[Duke]

The AudioKinesis Zephrin 46 is now official!

Form follows function here... but it turned out to be a rather interesting form after all:



You can't really tell from the picture, but that's a milled aluminum "AK" logo at the bottom.

Okay for anyone not familiar with Jim Romeyn's Late Ceiling Splash (LCS) configuration, which is used here with his permission, the idea is to get a lot of spectrally-correct, relatively late-arriving reverberant energy into the room without the room-placement requirements of dipole, bipole, or omni speakers.   The extra reverberant energy is fired from the floor up towards the ceiling, which gives us a nice long path-length-induced time delay.   This improves timbre, spaciousness, and the sense of envelopment.  The long delay preserves the imaging cues from the front-firing array, and then the ambience cues that are on the recording come from more realistic directions instead of primarily from the direction of the speakers (which is probably the worst possible direction).  So we get imaging and envelopment.  Additional benefits include eliminating the baffle-step and mitigating the floor-bounce notch. 

During the R&D path that resulted in the upfiring LCS configuration, Jim tried out variations on the bipolar theme, and found that the best results were from firing the secondary array up at the ceiling.   This was counter-intuitive to me; I would have thought that having the reflections arrive from a more theoretically ideal angle would outweigh the longer time delay from the long bounce off the ceiling.   So kudos to Jim for beating me at my own game, and my sincere thanks to him for letting me use his technology.

The original Dream Maker LCS system uses two boxes per channel, and correspondingly takes up a lot of real estate.   The Zephrin 46 uses a to-the-best-of-my-knowledge unique geometry that shoe-horns the upfiring array into a reasonable-sized footprint.    We do trade off some capability relative to the Dream Maker LCS system, but the price is a lot lower too. 

What you see on the front there is two prosound 6" woofers that have very good thermal characteristics (high power handling combined with good efficiency), vertically flanking a small-radial-horn-loaded compression driver.   That particular horn is a very gentle device, no abrupt discontinuities, so it has very low coloration.  its pattern is not quite as uniform as we'd get from a true waveguide, but I haven't found a similarly small waveguide whose radiation pattern and smooth transition at the throat (critical for low coloration) are as good as this little horn.   The radiation pattern shapes don't quite match up in the crossover region, but the net pattern coverages do, so there is no power response glitch in the crossover region.

Around back is the LCS array, which consists of four upfiring 1" soft dome tweeters and two upfiring 6" woofers... the total of four 6" woofers is where the number "46" comes from.   This is a different 6" woofer, one better suited to its particular role (a bit less efficient, but with parameters better suited to take advantage of strong boundary reinforcement).  The crossover for the upfiring array is tailored to its location within the room. 

Each of the arrays is configured to present a 16-ohm load, and each has its own set of inputs.  So the system can be configured as an 8-ohm load (both arrays in parallel) or as a 32-ohm load (both arrays in series).  The impedance curves are smooth enough and similar enough that either configuration works.  The 32-ohm configuration is primarily intended for compatibility with OTL amplifiers.  I'm a big fan of Ralph Karsten's Atma-Sphere amps, and wanted to offer really good compatibility with his little S-30, which becomes a 50-watt amp into a 32-ohm load, with distortion reduced along the way.   Most people will go with the 8-ohm configuration, which is still a benign enough load to work well with tube amps.

The Zephrin 46 is even more room-adaptable than my other designs, as both arrays have independent bass tuning adjustability and tweeter "tilt" adjustability.   I haven't figured yet figured out what my recommendations will be for toe-in angle.   

The Zephrin 46 will be making its debut in one of the Electra Fidelity rooms, 920, at T.H.E. Show Newport Beach in a little less than three weeks.  Tony Chipelo of Electra Fidelity is a dealer for the Zephrin 46, as well as for the larger Dream Maker LCS system.

Finally, here are some preliminary specs:

Type:  LCS (Late Ceiling Splash) system

Dimensions:  42" tall, footprint 16" wide by 18.5" deep; 80 pounds

Impedance:  8 or 32 ohms, user-selectable (tube-friendly in either configuration)

Efficiency:  92 dB/1 watt.  Voltage sensitivity (relative to 2.83 volts input) is 92 dB in 8-ohm configuration, and 86 dB in 32-ohm configuration.

Power Handling:  400 watts RMS thermal, and estimated 200 watts mechanical (where the woofers start to go into over-excursion; this is of course program-dependent).   

Typical in-room bandwidth:  Upper 30's to 20 kHz

Price:  $4900/pair

Ever since I showed my Jazz Modules in 2006, I've had dealers asking me if they could sell my speakers.   Not every dealer that comes in the room of course, but enough (and some of these guys have considerably more experience than I do).  I always turned them away, saying that my margins didn't have enough slack in them to begin to accommodate dealer markup.   But I always hoped to one day have a design that had a high enough performance-to-build-cost ratio that it would be as competitive with dealer markup as the Jazz and its descendents were without it.   Jim Romeyn's LCS technology gives me that opportunity - imo it adds enough enjoyment to the experience that, at $4500, the Zephrin 46 is even more competitive than the Jazz Modules were at that same price when they were introduced eight years ago.  I want people to be able to hear these speakers, and that means I need to get them out there, and a time-honored way to do so is through a select dealer network.  I don't want a whole bunch of dealers, only a few.  I realize that going from direct sale to dealers is the exact opposite of the way the industry is moving these days, but dealers are the hardest-corest audiophiles of all, the most kindred spirits for a speakergeek like me, and I think Jim's LCS concept is too good for me try to keep it all to myself (not that I could... it's Jim's technology, not mine).   So I think the time has come.   I will continue of offer some of my more conventional models direct, but my best and brightest designs will be sold through dealers. 

[mamba315]

Wow, quite the looker!  Looking forward to hearing more about this speaker after T.H.E. Show.

[JLM]

Interesting.

My failing memory recalls (don't know which vendor/manufacturer) recommended using two pairs of monitors, one on stands and one on the floor below aimed up.  Similar concept, but I'm sure with a lot less thinking, efficiency, and power handling behind it.

[jtwrace]

Look forward to hearing them at RMAF 2014.  Do you have any Polar Plots?   

[THROWBACK]

Duke,

Fascinating, both for the fresh ideas about speaker design and for the beautifully-written presentation that captures your enthusiasm without making you sound like a marketing hack. You truly love this stuff and it shows. I can't wait to hear these speakers at RMAF.

Two questions:
1. There seem to be some sharp edges in the cabinet. Would the design be improved if these sharp edges were rounded, or a felt ring put around the tweeter or something?
2. Have you thought about combining these with a woofer array to improve bass?

[James Romeyn]

Interesting.

My failing memory recalls (don't know which vendor/manufacturer) recommended using two pairs of monitors, one on stands and one on the floor below aimed up.  Similar concept, but I'm sure with a lot less thinking, efficiency, and power handling behind it.

Howdy,
That was me.  I took the pages down where I explained the concept in detail.  And you are exactly right re. the refinement items above. 

I had virtually no interest in different radiation patterns till I met Duke.  I must shamefully admit I became more interested in Duke's "Vertical Offset Bipolar" radiation pattern after Dr. Robert E. Greene awarded Duke's original Dream Maker his Golden Ear Award in 2008 TAS, noting especially its alternative radiation pattern. 

Later Duke explained VOB in great detail.  I built and employed six matching Dynaudio Esotec monitors for my Trinaural based system.  Six matched monitors allowed a lot of experimentation, leading to LCS.  (Prior to LCS I preferred Trinaural.  LCS provides better spatial effects than Trinaural.)   

Duke contributed at least two major LCS advancements beyond the obvious items you listed above, providing about 50% of LCS audible benefits, including the bass mode damping feature we did not employ at RMAF.

Brian Cheney's (RIP) last two speakers were the RM-V60 "Wing" (dipole) and his last, the RM-50 (bipolar).  The RM-50 seemed to be Brian's best work.  I encouraged Brian to switch to bipolar radiation pattern based on my experiments at that time with Duke's VOB radiation pattern.     

[Duke]

Wow, quite the looker!  Looking forward to hearing more about this speaker after T.H.E. Show.

Thank you!  This was the first time the guy who runs the CNC machine at the shop where the cabinets are built said that one of my speakers was "pleasing to the eye".  At which point the guy who does the actually assembly said "yes but they're a pain in the butt to assemble".   I still haven't gotten the bill for his time yet, but based on the hours he told me, ouch.

My failing memory recalls (don't know which vendor/manufacturer) recommended using two pairs of monitors, one on stands and one on the floor below aimed up.  Similar concept, but I'm sure with a lot less thinking, efficiency, and power handling behind it.

That probably would have been "James Romeyn".  Yeah, Jim totally ripped off his idea, but don't tell. 

Fascinating, both for the fresh ideas about speaker design and for the beautifully-written presentation that captures your enthusiasm without making you sound like a marketing hack. You truly love this stuff and it shows. I can't wait to hear these speakers at RMAF.

Thank you very much.   My main contribution was being friends with Jim; he did the experimenting (and continues to do so, passing his findings on to me).  I think at one point in his quest I even told him that I didn't think bouncing the sound off the ceiling would work as well as bouncing it off the side wall (which was the front-runner at the time), and he tried it anyway. 

Two questions:
1. There seem to be some sharp edges in the cabinet. Would the design be improved if these sharp edges were rounded, or a felt ring put around the tweeter or something?

Roundovers would probably be some improvement, but not as much as you'd think.  The tweeter section is directional enough that it just barely "sees" the enclosure edges.   That's one of the reasons I like good horns & waveguides.   Now down in the woofer section, yeah the woofers see the enclosure edges, but it's not going to screw things up as much because it's in a region where the ear is less sensitive to such things.  The ear's sensitivity to diffraction peaks at about 4 kHz. 

2. Have you thought about combining these with a woofer array to improve bass?

That should work.  With all the ports plugged and speakers pulled out into the room, the -3 dB point probably moves up to 70 Hz ballpark.   One of the ideas was a speaker that would wouldn't need subs, but could be adapted to take pretty much full advantage of them if the opportunity arose. 

Okay one of the weak points of many conventional satellite/subwoofer setups is the lower midrange region, where the little satellite up on a stand and out away from the walls has serious baffle step going on, so the lower midrange is weak.   But the area of weakness extends up higher than the subwoofer can comfortably help out with.   Some standmount speakers incorporate baffle step compensation, but they trade off efficiency to get it.  And it some cases it's trading one problem for another.

The solution that the Zephrin 46 offers is, the wrap-around from the woofers in the LCS array fills in the baffle step, so the lower midrange has realistic "body" to it, without the reverberant field being unnaturally "thickened" because we're also adding the right amount of mids and highs to it. 

Another advantage of the Zephrin 46 with the ports either plugged or low-tuned for use as satellite speakers is, unless you're going for very high SPLs, you don't need to use a protective high-pass filter.   If you have one that's sufficiently transparent then go ahead and use it if it's a net benefit, but if not, well that's one less problem you have to solve.

So when used in "satellite mode", the Zephrin 46 may have some significant real-world advantages over more conventional satellite speakers.   Adding the Swarm (or subs of your choice) after your piggy bank recovers offers a fairly logical upgrade path. 

[Russell Dawkins]

Wow! You've done it again, Duke 

This looks like a completely fresh take on a number of fronts. A pretty brave design, I would say–a speaker for speaker designers! I like the impedance options, too, not to speak of the promise of serious dynamic capabilities.

Any chance pf a peek at the rear? Have you thought of a grill cloth design?

Was the slanted lower front section dictated by function?

[James Romeyn]

I drive Dream Maker LCS with three modified Pioneer 912 receivers, one 912 sits in off-center rack and acts as preamp, the other two 912s sit next to each speaker channel and act as stereo power amp (one channel drives the Main Speaker, the other channel drives the Effects Speaker).

It's interesting to note that Zephrin 46 owners could put two stereo amps to good use as above.  In this case each amp would see a 16 Ohm load.  AFAIK, for every single SS amp, the higher the load impedance the less offensive are distortion components (think more even order/less odd order).  The downside is less voltage potential, but depending on the application, 16 Ohm load can be a net gain.  Certainly and undeniably (again AFAIK) the higher the load impedance the greater is perceived speaker cable performance (the cable is smaller ratio of total load impedance).  IIRC Duke used about 20' of 18 or 20AWG speaker wire to good effect on 16 Ohm MVW at 2013 RMAF. 

[Russtafarian]

Duke in Newport!  That's cool.  I look forward to meeting you and hearing this speaker.

Russ

[James Romeyn]

Duke in Newport!  That's cool.  I look forward to meeting you and hearing this speaker.

Russ

Our dear Duke will be in Newport in spirit only.  Myself, Tony Chipelo, and Jack Elliano would be honored to meet you.     

[Duke]

Wow! You've done it again, Duke 

This looks like a completely fresh take on a number of fronts. A pretty brave design, I would say–a speaker for speaker designers! I like the impedance options, too, not to speak of the promise of serious dynamic capabilities.

Any chance pf a peek at the rear? Have you thought of a grill cloth design?

Was the slanted lower front section dictated by function?

Thank you, Russell!

I have some cosmetic work to do on the rear before it's ready for photos.  You'll see two woofers and a neat little row of four dome tweets, plus various terminal cups and ports.   

I have a grille idea under development, nothing fancy, just trying to make good choices.   I hate grilles and am inclined to heavily punish the wallets of anyone who doesn't share my sentiments, but as a practical matter, offering grilles could double my sales, as I see this design potentially making its way into a considerably wider range of living rooms than my previous mancave-rated models. 

Here's the evolution of the shape:  I wanted to scale down the two-piece Dream Maker LCS and shoehorn it all into one box.  The first stage looked like the letter "L".   Ugly, and deep footprint.  So next I sliced the wedge-shaped notch in the backside, to get adequate "landing pad" area for the upfiring drivers without the footprint being so deep.  That looked better.  Hmmm.   Maybe a complementary notch in front, adding a "chin" just beneath the top section?  That looked pretty good.  I played around with the proportions and internal volumes of the different sections a bit, until it looked about right, and would do what I wanted.  The relatively gentle (22.5 degree) backward-slant of the "Z" section still allows the energy to be upfiring, perhaps not as totally unimpeded as the L-shape, but at least I'll probably sell some of these.

So it's not the only form that would deliver the function, but it's the best-looking one I could come up with.

Duke in Newport!  That's cool.  I look forward to meeting you and hearing this speaker.

Thank you, but as Jim said, I won't be there.  I'll be at RMAF in October though. 

[James Romeyn]

Please let me encourage visitors to try the following, providing the floor is clear for safe passage: stand with back as close as possible to the front wall, at L/R center.  Close your eyes.  It should sound like you stand on stage in the middle of the performer(s).  Now slowly walk in a straight line toward the sweet spot.  Your brief voyage should sound like you walk toward stage front, then off the stage, and among the audience directly in front of the performer(s).

Upon request, if you desire to really do the above with eyes fully closed, I'll lead you safely from the front wall center to stand directly IFO the sweet spot chair.  I might bring a stick or a golf club for visitors to hold, just for this purpose.       

It's more than a interesting artifact, and relates to the overall listening perspective and spatial effects everywhere in the room.  Yes, as ever, the sweet spot is just what the name implies.  But stand up and walk to the farthest corners of the room.  See if you've heard similar spatial excellence and musical reality in any system regardless of cost or complexity.  Notice how small is the magnitude of bass mode effects, even in room corners.

Stand outside the door and notice the system displays superior "density" and live music qualities even at this location.         

[dburna]

Here's the evolution of the shape:  I wanted to scale down the two-piece Dream Maker LCS and shoehorn it all into one box.  The first stage looked like the letter "L".   Ugly, and deep footprint.  So next I sliced the wedge-shaped notch in the backside, to get adequate "landing pad" area for the upfiring drivers without the footprint being so deep.  That looked better.  Hmmm.   Maybe a complementary notch in front, adding a "chin" just beneath the top section?  That looked pretty good.  I played around with the proportions and internal volumes of the different sections a bit, until it looked about right, and would do what I wanted.  The relatively gentle (22.5 degree) backward-slant of the "Z" section still allows the energy to be upfiring, perhaps not as totally unimpeded as the L-shape, but at least I'll probably sell some of these.

So one question on the shape: does the complementary notch in the front serve a purpose to improve sound, or is it solely a cosmetic pairing to accent the backward slant?  Just curious.

Ooops, make that two questions: are these designed to go right in the corners of a listening room?  Or solely against the back wall?

Thanks,  -dB

P.S. Kudos to you for continuing to push the envelope, Duke.  I can't think of any speaker designer who works as much to "advance the art" in every subsequent speaker model.  I wish there had been more forward-thinking speaker designs at AXPONA this year instead of so many me-too boxes.  Perhaps my ears wouldn't have bled quite as much had that been the case.   

[James Romeyn]

Howdy dB (hmmm, how appropriate for this forum),

So one question on the shape: does the complementary notch in the front serve a purpose to improve sound, or is it solely a cosmetic pairing to accent the backward slant?  Just curious.

Duke never mentioned the former, so I presume till he replies it's the latter. 

Ooops, make that two questions: are these designed to go right in the corners of a listening room?  Or solely against the back wall?

Placement options should accommodate anything from corners to way out in the room (Cardas spec).  Such placement options result from the user's ability to run all eight ports open, closed or any combination thereof.  Also, Duke, not I, designed a natural mode cancelling feature that should accommodate corner loading.  One of the great features of the natural mode cancelling effect is that users can focus primarily on spatial and mid-treble effects, then fine tune bass performance later by itself where ever the speaker ends up.  Speakers lacking this option require users to compromise one performance aspect (bass) vs. the other (mid-treble). 

My experience with Dream Maker LCS indicates that LCS works well with simple, classic treatments like Room Tunes at the "usual suspect" locations (first side point, front corners, front center, etc.) 

Sorry to repeat this, but visitors should hear not only superb density, but also distinct image layering, huge stage dimensions in three planes, extreme "detached from the box" effect, and also music energy elevated higher than other radiation patterns. This last effect is different from the effect resulting from simply elevating the drivers ala YG Acoustic's Anat and Sophia (which I very much like, BTW).         

[JLM]

Duke,

Grille could save money (minimal veneer needed) if you do a whole cabinet sock like Vandersteen I or 2.  And you wouldn't have to answer all those stupid shape related questions.   

[Duke]

Look forward to hearing them at RMAF 2014.  Do you have any Polar Plots?   

Sorry I overlooked this question! 

I ran a suite of off-axis measurements in two planes (vertical and horizontal) following the protocol that I learned from Earl, and used that to generate my target curves, but I don't have polar plots per se.  Just families of off-axis curves.  And sorry, I don't publish my in-house measurements. 

I can tell you that my target curve for the LCS array is not what anyone would call "flat", and neither is the "this sounds right to me" curve that I ended up with after the target curve got me in the ballpark. 

So one question on the shape: does the complementary notch in the front serve a purpose to improve sound...

Hmmm... let's listen... ka-ching! ka-ching!  That's the sound of future profits!  Sounds like an improvement to me!!

Okay seriously, it's just eye candy.  But I don't think it does any detectable harm. 

I did read something once that recommended head-sized speakers for reproducing the human voice.   Well I doubt that even Andre the Giant had a head as big as the top section of the Zephrin 46, but maybe my marketing department can put some such spin on it.  I'll have them look into it. 

Sorry, I had way too much fun with that question.

Ooops, make that two questions: are these designed to go right in the corners of a listening room?  Or solely against the back wall?

The wide range of port tuning adjustability makes corner placement more likely to work than with most speakers.  I didn't design with corner placement specifically in mind, but before too long we'll have tried it.  Might have to go all the way to both top and bottom sections sealed, but I'm thinking that one port still open (probably on the bottom section) would work.   

P.S. Kudos to you for continuing to push the envelope, Duke.  I can't think of any speaker designer who works as much to "advance the art" in every subsequent speaker model.  I wish there had been more forward-thinking speaker designs at AXPONA this year instead of so many me-too boxes.  Perhaps my ears wouldn't have bled quite as much had that been the case.   

Thank you, dB!  My listening pleasure is measured in units of you. 

I have been EXTREMELY fortunate to have people who are smarter and more creative than me in their respective areas of expertise give me the chance to work with their technology.  I can only take credit for recognizing when it was time to play the "if you can't beat 'em, join 'em " card. 

There are few things I enjoy as much as designing speakers.  I do try to push my boundaries a little bit every time even if it's in ways that are obscure, and I've had a lot of "learning experiences" that you don't see.  Probably anywhere from five to twenty a year that fail to make it beyond prototype. 

Duke,

Grille could save money (minimal veneer needed) if you do a whole cabinet sock like Vandersteen I or 2.  And you wouldn't have to answer all those stupid shape related questions.   

I think this is the first time I've had a speaker whose shape was actually a net positive attribute, so I'm going to show it off!! 

Regarding the grille, I'm leaning towards a super-Velcro to hold it in place, along with a bit more breathing room underneath the edges of the grille than what you normally see (credit to Brian Walsh for helping me out with this).   The grille may or may not happen before the Newport Beach show.   

[borism]

I have been following the LCS technology with great interest and have a simple question for Duke. Would drivers mounted on the top plate of the speaker provide similar benefits or is the additional time difference of sound travel from the drivers close to the floor (estimate ~ 3 feet difference) crucial?
Thanks,
Boris

[James Romeyn]

I have been following the LCS technology with great interest and have a simple question for Duke. Would drivers mounted on the top plate of the speaker provide similar benefits or is the additional time difference of sound travel from the drivers close to the floor (estimate ~ 3 feet difference) crucial?
Thanks,
Boris

Please excuse me for replying, just to hold us over till Duke chimes in later. 

Comparing Zephrin 46's placement of so-called LCS Effects system (drivers firing vertical) vs. your described placement.  Zephrin 46 has two features that I would definitely call "crucial."  The first is longer time delay.  The farther we are from the "10ms" ambiance target (Duke researches the source for this target number), the greater is the benefit for each ms of delay.  IOW, the farther we are from the 10ms target (11.3') the more "crucial" is each greater unit of measure (longer delay).   

The point with Zephrin 46's one box approach is to save cost and real estate vs. Dream Maker LCS.  The one-box approach reduces potential delay time for ambiance effects.  Your described system even further reduces delay to the point of it being "crucial" in my experience.

The other issue is Zephrin 46 provides helpful "Shadow" over LCS, thus decreasing off-axis output from the Effects and increasing ratio of Effects vs. on-axis from the Mains.  This difference is similarly "crucial" as the delay item mentioned above. 

I suspect the two combined differences would approach a night/day deficit.

Richard Shahinian's Obelisk tended in the direction of LCS, as does an up-firing system John Marks reviewed a few years ago (single 8" mid bass firing up IIRC).  Neither of these systems offer the same overall package nor performance, by a long margin.   

Hope this helps! 

[mgalusha]

I'm looking forward to hearing these in Newport.