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

The more I've used the dbx Driverack VENU360 app on my iPad to control and experiment with the speaker settings, the more I realize how superior its user interface is to the LCD display on the physical faceplate of the dbx unit for most all functions. It's just a lot simpler and easier to manipulate the touch screen controls of the app than it is to work the buttons and knob of the physical unit and usually many less steps are involved in accomplishing any given adjustment. I will be leaving the dbx LMS attached to my home network via its required ethernet connection.

Apple apps vary in how useful, simple to operate, and reliable they are. This one, apparently developed by Harmon, is an excellent one. The GUI is colorful, simple in layout, and informative, and the controls all operate quickly and easily with precision. Most adjustments only require touching and dragging. Typing usually is not required unless you want to rename something.

To avoid inadvertently changing any settings from what Sanders has preprogrammed, I have saved to the iPad's Photos library screen shots of every function's screen showing the preprogrammed settings. This also documents the settings I have custom programmed into the AEQ and/or PEQ filters to allow easy recall in the event of accidental erasure.
 
Last edited:
In his forum REG mentioned the concept of "anechoic stereo" when he was talking about the Gradient Experiment (scroll down to "All loudspeakers sound equally well" heading). As REG noted, in an anechoic room, the Gradient experiment found that all that really mattered for the speaker to be seemingly transparent was flat response. However, in an ordinary listening room, the speaker must be flat plus highly directional in order to seem transparent. REG noted the Gradient 1.3 came out of this experiment and that the 1.3 was designed to have minimal room interaction, producing something like "anechoic stereo" even in a regular listening room. REG mentioned he likes anechoic stereo, even if others don't. He also said that few recordings are actually recorded in such a way as to be meant to be listened to in rooms which do not add their own reflections or bass support.

I had said in an earlier message that speakers should "ignore the room." I was picking up on the "anechoic stereo" idea that the speakers should be designed to sound in a listening room as closely as possible to the way they sound in an anechoic chamber. A good part of this is to reduce audible reflections from room surfaces in the mid and high frequencies and doing what you can to smooth bass response. I agree with REG that I like this type of sound even if others may not.

Beside the Gradient 1.3 and subsequent Gradient designs (most of which I've owned in the past), a full-range flat-panel dipole speaker with very narrow dispersion would also seem to fill the bill of ignoring the listening room. Such a dipole panel would minimize early reflections off side walls, floor, and ceiling as long as the panel is vertical and aimed at the listener. It would also interact less with room modes since less sound is produced toward the top and bottom and lateral edges of the panel. The mid and high frequency room surface reflections from the back wave can be easily absorbed with a few pieces of foam on the walls behind the speakers since the back wave dispersion, like the front wave, is very narrow.

Now, I've been down the dipole bass road enough times (e.g., Carver Amazing Platinum Mk IV, Legacy Whisper, Linkwitz Orion, Gradient Revolution + added dipole sub towers) to know that getting high-level low bass and punchy bass with dipole woofers is very difficult indeed. Boxed cones just work and sound better down deep. So, for better bass, you add a woofer to the flat dipole panel and keep the crossover to it as low and steep in slope as possible, and apply EQ to smooth the bass. You end up with something like the Sanders 10e, the theory behind which REG so ably explained in 2017 in his Sanders 10e review.

I probably should have purchased the Sanders 10e a few years ago, but I had a probably atypical experience with the earlier 10c in a different room a decade ago; see my first couple of posts in this thread. It has taken me a while to work back my way back to the latest version of the Sanders 10 speakers.

In the interim I've reconciled myself to the idea that, with a dedicated listening room like mine and with me usually the only listener in that room, a speaker only has to sound great in the sweet spot. Thus, I no longer care if the highs are rolled off elsewhere in the room if the sweet spot can be this amazingly sweet.

And, if you want more highs elsewhere in the room, just undamp the back wave to taste. The sweet spot will remain almost as sweet, at least in my current small room listening to them from close up--58 inches panel to ear. The Sanders 10e sounds great in my small room even with no foam padding absorbing the back wave. With the Sanders speakers--unlike with all other speakers I've had in this space--my small listening room's own second-venue reverb is only really audible on sounds I know should sound non-reverberant, such as a radio announcer's voice. The D&D 8c was the best I'd previously heard in this room in that respect, but the Sanders seemingly are quite a bit better yet in this respect. They really need no foam padding at all even in my small room unless, like me, you really don't ever want your small reverberant listening space to announce its presence.

Actually, outside the room a bit down the hall the highs are not so rolled off, fairly similar to what I heard from the Dutch & Dutch 8c from the same position. It turns out that even with the Sanders' extremely narrow high frequency dispersion the sound still bounces off the diffusive surfaces behind the listening seat, out the open door, and down the hall. What I hear from my work desk a few feet down the hall is pretty well balanced for background listening while working. I wouldn't want super fidelity in my work space anyway since that would be too distracting.

And that's why I've concluded for now that the best one can do in speaker design is for the speakers to ignore the listening room to the maximum extent possible (helped along in this goal at the listening end by set up and room treatment) and that recordings meant to sound right that way will sound really right and the rest will just have to fall as they may.

As it turns out, with the Sanders in my room, "fall as they may" is usually at least as good sounding as this vast majority of recordings have sounded on any other speakers I've owned or heard. By "good" I mean low apparent coloration in the direct sound (which, with the extremely directional Sanders, is most of what you hear), extreme clarity, no apparent distortion on either transients or tones, instantaneous unrestrained large scale dynamic shifts, high SPL capability full range but sounding unusually satisfying at lower volumes, vanishingly low irritation from early reflections, and sound which is interesting, revealing, entertaining, and usually enthralling.

My recent forays into high-end headphones have also convinced me that there is nothing like a fine electrostatic headphone driver (e.g., my Stax SR 009S) for low distortion and clarity of sound without overbearing brightness or etch. And, while the stereo imaging and staging presented by unprocessed stereo headphones is not nearly as pleasing or natural as that from stereo speakers, totally removing the listening room from the equation yields the countervailing benefits of lack of early-reflection-induced distortions/grit/edginess and extreme stability of imaging and staging. Thus, moving toward speakers which interact less with the room seems a good move. Ideally, I want speakers which combine the lack of room effects and clarity the best electrostatic headphone listening provides with the pleasingly natural "out there" sonic presentation of speaker listening.

And that is pretty much how the Sanders 10e sound from the sweet spot in my small room. They are uncanny. They have very natural tonality, are matchlessly clear and clean sounding, have fantastically precise imaging as well as very large open staging when the program material calls for it, and have whip-snap dynamics that must be heard to be believed. They give me "big" sound in all senses, even in my postage-stamp-sized listening room. Yes, from 58 inches away, they look like gigantic see-through headphones. But that's the goal: they sound like electrostatic headphones (no room effects, extreme clarity, and totally stable imaging and staging) but with the "out there" speaker presentation we audiophiles crave, plus the best envelopment/surround effects I've ever heard from a two-channel system on recordings where phase is manipulated. Even when phase is not purposely manipulated, the Sanders 10e yield a more enveloping sound than any other two-channel set-up I've heard because they better reproduce hall ambiance and manage to get that ambiance to appear well beyond the speakers--behind, wrapping around to the sides, and even a bit behind me.
 
Any tall basically vertical dipole panel will have a low degree of floor or ceiling bounce. To quote from REG's Sanders 10e speaker review: "The panel operates as a pure dipole up to the point where the wavelength gets down to the minimum frontal dimension of the panel, which is 13 inches. This is the wavelength of 1kHz, and a little above that frequency the panel will become more directional in the horizontal plane with increasing frequency. (The panel is 40? high so directional behavior vertically happens much lower down. Floor and ceiling interaction is minimal above the bass.)"

Also, listening as close to the speakers as I do, when I place a mirror on my floor and look for the reflection of the panel from the listening seat, the reflection is from the part of the floor very close to my listening chair, close to my feet. That reflection is so close to my listening chair that my legs block the floor reflection to my eyes and ears in my normal seated position. I cannot see the mirror at all in my normal seated position.
 
Some might argue that a speaker like the Sanders 10e which bounces so little sound off the floor and ceiling will not sound natural on closely miked vocals or instruments since the listening room cannot contribute the expected bounce. This might result in a singer, for example, sounding somewhat "disembodied" or detached from the recording venue. If a singer were standing right in front of you in your listening room, the singer would not sound natural if there were no floor and ceiling bounce of the sound from the singer.

To this objection I say that a singer standing right in front of you in a small room is a "they are here" experience, something I'm not particularly interested in hearing, especially not when the room is as small as mine. Yes, I sometimes hear live unamplified singers and instruments in the living room concert series I attend, but the room is much larger with 40+ people present and thus the room contains a lot of damping and diffusion.

Any time you hear a closely miked singer in a concert setting, the singer is miked to be amplified through the PA system. While there may be floor reflections in that venue, they are not from the singer, but from the PA system amplifying the singer's voice.

The choice made by the recordist of such an event is to record the sound from the PA system at a distance thus capturing the floor and other room surface reflections, use a direct box to capture the close-up direct microphone feed of the voice with minimal room reflections, mike the PA system at close range to capture relatively little room sound, or mix these feeds together in some proportions.

Whatever the choices made by the recording folks, what I want to hear is what's captured on the recording as enhanced by the recordist's art, not the terrible small room second venue contribution of my home listening room's acoustics. With such recordings and with total studio creations which never existed in concert, I think you just have to let the chips fall where they may. Such recordings will never sound like a live unamplified performance in a concert hall.

But they may sound "good" if you just don't expect them to sound like a live unamplified event in a large venue. Sound engineers sometimes can do some very artful things. (My pet peeves are processing which robs music of most of its dynamic range or which drastically shifts the natural frequency balance of voices and instruments.)

I think Linkwitz is just wrong about allowing dipole speakers to use your listening room surfaces to "enhance" the space actually captured on recordings. That may work passably well for some people in larger listening rooms, but not for me in a small listening room. Even large home listening rooms do not sound to me like a hotel ballroom, much less a concert hall.

Of course, the argument is frequently made that in the final mastering process recordings are evaluated and tweaked to sound best in an environment believed to be closer to the typical consumer's untreated listening room where the mastering room and speakers will contribute some "sweetening" room sound, something which speakers like the Sanders will shortchange. But if this is true, the Sanders speakers should sound unconvincingly dry on most recordings compared to other speakers. I have not found that to be the case; just the opposite in fact.
 
Potential Use of the dbx Driverack VENU360 With Other Speakers

The dbx Driverack VENU360 is the unit Roger Sanders uses to control his 10e speakers. You can buy it separately from dbx, however, for about $1,000 and use it with any other speaker of your choosing. If you are looking for a modern hardware EQ (and more) electronic audio box with remote control and presets, I think this would be a valid alternative to the top all-analog Schiit Audio Loki Max.

Let me state right up front that I have not used the dbx unit with any other speakers besides the Sanders. But as a veteran EQ box user, I see no reason why it would not be dandy for correcting other speakers as well.

No, there are no nice tactile knobs to twirl on the dbx as there are on the Schiit, but if you want to use an EQ device from the listening seat so you can hear the results of your adjustments in real time, you probably will use remote control anyway and then the knobs on the faceplate are meaningless.

Yes, the Schiit has simplicity on its side. But the dbx costs $500 less, can provide excellent sounding AUTOMATIC equalization (the Schiit Loki Max has strictly manual adjustments with no on-board measurement software), and is MUCH more flexible in what it can do without having an overly steep learning curve as to how to manipulate it, at least if you use it with the free iPad app controller.

To use the iPad app, you will need to connect the dbx unit to your home internet network via ethernet. That is one area where the Schiit is definitely a simpler unit to use in a home audio system; it needs no internet connection at all to use in remote-controlled mode, much less an ethernet connection. Also, a hint: if you want to use the app to control the dbx unit, you probably will first need to go into the Utility menu via the dbx unit's faceplate controls, go to Network settings, and toggle the Network settings to have HDCP networking on.

For equalizing an ordinary speaker with a passive crossover, you would just turn off or zero out all the unused controls. From the app, this is a simple matter of touching each of the modules/functions shown on the home screen and making sure they are either turned off or not applying any correction. For passive speakers you ordinarily would only use the AEQ (automatic parametric), PEQ (manual parametric), or GEQ (graphic) equalization modules/functions.

Unlike the Schiit, the dbx can also "grow" with you if you need or want to try an electronic crossover for your speakers or time align the drivers while using an electronic crossover. Also, the EQ offered is much more flexible than that of the Schiit: there's graphic, manual parametric, and automatic parametric equalization on tap. Each parametric EQ band is user adjustable to your heart's content if you don't go the auto-EQ route or want to tweak things after using the automatic AEQ function. At least 14 bands of parametric EQ are available, 10 of which are used in the AEQ function.

Since the Sanders 10e as controlled by dbx unit is the most transparent to sources speaker I've yet encountered, I have to assume that the dbx unit is also transparent when used in the balanced analog in to balanced analog out mode with 24/96 internal digital processing--the way I've been using it with the Sanders 10e speakers. The dbx unit also seems quite noise free. I hear no hiss, hum, or other noise from the speakers when it is active even from just a few inches from the panels.

The AEQ process is the simplest to use I've yet encountered. You supply a mike stand, but the unit comes with a measuring microphone, the proper microphone clip to attach it to a stand, and a nice long mike cable. You set up the mike at the listening position pointing straight ahead, plug the cable into the front of the dbx unit and run the AEQ wizard. You don't move the microphone to different positions. The measurement process might take five minutes the first time, but only two minutes once you've done it a time or two.

The results of the AEQ process sound excellent for adjustments below 1 kHz. I have measured the results of equalizing the Sanders 10e speakers with the separate individually calibrated microphone which comes with the OmniMic V2 measuring system I use. The AEQ system of the dbx unit produces measured results which are quite smooth sounding and smooth graphing when I look at the results with OmniMic V2 using 1/6-octave smoothing and 2 dB per vertical graph division. The AEQ graphed results displayed on the dbx app closely match what I see from the graphed OmniMic V2 measurement results at all frequencies below about 1 kHz.

Above 1 kHz the dbx AEQ and OmniMic V2 graphs also closely agree. However, I totally agree with Roger Sanders that the AEQ corrections centered above 1 kHz should be zeroed out for best sound from his speakers. Whether that would be true for other speakers, I do not know. I do know that the AEQ results graph smooth and gently rolling off in the top octaves when measured with OmniMic. But whether because of some flaw with the dbx AEQ system, the extremely narrow high frequency dispersion of the Sanders 10e, or some combination of those or other factors, I totally agree with Sanders that, for his speakers, best sound is obtained by zeroing out any parametric corrections recommended by the AEQ which are centered above 1 kHz. If I don't zero out those corrections, despite the smooth looking graphs, the Sanders 10e speakers sound too polite and rolled off in the high frequencies.

If with your speakers the dbx unit also produces overly rolled off high frequencies if the AEQ is not ignored above 1 kHz, but you want to tame the highs from your speakers or particular recordings with the dbx unit, it is a simple matter to insert a presence range dip using a bell-shaped parametric filter, or just use a high-frequency shelf filter starting around 3 kHz and rolling off two or three dB above that. Something else which may work well is to just use the AEQ filters up there but manually adjust their amplitude so as to be less aggressive in their reduction of the high frequencies.

One other handy function the dbx provides is a master gain control for each channel. By inserting the dbx unit in the signal path just before your power amp you can use this master gain control to in effect act as an analog volume control on the input of your power amp to reduce the effective gain of your power amp. This allows proper system gain structuring by allowing you to use your upstream system volume controls (e.g., preamp or source digital volume control) toward the top of their range and will get the lowest distortion and noise from your electronics. Lower the master gain so that you can get just enough SPL on your "quietest" source material with your system volume control(s) turned all the way up. If you don't know what is your quietest source, I suggest using a Sheffield Lab CD, the BBC3 internet radio stream, or the WILL internet radio stream to adjust the master gain.
 
Last edited:
More on gain structuring with the dbx LMS:

Beginning on page 52, the dbx VENU360 manual has very specific instructions about how to properly set the gain structure of a system employing this device. It's not the easiest procedure to follow, however, since it involves disconnecting your speakers first and then running full-level pink noise through the system, adjusting each gain stage along the way so that the maximum output level of the upstream component exactly matches the input clipping level of the immediately downstream component. But if you really want to do it correctly, this is the way to do it.

Note, however, that if you are using the dbx LMS with the Sanders 10e speakers and Sanders Magtech amps, changing the gains settings Sanders uses for the dbx unit will probably result in the flashing output lights on the dbx unit no longer indicating the onset of clipping of the Magtech amps. Thus, you do this at your own risk if you adjust the dbx unit's gain controls away from what Sanders sets them for.

Page 53 of the manual also gives a shortcut tip that can help you get the gains structured pretty correctly much more easily. That tip says, in part: "In many cases, calibrating the gain structure between the mixer [source] and VENU360 may not be absolutely necessary as you can figure out where to set the VENU360’s analog input clip levels and input Router/Mixer gains based on the mixer’s [source's] max output level spec alone. For example, if your mixer’s [source's] maximum output level spec is +26 dBu, the VENU360’s analog input clip levels (located in the Utility menu) can be set to +28 dBu then the input Router/ Mixer gains set to +2 dB. The mixer [source] outputs and VENU360 inputs will now both clip at the same level (+26 dBu)."

In my case the specifications for my Lumin X1 source say that the maximum voltage output of the balanced analog outputs I use is 6 volts. The handy-dandy voltage to dBu calculator/converter says that 6 volts corresponds to +17.78 dBu or roughly +17.8 dBu. The idea of gain structuring will be to arrange the input clipping point of the dbx VENU360 to equal this +17.8 dBu maximum output level of my Lumin X1 which directly feeds the dbx unit.
Besides the master gain control for each channel which I discussed in my immediately prior post, the Utility Menu of the dbx VENU360 also allows wide adjustment of both the input gain and the output gain of the unit. Note that since these controls are part of the dbx device's Utility Menu, they are accessible only from the front panel controls, not the iPad app.

Now for best signal to noise ratio and lowest distortion through the dbx unit, the output gain of the dbx unit should be adjusted as low as possible while the input gain of the dbx unit is adjusted as high as possible. The lowest available output gain from the dbx unit's Utility Menu is +4dBu. I set the output gain to that level.

Now I picked the +20 dBu input gain selection as allowing acceptance of the full +17.8 dBu maximum output of the Lumin X1. To get the dbx unit to clip at the +17.8 dBu level, I added an additional 2.2 dB of input gain via the mixer input gains available in the dbx app.

Then I checked whether these gain settings allowed my amps to play music through my speakers as loudly as I would ever want from the quietest sources available. I used the WILL internet radio stream playing classical music with wide dynamic range to make this determination.

As it turned out, these settings did the trick. They allowed the Lumin's digital volume control to operate in its highest ranges, 80 to 100, to provide the highest SPLs I would ever want to hear from this quietest of all sources. All other internet radio stations and other program material from music files, Tidal, and Qobuz are at least as "loud" as WILL. Some are considerably louder, needing no more than a setting of 50 or so on the Lumin X1 volume control to reach the loudest SPLs I'd ever want.

With the gain structure set like this, hum and noise from the speakers when everything is turned on but no music is playing is totally inaudible when my room is at its quietest and with my ear within an inch of the electrostatic panels. That is true at any setting of the Lumin X1's Leedh-processed volume control, from 0 to 100. That is much quieter than most systems in my experience. Only when I was using Benchmark amps and DACs have I achieved results approaching this level of quiescent silence from the speakers.
 
Last edited:
Adding additional parametric equalization to compensate for diffuse field listening:

Some folks recommend adding a psychoacoustic dip in frequency response somewhere in the presence range (roughly 2 kHz to 5 kHz) to combat subjective overbrightness from speakers with flat frequency response in this area.

On some material--for example recordings of classical music in large concert halls--I find that a psychoacoustic dip centered at 3 kHz, with an ampltude of minus 4.2 dB and a Q = 3 to work well. I got the idea for these exact specs from the DSPeaker Antimode 2.0 DualCore manual, page 18, available here: http://www.dspeaker.com/fileadmin/datasheets/dspeaker/AntiMode20DualCoreEng.pdf

As that page says:

"The idea behind the 3 kHz dip originates from the human auditory system and diffuse field characteristics. The human hearing is less sensitive to diffuse fields in the neighborhood of 3 kHz. Because of a flat microphone response, many recordings have too much energy around the 3 kHz region, compared to the original listening situation (e.g. in a concert hall). A cutting PEQ at 3 kHz can be introduced to compensate. Whether or not this is beneficial is left to be judged by the user."

I don't know how those exact specs were arrived at. Perhaps it was just experimentation. Using the dbx Driverack VENU360 app, my current Sanders 10e-based system allows me to play around with the specs of the dip just by tapping and sliding on the screen of my iPad controller which I can hold on my lap as I listen. The specs are variable in real time as I listen. (With the Sanders speakers' extremely narrow vertical high frequency directivity, the glass of the iPad on my lap has little subjective effect on what I'm hearing--this was never true with other speakers.) I've tried varying the amplitude, center frequency and Q while listening. I keep coming back to these specs as sounding about right to my ears for simply miked classical music recordings. Others may hear things differently, but I think this is a good starting point.
 
I think that 3khz dip was called "The BBC Dip". Many British speakers had this integrated into there crossovers to give a more " British Sound" It resulted in a softer more gentle listenable sound.
 
With the Sanders speakers, absolute polarity changes are definitely audible as sonic differences. Whether one polarity is better sounding than the other is more difficult to determine and is sometimes impossible or at least ambiguous, however.

For most recordings, with most speakers, deciding what is the better sounding choice can be impossible or at least ambiguous. With many digital sources, the absolute phase of both channels can be reversed in real time mid-program so one can instantly compare the sonics of each choice. The Lumin App I use for playback offers this feature, for example. No physical switches or connections need to be reversed with such a digital phase reversal process.

But, first, many speakers are themselves not wired with all drivers in correct absolute polarity and this shows up on impulse and step response graphs of those speakers. Thus, some parts of the speaker's sound will be out of absolute phase with other parts. With other speakers, the pulses from each driver all go positive in response to a positive electrical signal, but the pulses are not time aligned. Relatively few speakers are both absolute polarity correct and time aligned. Absolute polarity changes will be most unambiguously audible on the relatively few speakers (Quads, Sanders, and the Dutch & Dutch 8c are among them) which are both absolute polarity correct and time aligned.

Then there are the recordings themselves. Most recordings will have somewhat messed up phasing characteristics because multiple microphones per stereo channel were used at various locations in space relative to the musicians, rather than one microphone per stereo channel where the two mike capsules are located in a quasi-coincident array. The various microphone pickups can be phase compensated during recording or post processing (I believe Reference Recordings, for example, does this) but I doubt whether this is usually done.

Then there are the wiring choices. With the balanced XLR-based cabling used throughout the recording and playback chains, there is always a choice of whether to make pin 2 or pin 3 the hot positive connection. By accident or design, balanced cables are not always wired up with the international convention (AES 26 since 1984) of pin 2 being positive hot. If some cables in the signal path are wired differently, this can create absolute phasing ambiguity, especially where many microphone/mixer channels are mixed down into one stereo channel. See this reference for discussion of this problem: https://www.prosoundtraining.com/2010/03/11/which-pin-is-hot-and-when-does-it-matter/
 
Yes, as earlinarizona says, some speakers incorporate a presence dip in their response already. The dip has been called the BBC dip or Gundry dip, among other names. If your speakers' inherent frequency response already incorporates such a dip, you obviously don't want to inject a duplicative dip unless you REALLY want to back off the immediacy of the reproduced sound. As far as I can measure, there is no such dip built into the response of my Sanders speakers. With the Dutch & Dutch 8c, the entire region from about 2 kHz to 8 kHz seems to measure a bit "relaxed" or down a dB or two. Adding a 3 kHz dip was sometimes helpful with those speakers, and sometimes not helpful at all. The inherent balance of the D&D 8c is very conducive to enjoyment across a very wide range of commercial recordings.
 
Tom and others, just for fun here, attached are good friends and clients system. Pat McGinty did the loudspeaker design, Larry Owens, aka the DEQXpert and Pat finalized the crossover, filter design, and room correction settings in the client's home and my wife Deb designed the rack in collaboration with the owner. The speakers are tri-amplified and use a DEQX DSP unit with a 32-bit DSP processor on board for digital processing.

Originally we used the incredible darTZeel NHB-108 amplifier for the mid-highs, however, the client's backup power generator was blowing fuses 1-2 x per week so we switched to all Magtech power. Yes, the Magtechs are bruisers with power to spare. That said, the incredibly low noise floor offered by the patented regulated power supply circuit Roger uses allows the amplifier to simply do what it is supposed to do (accurately amplify the musical signal) regardless of speaker load or power supply demand. Anyone who needs a bullet-proof, great-sounding amplifier should certainly consider it in the mix.

And yes Tom, despite not handling Sanders products in our dealership these days, your thread reminded me once again that the Sanders 10e speaker remains one of my favorite speakers of all time - especially for listening in the sweet spot in a small to moderately sized room.
 

Attachments

  • Gambee System.jpeg
    Gambee System.jpeg
    665 KB · Views: 14
Tom and others, just for fun here, attached are good friends and clients system. Pat McGinty did the loudspeaker design, Larry Owens, aka the DEQXpert and Pat finalized the crossover, filter design, and room correction settings in the client's home and my wife Deb designed the rack in collaboration with the owner. The speakers are tri-amplified and use a DEQX DSP unit with a 32-bit DSP processor on board for digital processing.

Originally we used the incredible darTZeel NHB-108 amplifier for the mid-highs, however, the client's backup power generator was blowing fuses 1-2 x per week so we switched to all Magtech power. Yes, the Magtechs are bruisers with power to spare. That said, the incredibly low noise floor offered by the patented regulated power supply circuit Roger uses allows the amplifier to simply do what it is supposed to do (accurately amplify the musical signal) regardless of speaker load or power supply demand. Anyone who needs a bullet-proof, great-sounding amplifier should certainly consider it in the mix.

And yes Tom, despite not handling Sanders products in our dealership these days, your thread reminded me once again that the Sanders 10e speaker remains one of my favorite speakers of all time - especially for listening in the sweet spot in a small to moderately sized room.
Do you think the darTZeel amp caused the fuses to blow? Or did I misread this?
 
No, not at all! The rebooting of the backup generator was the cause of the blown fuses. I forgot to mention the home was 300 feet from the Atlantic in South Florida so the generator was a real necessity. Let's just say as much as my friend loved his NHB-108 fuse changes was more than he wanted to deal with. The Magtechs were simply less sensitive to the occurrence and since we were already using two of them on the sub-woofers and woofers it was the logical choice to replace the darT. Seven years later here's our NHB 108 v2 still in the rack serving as our main showroom amplifier with nary a blown fuse:)
 

Attachments

  • Current System (1).jpeg
    Current System (1).jpeg
    53.5 KB · Views: 6
tmallin, great to read your walk through your Sanders Sound System. You have many good points on acoustics and system setup in generally and I read and learn with great enthusiasm. I though have some questions/suggestions. Take it for what it is worth :)

As you have 2 free outputs from the DBX, have you considered to implement 1 or 2 active subwoofers for even out the room response in the bass? That will also provide more total bass headroom and make the life easier on the Sanders speakers. Implemented correctly I have always found adding subs, have lifted the overall sound quality.

About EQ'ing over 1000 hz, one thing that can be done is measure in nearfield, about 1-2 feet away and then correct for speaker nonlinearities. Also working with room target curves I find is very important, to find the right spectral balance between low, mids and highs. A general rule is a downward tilt towards high frequencies is preferable. So 20 KHz should be down - 6 to -10 dB relative to 20 Hz. This is very much dependant on room size, listening distance, directivity on speakers, room treatment ect. I don't remember seeing any measurements of the in room frequency response, maybe you have a good target curve already.

As you already use DSP, have you considered using FIR filters or some advanced room correction software as Audiolense or Acourate? These are state of the art PC based software, that takes system fine tuning to the next level. It will also do subwoofer integration. Just some thoughts to consider, but you maybe all to busy enjoying the fine sound you have.

Best Sebastian
 
  • Like
Reactions: eurasian
Thanks for your comments and suggestions, schlager.

I have used subwoofers in this room before--the four units comprising the AudioKinesis Swarm. Yes, that sort of distributed bass source does provide measurably very smooth bass response, smoother than from the woofers of two speakers. But in my small room it also produces a lot of physical clutter. And, frankly, the additional measured smoothness was not significant in subjective bass quality terms. What matters in this room to my ears seems to be adequate bass extension and lopping off bass peaks. That can easily be accomplished with good speaker placement and electronic equalization. Bass dips, as Roger Sanders says, are not really that audible as long as they are not really broadband.

According to my ears and my measurements with OmniMic V2, As set up and adjusted, I appear to have plenty of bass headroom and extension at the target level down to 25 Hz. Even at 20 Hz the bass level is the same as the 1 kHz level. The bass sounds great, even better than it did with my Dutch & Dutch 8c speakers.

To further smooth bass response, Roger Sanders suggests setting up the speakers so that they are not symmetrical in the room. Symmetrical placement tends to amplify the effects of room modes since both speakers will be activating the modes in the same way. I'm sure this would work to cut down on the effect of room modes. But, as explained in an earlier post, in my small room, asymmetrical placement is not really practical or aesthetically acceptable to me.

Roger Sanders insists that flattening the 10e response with EQ above 1 kHz is not helpful. His point is that the electrostatic panels are inherently flat in response above that frequency and that measurements of his very narrow dispersion speakers above that frequency yield deceiving results. I have to concur. If I flatten the response to make the plotted OmniMic V2 curve flat or smoothly rolling off in the high frequencies, the speakers sound quite "dead" in the high frequencies from the sweet spot. This is true whether or not the back wave from the speakers is damped with foam.

The dbx unit's Auto EQ function will flatten the OmniMic V2's measured response incredibly well if I implement the two parametric filters recommended by the AEQ function above 1 kHz. But the sound belies the measurements. I have also tried measuring the panels from closer up than my 58-inch-from-panels listening position. This does not help; even directly on axis the panels seem to measure even more rolled off from a distance of one meter or less.

Thus, I have to conclude that Roger Sanders is correct: for his speakers, measurements above 1 kHz should be ignored and the response above 1 kHz should be left alone, at least in terms of foundation equalization. For that reason I've chosen not to publish my measurements of the speakers.

Particular recordings may benefit from insertion of the psychoacoustic presence range dip I've described in prior posts (centered at 3 kHz, magnitude -4.2 dB, Q=3) or a shelf filter Sanders recommends (high frequency shelf at 3 kHz, Q=3, magnitude of 2 to 3 dB). But the majority of recordings sound wonderful in the highs with no correction at all.

Acourate and Audiolens are indeed regarded as state of the art in terms of their ability to correct frequency response to a target curve with surgical precision. If I were to add those distributed subwoofers you suggested so as to be able to fill in the bass null I tend to get around 50 Hz from stereo woofers in my room, those programs would be able to achieve any given bass target curve with surgical precision and time align all the bass sources as well.

But I think I'd have to run the sound through a very capable additional computer to do the convolving required by those programs. I don't think they can run on the Roon Nucleus+, Lumin X1, or the dbx VENU360 I currently use. My "server" for my local music files is a 1 TB USB stick inserted into the USB ports of my Lumin X1 (for listening via the Lumin App) and my Roon Nucleus+ (for listening via Roon); I do not use a separate computer as a server for my local music files.

I tend to distrust heavy-duty digital processing and hard-working computers. Ordinary digital audio processing up through 24/192 is very easy for modern computers. But, In my experience, additional digital processing such as upsamping, resampling, conversion from PCM to DSD, etc., tends to add false brightness to the sound. This is so even for the pretty capable Roon Nucleus+ computer where resampling to DSD 512 just about maxes the thing out in terms of processing load. And whether it's because of power line interference or other reasons, even having my Apple iMac work computer turned on in a different room and running on a different electrical circuit degrades the sound I hear in the audio room. This is so even though that Apple computer is not in the audio signal path.

In addition, perceptual science research indicates that our ears/brain tend to perceive sound on a 1/3-octave band basis. Thus, even below 1 kHz, corrections where measurements look flat or smooth with 1/6-octave smoothing are subjectively "good enough." I can easily achieve such smoothness below 1 kHz with the available parametric filters of the dbx VENU360 (except for the 50 Hz null, of course). The measurable correction precision possible with programs like Acourate and Audiolens is thus arguably lily gilding rather than sonically significant.
 
  • Like
Reactions: schlager and dan31
Thanks for your response. You have advocated greatly for your stand points, so I'm not going to go further into that. I will say that my standart laptop is using 1 % pc power, for running the convolution file. I use the freeware DRC Designer to make my impulse correction files. It operates basicly in the same way as Audiolense and Acourate, correcting in both frequency and time domain. I have not experienced any problems with different types of music files or sample rates running the convolution. I take the output in digital format from USB to S/PDIF and use my minidsp as DAC and all is running smoothly.

Best Sebastian
 
  • Like
Reactions: eurasian
If Audiolense or Acourate work for you in your system, that's great. It's just that with a speaker like the Sanders 10e that is designed around a particular active crossover/equalizer, I would hesitate to second-guess the designer as to what is the best way to get the intended performance from the speaker when it comes to designing and/or implementing the crossover and equalization.

On the other hand, one can argue that except at the very highest price levels, perhaps the components are designed to a price point or are designed with the abilities of most potential users in mind. I doubt whether most potential buyers of Sanders speakers feel technically inclined and capable enough to throw out the designer's crossover and EQ box, start over with a custom computer-implemented crossover and EQ design, and get superior results.

Note that another manufacturer of hybrid electrostatic speakers, David Janszen, recommends an even lighter touch than Sanders has adopted in terms of equalization. For his discussion of the EQ topic, see this link: https://janszenaudio.com/blogs/issues-in-audio/room-correction
 
Last edited:
My response in post 76, was only to your comment "I tend to distrust heavy-duty digital processing and hard-working computers."
My point being, a PC/laptop does not represent a bottleneck in a hifi system running convolution. And with freeware as DRC Designer, one can get their feet wet with IR correction for 0 $. That is a cheap possible upgrade that I find intriguing to try out.
It's only an option to change the x-over, one can do IR correction without touching the x-over point. The software will only remove the excess phase component in the x-over, not change the x-over point itself.
 
Last edited:
Further thoughts on gain structure:

In post #66 I said:

Now for best signal to noise ratio and lowest distortion through the dbx unit, the output gain of the dbx unit should be adjusted as low as possible while the input gain of the dbx unit is adjusted as high as possible. The lowest available output gain from the dbx unit's Utility Menu is +4dBu. I set the output gain to that level.

Now I picked the +20 dBu input gain selection as allowing acceptance of the full +17.8 dBu maximum output of the Lumin X1. To get the dbx unit to clip at the +17.8 dBu level, I added an additional 2.2 dB of input gain via the mixer input gains available in the dbx app.

Then I checked whether these gain settings allowed my amps to play music through my speakers as loudly as I would ever want from the quietest sources available. I used the WILL internet radio stream playing classical music with wide dynamic range to make this determination.

As it turned out, these settings did the trick. They allowed the Lumin's digital volume control to operate in its highest ranges, 80 to 100, to provide the highest SPLs I would ever want to hear from this quietest of all sources. All other internet radio stations and other program material from music files, Tidal, and Qobuz are at least as "loud" as WILL. Some are considerably louder, needing no more than a setting of 50 or so on the Lumin X1 volume control to reach the loudest SPLs I'd ever want.

With the gain structure set like this, hum and noise from the speakers when everything is turned on but no music is playing is totally inaudible when my room is at its quietest and with my ear within an inch of the electrostatic panels. That is true at any setting of the Lumin X1's Leedh-processed volume control, from 0 to 100. That is much quieter than most systems in my experience. Only when I was using Benchmark amps and DACs have I achieved results approaching this level of quiescent silence from the speakers.

Further listening showed I needed a bit more gain on really quiet sources and that, subjectively, the system lost a bit of dynamic slam as well as a bit of depth perspective with these settings. The sound also seemed a tad less clean and clear at times.

Changing the input gain setting to +20 dBu with no additional input gain via the mixer input gains, and changing the output gain to +8 dBu remedied these issues in one stroke with no subjective change in the quiescent silence from the speakers.
 
As I've explained in prior posts in this thread, I have come around to the idea that extreme high-frequency beaminess (as the Sanders 10e speakers provide) is the way to go in speaker design, especially if you want to maximize the subjective "you are there" quality from the best-recorded classical music recordings. You definitely want the "you are there" experience from such recordings, rather than any sort of "they are here" presentation. And as it turns out, the Sanders design works especially well in my small room where I listen from close up (58 inches from panels to ears), with the panels aimed directly at my ears and with the back wave from the panels absorbed by foam on the walls behind the speakers.

On the other hand, classical recordings made like the average DGG, with lots of mikes placed very close up and above the instruments, will just join the 95% of other recordings which are not of classical music in a concert hall. With this vast majority of recordings, you will just have to let the chips fall where they may.

Actually, most recordings sound very fine on the beamy Sanders 10e even though most recordings have "no there, there" in terms of natural recording venue ambiance, or at least don't capture enough of it. For most recordings the engineers have added some artificial sweetening intended to mimic some sort of ambiance and the Sanders will well capture whatever the engineers put on the recording in terms of artificial ambiance or other spatial effects. Some of these spatial effects, especially in non-classical music, are quite fascinating when heard to best advantage as they can be with beamy speakers.

The sonically worst recordings for such speakers are ones where the miking is at close distance and everything is panned hard left and right and maybe with a center channel thrown in to avoid the hole in the middle, and there is no artificial ambient sweetening, just dry, close-up sound. But, then, those recordings sound pretty bad on most speakers, but not quite as bad as on the Sanders since your listening room will not add much of any ambiance of its own. Most monophonic recordings are not too bad, actually, perhaps because many monos were made back when the mikes weren't so close up, so some sound from the original venue is captured.
 

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu