my delay. It sounds unanimous about low powered amps even without any load below 50hz. My CJ MV-60 sounds pretty amazing so unless I can taste the magic of another circuit like single ended, or maybe a great tube like the 300b, maybe I should stick with the CJ. I.e. would an RM 200 really sound much better?

Also, as I look at amps, how can I tell from the circuit design, parts or measurements whether a particular amp can handle the impedance changes in the esl-63?

Thanks again.

We recently sold a customer who ownes these speakers a pr of Atma M60's then a few months later he upgraded to the new Atma Novacrons. When we went to set-up the Novacrons he still didn't have enough vol to get the speakers to sing. It wasn't the amps it was he was using a passive pre. Once we set him up with an Atma pre (about 12-14 dB of gain) everything came together so be careful of the gain of the pre is the lesson learned here.

(Dealer disclaimer)

Roger Modjesky's work is excellent. I have owned one RM 200 with those Quads. However I switched to the Atmas and never looked back. No comparison in finesse and resolving power. The Atmas could handle the difficult impedance curve of the Quads far better, it seemed to me.

I've got an RM 200 for sale right now!

I use the 63s in my second system and Atmaspere is absolutely right to my ears. Below 60 watts you won't hear what they can do. I use Gradient subs with them together with their dedicated x-over driven by a pair of old Manley 250, bought for very little money. The Quads are driven be the Atmaspere 60 watters. There is a pair here on A. for sale at a fair price.
This system ( front end is also Atmaspere ) is really heaven to my ears.

The ESL 63 works great with our amps, but usually its with either the M-60 or MA-1. The speaker actually needs some power and you don't do it justice with lower-powered amps.

I have heard that the designers of this speaker did not assume lots of watts and amperage abundance. I think you may have to find specialty forums for the info you want or even a brief conversation with Quad themselves.
I am sure the McIntosh amps do a good job and the Quicksilver as well but there may well be other alternatives.

Three watts is insufficient to drive the Quads. I do not know if an OTL amp would work with the Quads.

As has been said, a low-power amp is not the best mate for the 63. If you want an Atma-Sphere amp, you'll need at least a pair of M60's. Also look into the Music Reference RM-200 tube amp (Michael Fremer "Class A" listing in Stereophile, if you care). Unlike other tube amps, it's power increases into lower impedance loads like the 63.

I agree with Phillyb, I use Mcintosh MC501's, I tried Mcintosh MC2000, Quicksilver MS190(which were designed on this speaker) , Levinson ML2's, Naim 250 and others. The speaker has a weird impedance load and average sensitivity which is remaining of the amplifier.

This speaker will sing with high powered amps, like the McIntosh amps, MC402. The myth is these speakers can work fine with low power amps, even with my subs I use the MC402.