Love Is Always Better the Second Time Around: The Sanders Sound Systems 10e Hybrid Electrostatic Speaker

[schlager]

To really have a clear image/feeling (pardon the pun) of how a certain speaker sound in a certain room, one simply have to experience it. Or a least heard something similar. On two occasions have I heard panel speakers in a private room. One was with some of the newer Quad 2812 or 2912 I think. Right behind the speaker, the owner had placed 2 insulation bats 4" thick, in a V to absorb the back radiating sound. The room was almost covered in 2" dampening panels. It was like having 2 giant headphones on. Definitely to much for my taste. I blame the room and the general setup for that, not the speakers.

The other time was with Magnepans, I don't remember the model, in a normal living room. I was sitting about 2 meters away and with the right music, the speakers simply disappeared and you could hear deep into the recording in a clear cut center image. That was a special experience, that you don't hear to often. It lacked in dynamics and SPL though, which rendered rock and pop somewhat anemic. From what I'm reading, it sounds like the Sanders has solved those issues and that would make them special indeed.

What I find intriguing with dipole speakers is, that you can really play with the ratio of direct and reflected sound, based on the setup. Playing with speaker distance to the front wall and different damping techniques, will bring out a different spatial quality to the recording, more so that with monopole speakers.

 

[tmallin]

Your Magnepan impressions are quite similar to mine at numerous Magnepan auditions at dealers over the years, schlager: great space, but anemic dynamics, low maximum SPL, and a lack of deep bass. Only the largest models, the 20 and 30 series begin to address these issues to my ears and both of those are too wide to comfortably accommodate in the narrow listening rooms I've had over the years. By all accounts, Maggies work best in "shoebox" shaped rooms firing down the long dimension, but you need to give them space all around and sit a ways back because of the horizontally arrayed tweeter/mid/bass drivers (in larger models) to achieve a modicum of inter-driver coherency and thus avoid the dreaded vertical venetian blind effect.

Your Quad reaction is not representative of my typical reaction to those speakers over the years. However, I usually have not heard Quads in a heavily damped listening room. Also, as I state early in this thread, I actually like speakers to sound like giant electrostatic headphones but with enough of the "out there" presentation to have a pleasing speaker-like presentation.

But your reaction to the Quads may have been affected by the placement of foam damping batts in close proximity to the back of the speakers. While I like foam damping of room surfaces, I do not like the sound of foam damping panels very close to the speakers. I've done a number of experiments with that type of damping of the sound of many speakers over the years, including dipoles, most recently with the Sanders 10e. The woofer cabinet is flat enough near the back of the panel driver to allow me to stand a 2' x 4' piece of 4"-thick foam right behind the panel. This should provide excellent damping of the back wave and it does seem to do that; you can't hear any midrange or treble sound coming out the back of the panel if you walk around behind the speaker with the foam in that position. But the foam positioned so closely also provides an unintended loading of the panel diaphragm, a panel that is critically damped by mere air pressure. That changes the sound quality and frequency response of the sound coming out the front of the panel.

As I said, I've tried close up foam damping with many speakers over the years. It is tempting to try such configurations since you need less foam that way and since foam in free air tends to have more effect at lower frequencies than foam mounted against a room surface. However, I've never yet found a situation where such foam damping sounded as good as, much less better than, using more foam to damp the room surface areas near the first-reflection points of the speakers as viewed from the listening position. Most speakers seem to want free air immediately around them and even something as light weight and seemingly non-dense as acoustic foam creates midrange and treble spatial distortions and frequency response aberrations when used in proximity to speaker drivers, whether the drivers are cone, dome, ribbon or panel. Even an arrangement of foam panels mimicking the old Watkins Echo Muffs produce such distortions.

 

[schlager]

I think that the amount of damping and were it should be placed, has a lot to do with the speakers radiating pattern (power response) and room size. Listening distance also comes into play, so there is no clear-cut answers, only guidelines.

Normal cones and domes, has a very wide dispersion almost like a light bulb. They should not be placed close to the sidewalls. For damping technic, my starting point would be the Live End Dead End principal. What is an important point is, that using lightweight damping (carpets, curtains, damping foam with 1-2” thickness), will only damp frequencies from 1-4 KHz and up and thus living a sound rendering too dull or muffled. Using 4-6” thick panels will damp into the low midrange 200-300 Hz and give a sound with much more spectral balance. It will also reduce ringing and flutter echo to a larger degree.

Panels and most horns will narrow the dispersion with increasing frequency, so they can be placed closer to the sidewalls and don’t require the same amount of damping.

With constant directivity speakers, which I have down to 100 hz, you have better control over the speakers radiating pattern. You can get away with much less damping in those situations.

First reflection point should be damped, IF the reflection reaches the listener within 10 ms after the direct sound. Otherwise, it can be diffused.

Just my 2cent.

Best Sebastian

 

[tmallin]

I agree that these are reasonable guidelines, Sebastian. I would emphasize that early reflections (10 ms and earlier) are a much bigger problem in small and/or narrow rooms than in larger, squarer rooms. Most people seem to prefer the sound of speakers firing down the long dimension of the room. Even if your room is 25 feet or more long, if it is "shoebox" shaped and only, say, 13 feet wide, the side walls will cause early reflections in a lot of set ups. In my small 11' x 13' room, early reflections are almost impossible to avoid, at least with symmetrical speaker set ups.

Very narrow dispersion speakers and dipole radiators seem to help tremendously in this room. I need no padding or diffusing at all for the side walls, floor, or ceiling reflections.

The 40-inch tall Sanders 10e electrostatic panels are exceedingly directional in the up and down dimension, eliminating the relevance of the floor and ceiling reflections in the mids and highs since the crossover to the panels is at 170 Hz with a 48 dB/octave slope.

As to the side wall reflections, from my listening position, a mirror mounted on the sidewall so that I can see the speaker reflection shows an exactly edge-on view of the panel. I'm thus listening from a null in the sidewall reflection since dipoles put out very little midrange and treble sound from their edges. While the horizontal dispersion of the panels only begins to narrow significantly around 1 kHz due to the 13-inch width of the panels, that turns out to be sufficiently high in frequency, when combined with the edge-on view of the reflection from the listening position, to make the side wall reflection entirely irrelevant. That's a first in my experience in this small room.

 

[tmallin]

I am more and more convinced that folks who want to take control of equalizing their system for individual recordings ideally should use two equalizers daisy chained.

First, use an equalizer with a lot of parametric bands to set the "foundation" EQ for your system to smooth the overall response of the system in your room as heard from the listening position. Then use something with knobs and, perhaps ideally, remote control, to tune for individual recordings.

For the per-recording-tuning equalizer, one of the Schiit models should be ideal. You can use long interconnects and maybe an extension cord for the power cable to and from the Schiit to place that unit at your fingertips at the listening position. You may want to spring for the version with balanced ins and outs if you do that. For yet more money, you can use the Loki Max with remote control and presets and dispense with long cables, leaving the Schiit in your rack. But, you actually may prefer the tactile control of the lower-cost knob-twirling solution.

For your "foundation" equalizer, I can think of no better unit than the dbx VENU360 for $1,000. It will smooth the response of your system more quickly and easily than any other equalizer I have used, and I've used a lot.

For those with complex or active systems, it can be set up to allow you to bi- or tri-amplify your system and time align the drivers. Different crossover types and slopes are available. It also has both digital and analog inputs and outputs. Thus, it's quite flexible.

It takes only about five minutes to run the AEQ Wizard of the dbx unit once you have the included measuring microphone set up at your listening position. Choose "Recommended Curve" if you want the type of low frequency boost that most listeners prefer. This will smooth the listening position response quite well from bass through treble using 10 bands of parametric EQ, according to my OmniMic V2 measuring system. If, after listening to the results, you distrust the automatic equalization of high frequencies from the listening position, you can manually zero out those parametric corrections centered above any chosen frequency, such as 1 kHz. But, as I said, the AEQ results will measure very smooth even in the high frequencies via OmniMic V2. For further manual tweaking of the response, many more parametric bands, as well as graphic EQ are available, but the five-minute auto EQ will get you 95% of the way there and most users won't want or need to go further.

For passive speakers like Spendors, Harbeths, Stirling, Graham, etc., just turn off or zero out all the settings of the dbx unit other than the AEQ module. This is very simple to do if you connect the dbx unit to your home internet network and use the dbx app for the VENU360, which I highly recommend. Since it's a pro-audio unit, there are a lot of functions available that you just don't need. But, as I said, these extra functions are easy to turn off or at least zero out using the app, which is very fine. Think of the dbx app as the Google of equalization apps: very easy to use, fine display, great user interface, excellent results.

With my speakers, the Sanders 10e, with their very narrow dispersion, foam-damped back wave, close up listening positioning, and time alignment of bass and treble, I usually don't feel the need to equalize for particular recordings. Proper setting of the foundation EQ is all I need. As measured with OmniMic V2 from the listening position I have the foundation EQ set for rich bass (target curve is up 6 dB or more at 20 Hz, beginning the upslope at 500 Hz) and treble measures flat from 500 Hz up to 15 kHz, above which it smoothly rolls off a very few dB.

Yes, even very subtle tonal balance differences among recordings are quite evident, but while I can hear excess brightness, it just is not obnoxious the way it is with most other speakers in my room. This is a quality of the Sanders speakers I noted a decade ago when I owned the 10c iteration. The Sanders seem to merely "report" brightness in the recording; they do not add distortion or edge of their own and thus don't "demand" tonal correction the way most speakers do.

This may seem strange, but that's the way I hear it. It may be that what adds the "cringe factor" to excess brightness is not excess brightness in the direct sound from the speakers, but excess brightness in room reflections which, with the Sanders used as I do, can be much less than with other speakers.

With most other speakers, I need to either roll off the target response above 3 - 4 kHz, insert a presence range dip, or just relax the treble range between 2 kHz and 8 kHz a couple of dB (the built-in Dutch & Dutch 8c approach) to make the sound much more listenable with the vast majority of recordings.

 

[schlager]

Flexible EQ settings are very easy to do, if one run their music from a PC/laptop. Using APO equalizer (freeware) you can have all kinds of different EQ settings and just engage them on and off with the click of a mouse button. I use two different bass shelves for supporting "thin" recordings. I also have a small BBC dip that just takes the edge off some harsh recordings. The possibilities are literally endless. APO EQ can also run the impulse correction files in it's build in convolver, very neat.

 

[tmallin]

Just out of curiosity, with such a computer-based audio system, if the computer is in your lap, how is the signal path running through the computer so as to apply equalization to the audio components? Are you using the computer as a mere controller for the system, like I use an iPad to control Roon or Lumin App streaming? Or is the computer your actual music source/core, connected to the rest of the system components via ethernet cables, USB cables, some sort of sound card with audio connections, Wi-Fi, or what? What about the computer's power cord? Seems like you'd potentially have a mess of wires on your lap.

Sure computer software can do all sorts of time and frequency manipulation of music files. Even audio EQ boxes like the dbx VENU360 I use can save many different EQ set-ups to presets which can be quickly recalled by touches and swipes on an iPad screen. I've done that in the past. And I do have a presence range dip and high frequency shelf filter built and at the ready via presets in the dbx device if I want to use them. It's just that, with the Sanders 10e, as I explained, I rarely feel the need for such recording-to-recording equalization.

But with most other speakers I've owned and used in this room and elsewhere (not including the Dutch & Dutch 8c!), I definitely like to use at least high-frequency equalization from one recording to the next to make the sound more naturally listenable. There also is something to be said for the type of very quick, almost intuitive tactile control that knob twisting on a unit like a Schiit EQ or the Cello Audio Palette/Palette Preamp I owned in the past can provide. That's why I think the two-stage equalization may be of interest to many, especially those NOT running their music from a computer source.

I don't think someone using Sanders 10e speakers, in a well-set-up room, listening from the sweet spot, is going to have a computer in his lap, using a mouse. You probably are talking about sitting at a desk listening via headphones or some sort of desktop or stand-mount speakers.

But a flat desktop is not something you want near your ears when listening to fine speakers. You want as much free air near your head and the speakers as possible. With most speakers, even holding an iPad in your lap causes rather obvious deleterious effects on staging and imaging due to reflections off the iPad screen. It is only the Sanders' extreme vertical directionality which makes an iPad in the lap condition listenable and it is still sonically best to put it down on the floor beside your chair.

Pro-audio folks get away with having a mixer desk in front of them since the "desk" is angled down toward them. Reflections of sound from the speakers off the mixer are angled down and away from or past the operator's ears, in contrast to the situation with a flat desk of the type one usually sees pictured in computer audio installation ads and set ups. Such a flat desk wreaks havoc on the sound one hears sitting near such a desk because the reflections back to one's ears are very early and very strong.

 

[eslguy]

Very much appreciated tmalin's thorough and informative discussion on the 10's' I must say I am in almost total agreement with all he has stated. That is pretty rare to get two people to agree on anything in audio and especially loudspeakers. I have never actually met Roger but have "known" him for over thirty years. Back in the 70's and eighties Roger published articles in Audio Amateur or Speaker Builder magazines on how to build electrostatic loudspeakers. Having minimal experience in audio or electronics (other than building a few Heathkits) I found these articles I found these articles compelling. I had heard the electrostatics of the day - Quads, Dayton Wrights, etc and was captivated by the esl sound, His original articles are available on his website. Unfortunately I was not in a financial position to purchase any of them. I contacted Roger and indicated I was a poor and pretty ignorant person electronics wise but wanted to build his esl's but didn't think I could do it. He assured me I could and he would help. That was in the days before the internet so "tech support" was via snail mail or phone. Over a period of two years he coached me along. The project was quite primitive at the time. Panels were metal garden panels from Ace hardware, some mylar sheets, graphite to rub into the mylar, a photocopier power supply to provide the voltage, sourcing transfers - 6-12 months just sourcing parts. Built the electronic xover from a Marchand kit. Then on to building the Transmission line woofers. Kind of a big woodworking project. The end project were stacked ESL panels (about 7 feet high) and separate TL woofer (again about 7feet high) with a 12 inch Dynaudio woofer feeding into dual 8-12 foot transmission line. After the 2 years of work I trepidaciously fired them up and to my amazement they actually worked and more amazingly sounded great. I ran this system for about 10 years with various electronics (more on that to come) After more than 10 years of audio bliss the esl panels began arcing. I was rather heartbroken however I felt 10 years was pretty good for a project I didn't believe I could do. Remembering the work it took to build and being a bit busy I thought I would just go out and buy some "regular speakers". As I toured the audio shops (you remember those) I was confronted with the Eros electrostatics - it was love at first listen. I was trying to figure out a way to afford them when it struck me - Roger, you taught me to build electrostatics so I am going to do it again. I contacted Roger and he apologized that now he was running a company actually commercially building his designs he unfortunately did not have the time to be my "tech support guy". He did however pass me over to his friend Barry Waldron who helped thru iteration 2 esl project. This was much easier as Barry offered to "source" some parts for me - basically esl panels, transformers, power supply - basically I just built the frames. (I strongly suspect these were Eros components but never asked) I then went a bit crazy and decided I wanted a 3 way system. So I built another TL box this time with 2 Dynaudio 6 inch drivers each feeding a separate 8-12 foot transmission line. Being even crazier and wanting to know if esl beat dynamic and having a 7 foot box to work with I put a Dynaudio 330 Esotec driver between the 6 inchers. Built another Marchand xover now for a three way system. So 3 floor to ceiling units per side with WAF of minus infinity. Barry labelled this system as the PMS system (poor man's Statements in homage to ML's flagship system of the time) To my horror when I fired them up the panels were arcing. I then realized that I had put a new amp into the old system that went into oscillation trying to drive the capacitive load and the same problem was now happening. Swapped out amps and no problem. One of the advantages of an electronic xover is it is easy to change points and speakers. I tried the system with the Dynaudio tweeters and they weren't even close to the Esl's. Ran this system for 10 or 15 years (who says esl's are unreliable) with 0 issues. Sold them due to age and plan to downsize. However missed that sound hence I have had the 10e's for a couple of years. They are truly as good as outlined in this thread, The new ESL panels are far superior than the old and the Driverack is a much better xover system. I am also amazed at how good the bass is given the size. My home built TLs had amazing bass as you might expect from 3 drivers per side driving very long transmission lines -an organ lovers dream bass. The 10e bass is however quite impressive. I won't say much more about th 10e as I think tmalin has said almost all. I just thought I might add a bit of history and acknowledge Roger's contribution to audio and especially to me

 

[eslguy]

Hi Tom
Our mutual love of the 10e's prompted me to review your numerous other contributions in other threads. I was very impressed' with your obvious knowledge and experience it is a welcome opinion. I gather we are both "old guys" who have been at this a long time so willing to share our experiences and opinions in a respectful manner to reflect our personal perceptions. One of your posts suggested "big bands" were perhaps a listening reference (correct me if I have misinterpreted). I have played in big bands for many years so this for me is a partial reference as I can relate directly to my live and recorded experiences, I am definitely not an audiophile just seeking what brings me close to the music. Having said that I have had deep love of audio and the luxury of numerous audio systems. The reason I am not an audiophile is I believe that the focus needs to be the music. Again having said that the pursuit of audio excellence is a noble pursuit I have certainly pursued that. I have the luxury of numerous systems - 2 home theatre systems - one based on Paradigm floor standers and one based on Maggie 3.7s. For stereo I have Soundlab 545's and of course my beloved 10e's. and a number of small systems. We are both blessed by pretty good systems but in the end it is what brings you musical delight. I fear many pursue the belief that you need to spend a lot of money to pursue the ultimate and although this is a noble cause for most unnecessary strain of resources. . As I type this I am listening to my $200 Monsoon desk top speakers. They reside in the same room as my 10e"s so I could easily switch. Are the Monsoons as good as the 10e's - of course not. But they are a delight and bring me great happiness.
Point for me is the pursuit of audio perfection is a noble goal (however unobtainable) but the end goal is to just get the musical experience that brings you closest to your "sweet spot"

 

[schlager]

Tom thanks for your reply. The laptop has one USB cable connected to a USB hub and of course a power cord, so two cables going out from the laptop. I find it manageable. From that I can control my minidsp and my music files stored on hard drives or streaming. The EQ APO with all its settings, included the convolution file for impulse correction, is running directly from the PC with no latency. If I watch a video, sound and picture is synced.

Running the signal digital from laptop to minidsp the degrading of the signal, in theory, should be zero and I don't find it problematic at all. Lot's of audiophiles are using laptop/PC in this way and I think a lot more will resort to such, in the future.

Of course for critical listening, the laptop is put on the floor beside the listening chair, as you say it courses unwanted reflections.

For most music all that is "ON" is the convolution engine for the IR correction file. Sometime a "thin" sounding rock or pop album needs a little boost in the low bass. That is all. Easy peasy.

 

[tmallin]

More on gain structuring with the dbx VENU360:

One way to actually improve the signal to noise-plus-distortion ratio of any audio system is to take care of gain structuring. This is not tweaky or controversial in any way. Pro audio systems are usually set up to have proper gain structuring, but most consumer audio set-ups do not easily allow this.

In most cases of consumer audio, proper gain structuring will involve putting a volume control just upstream of your power amplifier, if it doesn't have one built in. Few consumer amps have a volume control on the input, though most pro-audio amps do.

Most consumer amps have far more gain than you need. You can tell whether this is so by testing whatever system volume control you are using. If that volume control will allow you to turn your system up to a far higher SPL than you ever need even with your quietest sources, you could further optimize your system's gain structure by "turning down" the gain of your amplifier. This will also prevent amplifier clipping and speaker damage.

Tube amps are notorious for having very low input sensitivities, usually putting out full power with just 0.775 volts of input. Even transistor amps seldom have input sensitivities above one or two volts. That means that that much voltage will drive them to full output. Standard CD players have an output of 2 volts for the maximum undistorted signal. Many high-end consumer sources, such as preamps, DACs, and streamers, have much hotter signals than that (my Lumin X1 puts out 6 volts max).

By proper gain structuring it's usually possible to add several dB of actual signal to noise and distortion ratio to your system. You can hear this by putting your ear close to the tweeter. Without proper gain structuring, you will frequently hear some hiss without any music playing from within a foot or more of the tweeter at any setting of your system volume control, even the minimum setting. With proper gain structuring, you may hear no hiss, or at least greatly reduced hiss.

One of the things I really like about the dbx VENU360 unit Sanders uses as his (crossover, EQ, time alignment) control unit for the 10e speakers is that it has very flexible levels of input sensitivity and output voltage adjustable in the analog domain through the device's Utility menu. Since this unit directly feeds the Sanders Magtech amps it can act as an input sensitivity control for the amps. The difference in quiescent hiss audible from the stat panels is quite noticeable. As delivered, there was a slight hiss audible, but with proper adjustment, absolutely no hiss is audible at any setting of the system volume control (the Lumin X1's Leedh-processed digital volume control in my case) at any setting of that volume control when no music is playing.

A bit of the theory behind gain structuring and how to do it with the dbx unit are explained on pages 52 - 54 of the dbx VENU360 manual; note especially the Tip on page 53.

Generally, proper gain structuring relies on lots of gain taken at the initial stages of the sonic chain. Many pro-audio mixers thus have a high maximum output of 18 or 24 dBu, so a consumer audio preamp with a high voltage output makes sense--IF you can decrease the voltage further down the signal chain before the power amp gets hold of it.

The output impedance of most active preamps does not vary with the setting of the preamp's volume control; the output is buffered to have a uniformly low output impedance. Only a so-called passive preamp (rare these days because of potential problems driving cables and downstream equipment) would have a variable output impedance depending on the setting of the volume control.

Any preamp with a constant high output impedance should be rejected out of hand due to difficulty driving cables and downstream equipment. One exception would be if you use low capacitance interconnects and the downstream equipment is designed to have extremely high input impedance, such as a megaohm. A few companies do that to make their equipment as immune as possible from the electrical characteristics of the driving components and cables.

The debate over the worth of an analog preamp in an all-digital system goes on and on. More and more high-end audio manufacturers recommend direct connection of their digitally volume controlled streamers to power amps.

As I recall, the last time I used an analog preamp with more than unity gain available was the first time I had Sanders speakers about a decade ago. I had the Sanders preamp for a short time before I realized its additional gain was totally unnecessary and not otherwise sonically helpful in any way.

However, many audiophiles prefer eliminating the digital volume control in favor of using an analog preamp. Sometimes I think this may come down to the strong preference of most audiophiles for having a system with a lot of excess gain available from the system volume control. Many audiophiles seem to think that their volume controls should be running in the 9 to 11 o'clock region so that there is lots of "reserve" volume. Nothing could be further from the truth, both for analog and digital volume controls in a properly gain-structured system. Both should be run as nearly wide open as possible (assuming doing so does not drive the unit with the volume control into clipping distortion) since both analog pots and digital volume controls at least theoretically will have less effect on the sound toward the top end of their ranges.

Proper gain structuring will result in your system volume control working in the top third of its range for serious listening with most sources. Sources differ wildly as to how "loud" they are and thus you need to allow for enough excess volume control gain to get the maximum SPL you want from your speakers with your quietest sources. Among the quietest sources I've encountered are Sheffield Lab CDs and the signals from a few internet radio stations such as BBC3, WILL, and ABC Jazz.

I find nothing amiss with digital system volume control these days. With the volume processors running at 24 bits resolution as they do, a digital volume control has plenty of volume attenuation available before the signal is meaningfully compromised. In my system, the digital volume control in the Benchmark DAC 3 sounded just fine, only slightly bettered by the use of the downstream HPA4 relay-controlled all-analog volume control. I later found that the Lumin X1's digital volume control also seemed fine sounding--at least as good as running the Lumin with its digital volume control disabled through the Benchmark HPA4. And since then the Lumin volume control has been made yet better sounding with the addition of Leedh digital processing which can be switched in and out for comparison.

Yes, the hierarchy of system set-up is first to buy inherently good sounding speakers (whatever that means to you), then move the speakers and listening position to get the bass response as smooth as possible while also optimizing the spatial aspects of the presentation, then apply room treatment to absorb or diffuse annoying mid- and high-frequency room surface reflections and further optimize the spatial aspects of the presentation, and then use electronic equalization to smooth out the remaining response wrinkles, especially in the bass. No one is disputing that.

Perhaps this got lost along the way, but the gain structuring I'm talking about is only possible because the dbx VENU360 electronic equalizer I'm currently using, the loudspeaker management system unit that comes with the Sanders 10e speakers, allows such gain structuring. Unlike any other equalizer I've owned (and I've owned quite a few) it has selectable analog domain input sensitivity and output voltage controls.

If you think gain structuring doesn't matter, that's fine. Think what you want. But pro audio folks have gain structured their recording and playback systems for many decades. If you think increasing the signal to noise and distortion ratio of your system by, say, 12 dB is meaningless, that's your prerogative. Some believe that a signal to noise and distortion ratio of 70 dB is just fine because if the noise is any lower it's inaudible anyway. With digital sources, you certainly don't need gain structuring to reach that level of S/N+D.

Using the dbx unit I'm using as an EQ device makes gain structuring my home audio system very easy. If you are using a pro-audio type of amplifier--most of which have an analog volume control built into their input--it also will be very easy.

But if you are using a typical consumer/audiophile amp with no volume control on its inputs, to properly gain structure your system you would have to add an analog volume control to your amp input. One way to do that is to buy a fixed resistor or variable pot and plug that directly into the amp input and allow your upstream components to feed that. A couple decades ago I owned a pair of variable pots for this purpose. They were called EVS Ultimate Attenuators. Cheaper and more compact solutions are certainly available, especially if you opt for a fixed resistor control to knock down the voltage going into the amp.

 

[schlager]

Gain structure is important, especially if you run compression drivers on horns. That will reveal any noise, hiss or hum, from the upstream components. I personally use a 10 Kohm resistor in serie with the input on my tweeter amp. The hiss is reduced so it is barely heard at 1 meter. In LP it is dead silent. Amps with gain control is easier to handle regarding S/N.

 

[tmallin]

Each audiophile needs to think about how to make the playback set-up deal well with recordings as they are. One hopes this can be done while also letting the few really good recordings sound as good as they are.

One compromise that needs to be made, I think, has to do with the subtended angle between the speakers. Some of the best recordings I listen to are those made via Blumlein and other quasi-coincident stereo microphone pair techniques. These usually require a subtended angle between the speakers as viewed from the listening position of at least 90 degrees to sound their best.

Unfortunately, most other recordings--the vast majority--seem to be standardized around a 60-degree subtended angle set up. There's no theoretical reason why that should be the case, but apparently it is so.

Years ago I did for a while standardize my set-up around a 90-degree subtended angle to get the best from Blumlein recordings. This makes most all recordings have a "bigger" presentation because of the wider subtended horizontal angle, and that can be exciting. However, it also changes the tonal balance a bit for the brighter so that some high frequency EQ is needed with most recordings to make the highs less aggressive. (With wider separation, the speakers are aiming more straight into your ears and our hearing mechanism interprets such sound as brighter in the highs than sound which comes more from the front.) Even with such EQ, I eventually grew tired of hearing most recordings sound too "pulled apart" horizontally with "holes" in the staging caused by the wider-than-intended subtended angle. But YMMV.

Now, if you position your listening chair so as to provide you with the 60-degree separation that works best with most recordings, you can get a bit closer to 90 degrees without moving your chair if you sit on the front edge of the chair and lean forward. With many speakers I've owned, this works okay sonically, but it is not really a comfortable sitting position.

Even when the sound improves considerably with this technique, it rarely produces quite enough angular separation and also tends to move your ears lower as well as making it so that the speakers are no longer toed in as much toward your ears. You can move even closer to the speakers by moving your chair (workable if the chair is on carpet and the chair legs make depressions in the carpet so that it is easily moved back to the 60-degree position), or putting another chair in front of your usual chair, or by kneeling on the floor in front of your regular listening chair (although that may not be too comfortable, either).

With speakers where, at close range, the vertical listening height is important, the leaning forward technique can be a problem. The Harbeth M40 series is in this group at least with near-field listening; the best focus and smoothest sounding response is just a bit lower than the tweeter. The Stirling LS3/6 usually benefits from a lower listening height (the ideal listening height being no higher than the lower tweeter), so this leaning forward technique works quite well with those speakers. It also works very well with coincident driver speakers like the Gradient 1.4 and also worked very well with the Dutch & Dutch 8c.

With really narrow horizontal dispersion speakers like the Sanders 10e, you will move enough off the horizontal axis of the panel beams to quite noticeably roll off the top couple of octaves if you set the speakers up for 60 degree separation and move your listening position to form a 90 degree subtended angle. The same is true if you set the speakers up for 90-degree separation and move your listening position to a 60 degree angle.

The only way I've found around this problem with the Sanders 10e speakers is to remove the damping from the walls behind the speakers. This effectively lessens the off-axis high frequency roll off enough to allow appreciation of the better imaging and staging from Blumlein recordings heard with 90-degree separation without moving the speakers. It is one reason to adopt an undamped rear and side wall set up for these speakers. The sweet spot will be a little less sweet, so it is a compromise. Only you can decide whether this compromise is worth it to allow you to hear your Blumlein recordings at best advantage. Alternatively, with the 10e, you could set up for 90-degree separation with the side and rear walls damped as I have done and just live with the unnaturally "pulled apart" and "holey" staging you will hear from most recordings which are optimized for 60-degree separation.

I wish all stereo recordings were optimized for 90-degree separation!

 

[tmallin]

Stripped Naked

I had a nagging suspicion that the very nice looking cherry wood trim strips which edge the electrostatic panels and otherwise provide an attractive "finished" look to the Sanders 10e speaker cabinets could be adversely affecting the sound. While they are beveled/rounded, they do stick up quite a bit from the front edges of the panels and even a bit more than the metal rails in back of the panel.

In addition, as I think I've mentioned before, these speakers are electrostatic air cleaners extraordinaire, attracting lots of dust and grime from the air of my admittedly quite polluted (trains, planes, and automobiles) suburban Chicago environment near O'Hare airport. The surfaces of the panels go from black to gray in a week or two and thus require periodic damp sponging off to return to their pristine appearance. The cabinets and wood trim go from their nice cherry color to a grayed out version just as quickly. And the bass cabinet grill cover seems impossible to keep black without using a vacuum cleaner, which I didn't want to do with the grill attached near the woofer cone.

So one day I decided to combine a "deep cleaning" project with a sonic comparison of the speakers fully dressed with their sound stripped naked of the trim strips and bass cabinet grill cover.

My suspicions turned into frank surprise at how obvious the sonic differences were.

On the one hand, with the wood trim strips and woofer grill in place, the sound is always vertically placed at eye level. There is a more obvious roll off of the top two octaves or more as one shifts head position left and right more than a very few inches. Centered voices and instruments are more resolutely centered.

On the other hand, stripped naked of the wood trim strips and woofer grill, the sound is definitely yet more open and enveloping to the sides and rear of the listening position. The staging is wider and deeper. The sound is yet subjectively clearer and cleaner in both the area covered by the panel and in the bass.

The bass differences reminded me that in my time with the earlier Sander 10c version a decade ago one of the problems I encountered was mechanical noises including rattling, buzzing, and general fuzz around the sound, seemingly caused by the bass cabinet. I attributed this at the time to overdriving the woofer in my attempt to achieve flat-to-20 Hz bass, as well as spiking the speakers to the concrete slab floor beneath the carpet.

Now, I've never encountered these more obvious problems with the current 10e, but occasionally I have heard a bit of buzzing or fuzz on particular plucked bass notes. I've tracked that down to vibrations from the trim strips which could be remedied at least temporarily by slight physical manipulation of the trim strips in terms of the velcro fastening of the strips to the metal rails. Just tugging a bit on the trim strips usually fixes any problem I heard. But without being able to see the woofer cone hidden behind the grill cloth, I did not know if this corresponded with extreme cone excursion, or just some resonance in the cabinet or trim strips themselves. Certainly the output lights on the dbx VENU360 never indicate that the woofers are being driven to extreme levels when this occurred or at any other time.

Stripped naked, the bass is surprising rather obviously cleaned up generally and the occasional buzzing/fuzz is totally gone. It is obvious from close viewing and touching the bass driver surround that extreme excursion is not occurring. I thus have to conclude that either the trim strips, the woofer grill, or a combination are causing the slight occasional distortion I heard. Stripped naked, the bass quality is constantly at a new level of subjective cleanness and there are no sporadic buzzes or fuzz even at higher levels than I usually listen.

No, stripped naked, the speakers are not as attractive to the eyes as they are fully dressed. I don't think there can be much disagreement about this. Fully dressed, with the trim strips and woofer grill, the speakers are truly handsome examples of industrial design and would present well even in a home surrounded by the finest furniture. Stripped naked, the speakers, while probably not in the "ugly" category for most viewers, are rather obviously "industrial" in appearance, and are not "industrial art" in the way that even the largest Wilson speakers might fit into fine decor. No, the stripped naked benefits are strictly sonic.

The occasional slight vertical and horizontal wandering of the apparent center of the stripped naked presentation is not as reassuring of the focus of the speakers as their subjective sound fully dressed. But this presentation center wandering seems program related and usually is only audible on low-quality internet radio signals, disappearing with the higher quality Tidal/Qobuz version of the same material played at the same volume setting.

Stripped naked the speakers seem yet more revealing of differences among program sources, laying bare differences among Qobuz Hi Res and Tidal MQA versions and CD quality versions of given program to a degree not previously heard. Great recordings are more jaw-droppingly wonderful sounding than ever before. Poor recordings, even those with obviously too-bright highs, can still be appreciated because the speakers only report, don't add to, the problem. There is no general brightening of the average balance of a particular recording or of recordings generally, and that is a good thing in my book.

My measurements with OmniMic V2 show the impulse response as least as clean stripped naked as it it fully dressed. The monophonic impulse response (both channels fed the same short sine sweep signal simultaneously) show a visually identical same impulse response with thus "perfect" matching of the left and right speaker distances and time response. If the left and right are not so matched, the impulse response "smears out," which is very easy to see visually on the graph. This is an easy way to see how well your left/right speakers match both inherently and in terms of the physical set up in the room. The Sanders matching shows the best I've ever achieved on this test, graphing as essentially "perfect."

Frequency response wise, stripped naked results in a slight narrow-band decrease in output centered at about 1.2 kHz. I correct for this with a 1 dB boost with a parametric filter centered at 1.2 kHz with a Q of about 7. Otherwise the frequency response graphs seem very similar visually at 1/6-octave resolution with 2 dB per vertical division, my usual OmniMic frequency response graph setting.

I am leaving the speakers stripped naked for the time being. They sound yet better, yet more glorious, to me this way. And they certainly are easier to clean this way. As it turned out, the woofer grill cloth had to be vacuumed from the inside to remove the gray dust ring from where the woofer surround had been in proximity.

 

[EvilTed]

You'd be better served putting the clothes back on the speakers and actually setting them up as Roger recommends - wider apart and close to the rear wall so that there is less room interaction.
The side strips and woofer cloths have zero affect on sound anyway, but if you want the best sound quality, you should get Roger to swap out those aluminum coned woofers for the original paper coned ones (if he still has the stock).
They are superior in sound quality.
He only went to aluminum because people kept blowing up the paper ones (god knows how).

FWIW, I own Sanders Model 10e and a pair of Magtech's.
I owned my first InnerSound speakers in 2002 and have upgraded every time but stayed with Roger, because in my opinion he makes the best speakers, period.
I also own a Lumin X1 and IMHO the Lumin sounds way better through a tube amp than it does direct using Leedh processing.
Leedh just sounds digital to me and is not very organic, but hey, each to their own.

I have mine setup in a loft with 18' ceilings, hardwood floors and zero damping panels of any kind.
Why? because this system and these speakers seem to work wonderfully in any room (and I've had them in over-damped rooms and in apartments with lower ceilings with concrete floors and walls.

The 10e speakers clean up and sound the best when you put them on IsoAcoustics Gai III (at least to my ears - never any buzzing and I play a lot of early Rocksteady and Reggae).

Cables is always a matter of personal taste but after many years and many $1000s my current cabling is:

Interconnects: Belden 8402 tinned copper balanced cable made up myself using leaded solder.
ESL Panels: Belden 9497 tinned copper which is a near perfect match for the ESLs electrically. I use naked copper spades soldered with lead solder.
Bass: Western Electric 10 gauge NOS tinned copper cables from the 80s, again using naked copper spades soldered with lead solder.
Bare wire does not sound any better than a naked copper spade.
Bare wire will eventually tarnish and you'll have to strip it and start over.=

Anyway, enjoy the Sanders and stick with them.

 

[EvilTed]

Oh and the best way to get precise imaging is to use the same mic stand you use to set the sound (I hope you use a mic stand) and get an attachment for a laser measurement device.
Using this device you can set the distance precisely from the panel to the midpoint of your chair.
You will still need to use the flashlight technique to set the angle of the panel to the midpoint of your chair.
You will be surprised how inaccurate you have them set without a laser and immediately hear the difference once you do.

 

[tmallin]

Thanks for the mike stand laser measurement tip. I may well try that. I do use a laser measuring device already, but not on a microphone stand. I measure from the head of a quilting pin driven into the exact center of the top back of my listening chair. That pin head provides a fairly stable reference for the back of my Leica Disto measuring device.

Roger currently recommends 60-degree subtended angle set up for the 10e. That's what I use.

Using the speakers close to the wall behind them will of course ease the load on the woofers and allow them to produce more and deeper bass. The old paper-coned woofers would be fine in such a situation, of course.

However, Roger's recommendation of near-wall use for the panels is a head-scratcher to me. You do NOT want strong, early reflections from the wall behind the speakers in the mids and highs. This is true for most any speakers. This can only blur the otherwise pristine panel sound, especially since the rear output of the dipole panel is as strong as the front output.

Ceiling height is irrelevant since the speakers are extremely directional above the bass in the vertical dimension given the 42" panel height. Even floor reflections are irrelevant for this same reason and the floor is of course much closer.

Naked copper connections initially sound great, it's true. But naked copper spades (I assume by naked you mean unplated copper) will also quickly tarnish and need cleaning. Gas tight connections (which thus do not tarnish) on spades are difficult to make using ordinary speaker binding posts. Gas tight connections to solid copper wire are almost automatic with electrical circuit breakers which have a knife-edge connector which automatically dents the solid copper wire ordinarily used in house circuits when the screws are tightened down. The nick in the copper stays clean over many years. You can only reliably make gas tight connections even with soft copper spades by clamping a leg of the spade into the hole in the post and clamping down hard on that, bending and otherwise deforming the leg of the spade. The same technique works more easily with bare solid copper wire or banana plugs.

Adding a preamp, tubed or otherwise, will result in not using the digital volume control, it's true. Each person must determine whether the added analog stage is sonically an overall better solution than any deterioration caused by using a digital volume control, LEEDH-processed or otherwise. To my ears, digital volume control as implemented by the Lumin X1 is the best-sounding volume control solution I've heard, even better than the all-analog Benchmark relay-controlled stepped attenuator solution in their line preamps, or the all-analog stepped attenuators EVS used to market as their Ultimate Attenuators. I have not heard tube preamps as competitive volume adjusters for decades. Perhaps you are using the tube warmth and fuzz to compensate for the high-frequency edginess caused by your improper close-to-wall speaker set up.

My Sanders 10e's are necessarily mounted on a carpeted floor. Spikes are really not an option when heavy speakers like the Sanders are mounted on a carpeted floor and you want to minutely adjust the position of the speaker for toe-in and distance from the listening position. I use the flat-bottomed hardwood floor feet Sanders now provides. If you mount the speakers on a hardwood surface, you could use spikes and still nudge them into position, especially if you used some sort of flat-bottomed cups beneath the spikes. I'm sure the Gaia's work great on your hardwood floor, but over the decades I've consistently preferred the sound of "floating" speakers on carpet or cushioned footers, not spiking them to the floor.

I surely understand why some listeners would prefer a very live room with these speakers. That aids the uniformity of frequency response in the rest of the room outside the sweet spot and makes the speakers' presentation even "bigger." In my quite small room, however, damping the corners behind the speakers yields a better overall sonic result, to my ears.

 

[EvilTed]

The manual that accompanies the Sanders 10e hasn't really changed since the InnerSound days in terms of speaker positioning.

It states:

"The speakers are designed to have a hard reflective wall behind them -- this will disperse the high frequencies throughout the room so they sound good when you are out of the sweet spot. So do not put sound damping material behind them unless you only listen in the sweet spot and don't care about the sound when you are off - axis."

Personally, the sound fills the room really nicely, anywhere in the room.
In the sweet spot it is noticably better, yes, but that is for critical listening,

It also states:

"The speakers may be positioned close to a wall -- any wall, side or rear walls work equally well. You do not have to place them out in the room."

It goes on to state that:

"It is much better to place the speakers randomly in the room so that they are each at different distances from walls and corners"

I meet all of the above criteria and have no problem with buzzing. Never have.

It's not me that has the speakers improperly positioned
You are treating his speakers like Maggies or other panels. They are not.
The bass units are transmission lines and like being close to a rear wall.
As mentioned, his panels work well close to the rear wall.
Pull them out by all means, but the sound will be thin and anemic and you will be missing out on the part which makes the Sanders special - the powerful bass married to the stunning midrange of the ESLs.

I've had 20+ years experience with Sanders.
I know these speakers better than most people.
I've lived with them and used them daily for over 20 years and know how to get the best out of them.
The other important part is I've never felt the urge to change them.
I've upgraded them, but will never part with them.
They are that good.
I'll wager I'll be using them long after you have moved on to the next best thing

The Lumin X1 is excellent.
I tried it with the Leedh, but didn't like the sound out of it.
I use a Audio Sensibility Signature DC cable with it.
I actually gave Steve the wiring diagram to be able to build it.

Anyway, I prefer a more organic liquid sound that tubes and ESLs give.
An analog front end with a stone bodied Koetsu, a good tube preamp with some vintage tubes and the Sanders is pure magic.
The Lumin I use for background listening

 

[tmallin]

Okay, so we have a long-term devotee of the Sanders speakers who loves them set up in a completely different way in a much different system context. This should go a long way toward encouraging potential buyers to investigate the speakers for themselves. Even if you're averse to room treatment, love tubes and analog sources, and want a decor-friendly near-wall set-up, these could still be long-term keeper speakers for you. What's not to like?

 

[tmallin]

In my small room, the key to good stereo imaging and staging seems to be minimal reflections off the room surfaces. With the back wave of my Sanders 10e damped with 4-inch foam on the walls behind the panels, there is much less mid and high frequency bounce off the room surfaces than with any other speakers I've had in this room. I can hear this by walking around the speakers; not much above the lower midrange is audible more than 20 degrees or so off axis.

Above the 172 Hz crossover, the 40-inch floor-to-ceiling electrostatic panel dimension pretty much eliminates any possibility of floor and ceiling reflections in the mids and highs. Even side to side, the 13-inch width dimension of the panel means that the panel is quite directional from 1 kHz on up. And since I sit where the wall reflection of the panel is seen edge on, there's really not much mid and high frequency sound bouncing off the side walls and reaching my ears directly, either.

The imaging and staging is quite precise, with a lot of variation from one recording to the next, depending on the miking and processing of the recording. And of course the sound is quite clean due to the electrostatic driver and lack of blurring reflections.

"Full range" panel radiation also distributes the sound origin over a uniformly wide area. There is a lack of a treble "hot spot" that ribbons and especially domes can create. The actual treble source is not focused on the one-square inch of a typical dome tweeter. I can hear this clearly by covering my one ear, putting the other ear just a couple of inches from the front side of the panel, and moving my head across the width of the panel. The treble subjectively emanates rather uniformly from all areas of the width of the panel in this test.

This large-panel treble radiation produces a very natural, open sound, and a sense of being easy on the ears in the treble at higher volumes in a way that ribbons and domes are not. With the Sanders, high volume listening does not cause temporary tinnitus the way such listening does with, say, Magnepans with long tweeter ribbons.

The distribution of sound radiation over the area of a large flat panel heard from fairly close up may work against imaging/staging precision because the distance from various spots on the panel to my ears varies. However, to my ears the lack of room surface reflections carries the day. The Haas/precedence effect makes the stage appear centered vertically at eye/ear level, while the distributed treble sound radiation prevents the treble "hot spot." Or so it seems to me, subjectively.

I'm trying to think of other speakers that have the "no tweeter" sound to the extent the Sanders 10e so obviously does. While I have not emphasized this aspect of the sound of the Sanders 10e before now, in fact this was one of the first things I noticed about the sound of these speakers once they were set up for near-field listening in my room.

With most speakers, if you listen in the near field you have to work hard on your set up to avoid hearing out an obvious tweeter location. Yes, with most speakers I've owned, good set up will avoid the tweeter sticking out even in the near field. With cone-and-dome speakers, this usually involves getting the tweeter level set correctly, using the proper amount of toe-in, and then adjusting your listening height to be just right with respect to the tweeter.

For example, with the Harbeth M40.2, the magic spot is a listening height an inch or two below the tweeter center. With the Stirling LS3/6, the coherent listening height where the tweeter location (either lower or upper tweeter) does not "stick out" from the overall sound is no higher than the height of the lower tweeter. The Dutch & Dutch 8c is rather easier in this respect than other cone & dome speakers I've used. I think the reasons for this include the fact that the design axis is well below the tweeter, the tweeter is time aligned with the woofer, and most of the high frequency range is pleasantly "relaxed" with respect to the woofer; these characteristics all militate against the tweeter sound "sticking out."

But once you eliminate the really small true source of treble, the "no tweeter" sound is taken immediately and strikingly to the next level with no real work at all. Just aim the panels of the Sanders 10e at your ears and sit so that your ears are at least a foot or so above the bottom of the panel and--voila!--no tweeter sound at all while still having some of the clearest, most extended, lowest distortion treble around.

Most panel speakers actually are divided up into frequency-specific strips or areas. The KLH Model 9 had small-area squarish tweeters. Many Maggies have a long ribbon tweeter. Then other panels are composed of time-delayed annular rings like the Quad 63 meant to create the impression of a point source or otherwise arranged to focus the panel in a point source manner like the Muraudio. Or the panels are faceted and at close range thus have a certain amount of vertical venetian blind effect which prevents an entirely coherent sound (Sound Lab, Acoustat). I find that most Martin Logan curved electrostatic panel speakers both blur and "edgify" the sound considerably. I have not heard the King Sound or Stax flat panel electrostats.

In my experience the Sanders flat panel electrostatics are uniquely capable of this "no tweeter" sound while yet having truly top-notch treble extension and subjective treble quality. There may be others out there with this quality but, if so, I have not heard them.