Three Way Evil Design Study

[kvardas]

Feeling some major dark reverberations in the force...

[Evil Twin]

My master, the delivery is complete

Soon the salvaged Seinnar droids will begin work on the initial fabrication steps- the additional tooling and assembly fixtures have also been delivered.

[Juhazi]

Should we prepare ourselves for some remastered 3D? :E

Seriously, I'd really like to see your statement of this project, EvilTwin! The cab design is timeless an a high WAF. And no curves, relatively easy to make.

[Fishy]

I'd still like to know the design documents for those little speakers that Face made.

Fish

[Evil Twin]

All in good time...

Should we be prepare ourselves for some resampled 3D? :E

Seriously, I'd really like to see your statement of this project, EvilTwin! The cab design is timeless an a high WAF. And no curves, relatively easy to make.

The final documentation of the build plan is being completed...

The target is a classic configuration known to many, and the long time object of consideration with variations by this thread.



But after consultation and further design exploration, a somewhat radical implementation has been chosen, eschewing the use of conventional construction materials, and pursuing an all LBL constructed design, using Natural Fossilized LBL ply sheets as shown.




Past experience using LBL selectively in projects has highlighted its superior properties; the hardness, stiffness, and low resonance and durable, attractive surface make it a very interesting material. Discussion of how the material is used to advantage in the final design will be presented shortly. In the past difficulty has been finding a vendor for sheets of the necessary size. That has been solved, and 530 lb. of material were delivered to the orbital assembly facility yesterday.

It is expected that a construction can be completed without the use of veneering, further simplifying the build process, and finishing with a combination of tung oil and satin polyurethane varnish.

Woofers and midrange will be of the underhung Neodymium construction for low distortion; only the tweeter choice is yet to be finalized, pending the results of some measurements and audition.

[dsrviola]

"I'd still like to know the design documents for those little speakers that Face made."

+1

[gbegland]

Oh man, that is gonna score big points for the dark side if you make the baffle out of multiple layers, slice through it on an angle and reveal all of those lovely bamboo cross sections.

Greg

P.S. Was the source in your sector Cali Bamboo? How much does the dark lord negotiate his LBL for?

[5th element]

How much does the dark lord negotiate his LBL for?

The Dark Lord does not negotiate, he tells them how much he thinks he should pay. :roll:

[bbggg]

I thought that was Chuck Norris. :B

[Evil Twin]

Oh man, that is gonna score big points for the dark side if you make the baffle out of multiple layers, slice through it on an angle and reveal all of those lovely bamboo cross sections.

Greg

P.S. Was the source in your sector Cali Bamboo? How much does the dark lord negotiate his LBL for?

The Force is strong in this one...

[Bear]

Oh man, that is gonna score big points for the dark side if you make the baffle out of multiple layers, slice through it on an angle and reveal all of those lovely bamboo cross sections.

Greg

P.S. Was the source in your sector Cali Bamboo? How much does the dark lord negotiate his LBL for?

Plyboo from a distributor here in Houston is around $200/sheet for 0.75", IIRC. Compare that to Baltic Birch at around $50 for a 5'x5' B/BB sheet of 18mm (3/4 nominal).

[---k---]

ouch.

[Evil Twin]

That is very much the going rate, it seems...

[Evil Twin]

Some progress...

Large sheets require divide and conquer...





At the end, most of the base cuts were completed; front panel pieces and sides/backs ready for lamination.

[Evil Twin]

Business in the old Imperial sector is slowing progress...

But not stopping it completely...

The plan is nearly complete. Simplicity of construction is preferred....

[David G]

Why tilt the top section forward? If you keep it sloped backward in line with the bass section you will get a degree of time alignment.

[---k---]

I'm not seeing the simplicity, but I like it! :T

[Evil Twin]

There is a good and perhaps non-obivous reason-

Consider a seated position about 2-1/2 meters from the cabinets.





Of course, the crossover phase response influences everything, but using a Duelund target transfer function with a center frequency of 900 Hz and adjusting alpeh for nominal crossover points of 300 Hz and 2800 Hz, if the target is followed accurately down to at least - 20 dB, within this range all drivers will be in the same relative phase, something not accomplished with more conventional approaches to three way systems.



Standing will compromise this somewhat, but given the difference in the distance/length required for a low frequency phase shit, I'm not expecting serious issues. And due to the direction of the phase shift, the response holds on better moving above the ideal tweeter/midrange position than moving below it.

[David G]

Have you factored in the relative acoustic center of the midrange to the tweeter? Generally its about 1 inch behind.

[sdl2112]

I'm a little confused...is this a collaborative design with Jon Marsh? I see the same pic of the bamboo. Have you decided on the tweeter yet...I thought he was going to use the Transducer lab N26C. Will the design be ported, if so where will it be located?

[Face]

I'm a little confused...is this a collaborative design with Jon Marsh? I see the same pic of the bamboo. Have you decided on the tweeter yet...I thought he was going to use the Transducer lab N26C. Will the design be ported, if so where will it be located?

A friend and I just measured a pair last week, the FR was less than stellar.

[bbggg]

Have you factored in the relative acoustic center of the midrange to the tweeter? Generally its about 1 inch behind.

There is also electrical delay, with the driver following a low pass filter usually lagging behind the driver that follows the corresponding high pass filter.

[JonMarsh]

Curious...

A friend and I just measured a pair last week, the FR was less than stellar.

Some others over on Tech Talk have gotten fairly good results for SPL and distortion. Packing for more business travel tomorrow, but hopefully I'll have time to measure some of mine next weekend- planned for three projects.
Where did you see issues?

[5th element]

I'm a little confused...is this a collaborative design with Jon Marsh?

The clue is in Evil Twin's name :W

[benchtester]

I'm a little confused...is this a collaborative design with Jon Marsh? ....

You could say that.. But trust me you don't want to be around when they have one of their internal squabbles. :E :E :E

The clue is in Evil Twin's name :W

[cjd]

You could say that.. But trust me you don't want to be around when they have one of their internal squabbles. :E :E :E

Isn't that when the Emperor steps in? ...

[Chancellor]

Without doubt,

In this day and age collaboration is the order of the day...

Whether resources and talent comes from the old Imperial sector, or the Kallang Sector in Singapore, or somewhere in the Left Coast of a decaying republic, in the end it is the results which count.

While some favor the role of the auteur as the only guiding force and direction, in reality when a number of viewpoints are brought together and engaged fully, a larger picture and higher concept may be developed- one that still may require the intervention of the auteur, but not before the creative process is expressed through it's many channels for the Force. Each vessel and personage will bring a different perspective and possibly unique skills...

Hence, the whole can be greater than the sum of the parts.

[Face]

Where did you see issues?

I sent my friend an email, I'll see if I can get a FR graph for you. It did have a low FS and very low distortion, but the FR was not very smooth.

[JonMarsh]

On a true IB tweeter test setup, Jeff B. got very reasonable results-



And also got the kind of results you'd expect when mounted on an enclosure (16 X 9 baffle), and BSC and diffraction peaking raise their heads...



Curious to see any other data, obviously I've got to do my own once I can find some time... next weekend, hopefully.

Nearfield tells the tale... all else is implementation.

Otherwise, it's back to some 6640. What I'm really curious about is that I haven't seen anyone measure the N26c above 20 kHz.

[Evil Twin]

All of this speculation will come to an end once we have the data in hand.

Should the characteristics prove to be significantly flawed, we have more "traditional" solutions on hand from Accuton or Scanspeak. My instincts suggest that the Force is strong with this unconventional newcomer - in today's political climate it could be appealing to some if more high performance components were available from this hemisphere's sector. For that reason, several designs are in progress with the N26c.

[Face]

I emailed him for the graphs, just waiting. From what I remember, they don't look as pretty as the graph above.

[meb46]

Jon - looks like some great progress... you internals appear to be a "lot" simpler than mine... and I suspect a lot faster to assemble

I have managed to get the 3mm skins onto my top cabinets and now slowly working through the bottom Bass Cabinets. All cabinets essentially assembled and ports mounted, just the skinning and then the finishing to go... will get some more photo's up shortly...

Progress is very slow at the moment, travelling for 2 weeks on business, now I am heading away to Australia for a couple ofweeks to get married... hence not much speaker attention these days. Will get things moving again after I get back

[JonMarsh]

Jon - looks like some great progress... you internals appear to be a "lot" simpler than mine... and I suspect a lot faster to assemble

I have managed to get the 3mm skins onto my top cabinets and now slowly working through the bottom Bass Cabinets. All cabinets essentially assembled and ports mounted, just the skinning and then the finishing to go... will get some more photo's up shortly...

Progress is very slow at the moment, travelling for 2 weeks on business, now I am heading away to Australia for a couple ofweeks to get married... hence not much speaker attention these days. Will get things moving again after I get back

That's some great progress on your part- I completely empathize with the business travel- I'm in Houston right now, some meetings at HP about a custom project this AM.

Congrats on your upcoming marriage; hope you two have a great time and generate some marvelous memories! I've got three weeks in the US in April, then a work trip to Europe in late April/early May, and a personal trip to Europe for most of the rest of May- "Hiking from Vienna to Prague", that's what the Sierra Club calls this event. Been looking forward to this since last September.

Just have to see how much work I can squeeze in over the next few weeks... including getting the Arvo Part Picante in playable condition so I have something to listen to!

[mante]

Just have to see how much work I can squeeze in over the next few weeks... including getting the Arvo Part Picante in playable condition so I have something to listen to!

Should be a little faster with the MiniDSP.

Sorry to get practical here, but is there a minimum room size you'd recommend for these Avalon inspired speakers?

Cj

[Evil Twin]

The sealed version pictured above should be more "friendly" to smaller rooms due to the difference in LF roll off- the larger ported version in construction in Singapore will be anechoicly flat to a lower frequency range, requiring greater distance from room boundaries if subjectively flat LF response is desired. If a more Rubenesque response is preferred, then closer positioning to walls will be allowable. Time permitting, I will post a comparative analysis in the near future.

[Face]

Here you go Jon.

[JonMarsh]

Interesting- thanks for sharing-

Any info about baffle mounting, measurement mic and distance? I usually do both ultra near field, which reveals the IB response, and of course, measurement on the target baffle design, which shows what I have to work with in a specific application. In one like what's discussed above, the tweeter will not be especially flat looking, for raw response, because the BS will be at a fairly high frequency- which creates more work for the crossover designer, but has potential to improve the off axis power response.

[JonMarsh]

Consider a seated position about 2-1/2 meters from the cabinets.





Of course, the crossover phase response influences everything, but using a Duelund target transfer function with a center frequency of 900 Hz and adjusting alpeh for nominal crossover points of 300 Hz and 2800 Hz, if the target is followed accurately down to at least - 20 dB, within this range all drivers will be in the same relative phase, something not accomplished with more conventional approaches to three way systems.



Standing will compromise this somewhat, but given the difference in the distance/length required for a low frequency phase shit, I'm not expecting serious issues. And due to the direction of the phase shift, the response holds on better moving above the ideal tweeter/midrange position than moving below it.

I wonder if you could pull off that target function for the tweeter with a RAAL 70-10D? They recommend no lower than 2800 Hz L-R 4th order, it would be OK at 2800 Hz, but a bit higher in the lower frequency range- which is rather attenuated, anyway. It would be close, though...

[Face]

I would try to track down a pair of OEM RAAL's to use instead.

[JonMarsh]

snooping around on the web...

and looking at measured distortion data suggests to me not crossing lower than 4 kHz. Sigh... reality strikes back again.

[gbegland]

and looking at measured distortion data suggests to me not crossing lower than 4 kHz. Sigh... reality strikes back again.

I don't even recommend the big ones at 2.8KHz for high spl.

Greg

[Face]

OEM will be fine down to 2.5K and a little lower for the average room.

[agrippa]

Is this one of the oldest threads still running?

I have lurked over this one, off and on, for years

[dlneubec]

Wow, those N26C results look very different from mine. Below is one in a waveguide 0-15-30-45-60 horizontal off axis, followed by one infinite baffle juxtaposed with an early waveguide.

BTW, this is for the N26C-A. There is now an N26C-G, which is about .7db more sensitive, but more importantly, has a rising top octave, which may do a little better yet in the waveguide. I will have a pair of the "G" version soon.

[JonMarsh]

Is this one of the oldest threads still running?

I have lurked over this one, off and on, for years

Well, as Dr. Tihanji in Munich used to say, "Slow work takes time..."

:roll:

Jon

In Vienna the last few days, leaving for the Czech republic tomorrow. Beautiful weather here for this time of year- couldn't be better so far!