The Satori MTA- a two way, but with a checkered past and lots of different DNA

[JonMarsh]

This is a project that you might say has multiple ancestors and source of design DNA, as far back as 20 years ago, and as recently as six months ago. And, just to get this out of the way, it's NOT another 6-1/2" two way, but rather, an upgrade? A 7-1/2" two way... and with well pedigreed parts.

And it's being driving partly by the desires of one long term HTG member, Darryl Glenn, and by my own speculations and itches.

So, what are the historical antecedents? (This is sort of like answering a question such as, "who in your family do you resemble?" Which can be very complicated, or obtuse, if who you resemble was never living when you were born, and a key example goes back to the mid 1800s, in my case). But no time span like this.

So, I'd list these items as being in the genetic soup of this project:

• M8ta, circa 2006/2007, for a roughly 8" two way in a tower format speaker similar to Avalons- too bad Charlie Hansen passed nearly 10 years ago, he could tell you about the prototype speakers I gave him when my development and business partner died from an aortic aneurysm when I was away at CES in 1980 scoping out the competition. Charlie was working for me in a small professional audio outfit in Boulder Colorado while attending school at CU. More than 45 years ago.
• NatalieP, also a two way, with smaller drivers, and an unusual crossover with a low turn-over frequency to get the most out of an Usher tweeter made for PE, the RS28a.
• SMJ Audio Calliope, with the SB Satori MW19P-8, and the Peerless DA25TX-008 Corundum dome tweeter. IN and Out of availability, never goes out of style.
• Ardent D, for overall enclosure concept, materials, and finishing options (possible to get a nice result with just golden Watco oil and wipe on polyurethane. Some modifications to dimensions and configuration, of course.
• Modula Magnifique, based on the concept of a Bessel fourth order high pass, given specific requirements for driver Qts, enclosure size, etc. are made. Complete analysis still pending.

So, I'd say the enclosure details are pretty close to being fully nailed down, with one special option set-

• for the more budget minded, a reflex port in the base
• For those favoring passive radiators, a pair of the new Dayton SS10's might be optioned... or the RSS265PR. The SS10's if you want a 4th order Bessel high pass. The RS265PR if you want options for a flatter measuring system, with a heavier low end.

For now, a few renders of the enclosure.


Just the facts, ma'am, as seen with basic visualization.



An X-Ray view of the ported version:





And a very basic rendering visualization:




For reference, this build will be done similarly to the Ardent D:

• Maple ply front panels, as used for piano tuning plates
• Maple ply top panel
• Bamboo board, not ply, the kind made with 3/4" wide strips, for the side walls and back.
• Back from combination of bamboo board and 1/2" Baltic Birch Ply, new old stock in my case.
• Internal bracing materials primarily from 1/2" and 3/4" Baltic Birch Ply.

[JonMarsh]

Well, a lot going on today, and I forgot to mention the drivers:

• Woofer: Satori MW19TX-08 Textreme woofer
• Tweeter: Peerless DA32TX-008

Of course, a pic...

[JonMarsh]

Yes, I get a bit of a buzz just looking at them and handling them!

[JonMarsh]

Let's talk possible low frequency alignments- something sexier than just sealed. Of course, this is an area where DIYW can always rule, but perhaps some examples will be useful.

I examined two types in detail-

• A downward firing double flare port
• A passive radiator setup- many possible variations, but I've been having fun modeling with the new Dayton SS10, I have four currently on hand; though truthfully, I think it is more suited to Evil Twin, as it has some interesting characteristics, but will lead you into a design that seems influenced by the Dark Side.

Let's start with the vented alignment-




The port diameter is a critical parameter- too small and the high air velocity results in compression and distortion at higher SPL; too large and port resonances at significant amplitude will occur in the working pass band.

Like this one:


With the 3" diameter port, things are fairly reasonable, if not quite perfect, and port resonance is fairly low and port air speed is within reasonable bounds.







This is for a port construction flared at both ends. ET says don't even consider not using flared ports..


Lastly, the LF impedance curve, it's almost like a smiley face...

[JonMarsh]

Now, with ET looking over my shoulder, I'm going to take a trip on the dark side- PR's on the back side of the enclosure, and to make the challenge even more interesting, let's consider one of Dayton's newest high value parts, the SS10. I ran through many alternative configurations, and even flirted with a full 4th order Bessel alignment, but in the end settled for something a bit more conventional... but given the parameters of the SS10, not without it's quirks. PR alignment quirks include the relatively large steps in MMP depending on how many weights one uses and how large each weight is; sometimes one might prefer smaller steps, and that is possible, but requires visiting your hardware store. Here, we're keeping it simple... sort of.



These PRs will always be loafing, but hey, so are many modern cars at legal speed limits, right?

Baseline tuning target for Fb is similar to the ported alignment, but there are no port resonances or similar issues. For various reasons, I wound up with a Fb of 30Hz.




This was partly to wind up with a similar max woofer excursion for the same input power limit at low frequencies as the vented design, and Qmp of the PRs is a factor in that choice.



Another smiley face impedance plot, sort of, this time with a bit of a wink and a nod...




Yesterday I also ordered the maple ply for front panels and top, and two more Dayton SS10's... glad I have all the other needed wood on hand!

[JonMarsh]

And after two design reviews, have switched to RSS210-PR, have a near identical curves- the SS10 are just a bit too big for this enclosure, and given the limits of width for the maple ply, expanding it isn't feasible.

[technodanvan]

Hi Jon! If you'd like to experiment with optimized ports I'd be happy to print a few for you. Max length would be under 350mm on my printers, but I think the tool linked below technically allows for a two piece solution that screws together so we could go a little nutty if you wanted.

https://freespeakertools.com/port-optimizer/

It's an interesting tool to play with. It would depend on your SPL spec, but when I entered in your values at 100dB the result is a bit smaller than what Unibox shows.

[JonMarsh]

Darryl might be interested in that! I'm doing enclosure plans for both the ported configuration and a triple Passive Radiator version- was working on the latter this afternoon in Shapr3D, that's what I'll build here for verifying a crossover design and having a build for a secondary system.

I have a crossover roughed out based on measured data from test fixtures, not a finished cabinet, but it's proving things out, looks somewhat similar to the Calliope, but different values, of course, with a lower crossover point.

[JonMarsh]

Working on the 3D cabinet design for the Passive radiator version, I came to the conclusion that the SS10's are just a bit too big- just like the RSS265 were on the original Ardent D build. But, thinking back to what I recall of the parameters and balance of the RSS210 PR, I did an analysis for that, and it seems to be the sensible way to go... to whit:













Not quite Twin Sons of Different mothers, but pretty close tuned this way...

[JonMarsh]

Mainly for Darryl's (and a few other curious parties) benefit I'm going to show the cabinet updates and the concept for how it can be tuned or oriented to an end users preferences, as regards the vertical angle. This was only explained a bit in the Ardent D project, because it was basically a conventional format acoustical prototype for the stacked laminate enclosure we planned for Sven and that SMJ Audio (Steve Manning) built.

So, since I'm building a Passive Radiator version, that's what I'm showing drawings for, but they key adjustable part of the architecture is the supporting base, which allows the front and back bottom of the enclosure to be at different heights. Aspects of this came from a sub Brand of Tidal Audio, VIMBERG, that I admire as regards design esthetics and results.




So lets look at a few views of the updated drawing:

First, a front view... Note the base structure, a high density phenolic base plate, made from Leecraft BK3, which will attach via bolts at a number of locations in the bottom of the cabinet equipped with threaded inserts, M6 RAMPA being my preference. Of course, DIYW. The elevating blocks can be varied in height, varying the vertical angle of the cabinet. The material I used was scrap pieces of maple ply, due to it's strength and ease of working (basically no rip outs on routing or drilling operations)



Here's a photo of what one build example of the base looked like:






The outriggers are a basic product made in China, available on Amazon, and easily converted to different interface products with the floor.

Rear view makes the concept a bit clearer, with rear panel openings for the RSS210 passive radiators- not shown is an input plate connector. This could be on the underside of the phenolic plate:



And one more view, from the right side, with the view of the nominal cabinet axis included:

[JonMarsh]

A few notes for Darryl and others that might be interested...

• There's nothing "mandatory" about the wood choices I am using; they're based on experience with other cabinets, including the ARDENT D, and guided by consistency of strength, workability (once you route with Maple you may fine many other materials annoying at best), and supporting a very simple finishing process that yields attractive results- basically, just Watco Golden Oak oil and wipe on poly. No veneering, no spray painting of solid colors or working the steps to do a piano black finish. I've done plenty of these, and this path makes life easy.
• Oh, and good luck finding different easy to construct approaches that wind up with an enclosure anywhere near as stiff and rigid. But if you do work out something you're pleased with, by all means post about it in a build thread!
• I'll provide PDFs of piece part fabrication drawings. Normal drawing size I use is "C"; it prints well on my Epson T3170 Printer-Plotter, and scales acceptably to A size prints.
• I use cut off scraps of the maple ply for some of the structural pieces, like the front to back braces at the bottom of the enclosure, and for the external base support struts that determine the vertical angle of the cabinet. Of course, if you like the 10 degree tilt back, then you can probably just eliminate those parts and incorporate footers into the bottom of the baseplate. This would make it more like a Wavecor Ardent, and if you use an elevated seating position like Sven prefers at home, this might be the optimum choice.
• Of course, as always, it's DIYW, Do It Your Way, whatever works for your circumstances and desired results. I'm just the wires and sparks guy, not a bonafide wood worker, like Steve Manning, for example. Though I probably have more routers than him, he doesn't need any stinking routers with his CNC!
• Speaking of Maple Ply, UPS reports they are delivering my order this afternoon. Woo Hoo!
• And RSS210-PR's arrived yesterday!

And last, a photo showing the top of one of the original Ardent D's, showing what things look like after it's all put together and finished:

[JonMarsh]

Maple Ply arrived on schedule this afternoon. It’s nice when stuff goes according to plan

[JonMarsh]

I'm going to post one more PDF screen capture right now- in this one, I took off the edge fillets, that were implemented in building the original Ardent D, as it allows the dimensioning for the raw pieces to work more easily.

Note, the original 3D design file for the Ardent D was prepared in Shark CAD; since around 2022, I've only been using Shapr3D, developed in Hungary, first for the iPad Pro, then Mac, then PC.


I'll create piece part PDF drawings using this version of the cabinet design. All the major ones are already complete; just some small parts to finish up.



Note, the only screw/bolt holes I'm including in these drawings I'm including in these PDF drawings are for the 3/4" BK-3 baseplate, which can be used as a template for locating M6 threaded inserts in the assembled cabinet bottom.
This version includes the recommended mounting for 3x RSS210-PR; for the ported version, do not put these holes in the back panel, but install a Precision Port 3" flared speaker cabinet port tube in the base of the cabinet, or as otherwise desired.

[Efalegalo]

Jon - this is looking good.

Curious - can you comment on the virticle distance between the mid and tweeter? Would that not make the verticle response very narrow?

EDIT - maybe I've answered my own questions. Looking at the plans, the spacing between the tweeter and woofer approximate 350mm, which coincides with 1.2x what I can only assume would be a crossover freqency of approximatley 1,200Hz. I'm assuming such a low crossover point simply based on selecting the DA32 (and not DA25)

[JonMarsh]

OK, some early "sharing" here, so that folks keep in mind the work required for this build, should you choose to accept this "Mission Impossible".

(well, it's far from impossible, but it's not a simple 6-1/2" two way in a box with 90 degree corners...)

I've added some "hints" and directions into this print, should answer some questions, and establish a realistic baseline. Of course, I'm assuming overall a reasonable accessibility to tools, including a variety of routers, a high grade miter saw, a capable high precision table saw with a reasonable deck size, and a good quality biscuit joiner, like a Makita or DeWalt.

No CNC was harmed or used in this plan or past contructions.

THESE are INITIAL DRAWINGS FOR A PROTOTYPE CONSTRUCTION FOR A DESIGN THAT HAS NOT YET BEEN COMPLETED AND TESTED (Meausrements in cabinet and crossover work pending).

So, consider this for information purposes only, though obviously, it's what I'm building, based on past experience and fairly standard design work, as shown earlier in this thread. This JPEG was created from the PDF. Message me if you'd like a copy of the PDF.

[JonMarsh]

Didn't mention this, but the routing operations should be done on the 1-1/2" thick panel pieces, how ever you choose to fabricate those. This is why we want precision operations so that stuff lines up properly when assembled. I can post about those steps if any one is interested.

[JonMarsh]

All the other piece part drawings are completed and printed out as of late this afternoon.

[T-35]

Jon,
I’m so excited about this project! It’s been my dream to build the M8ta’s and I asked you one question about them and you didn’t hesitate to help me out and steer me in the right direction.
I can’t thank you enough, Jon!

[JonMarsh]

I'm glad you're enthused; of course there are ways to do this and not spend as much as my build will cost- for example, the MW19P-08 are very close in performance. My focus here is a high level of refinement and musical transparency coupled with a nice form factor that hopefully will rival some of the ludicrously priced stuff on the market these days, by judicious design choices, including drivers, cabinet materials and attention to detail in execution.

Heck, in my 35 years with my last employer I did reach the level of Distinguished Engineer, and I do my best to bring the same thought processes and attention to these projects, as music and audio is a long time first love since the 60's, when I got my start as a gigging musician.

[JonMarsh]

Jon - this is looking good.

Curious - can you comment on the virticle distance between the mid and tweeter? Would that not make the verticle response very narrow?

EDIT - maybe I've answered my own questions. Looking at the plans, the spacing between the tweeter and woofer approximate 350mm, which coincides with 1.2x what I can only assume would be a crossover freqency of approximatley 1,200Hz. I'm assuming such a low crossover point simply based on selecting the DA32 (and not DA25)

Efalegalo, you know I just love it when folks pretty much answer their own questions!

Of course, your concerns are valid, but I would note that the various earlier versions built many years ago always used tweeters capable of much lower operating frequencies while still having reasonable distortion... like a specific Hales tweeter, the Dayton RS28a, and the Millennium Excel.

Regarding the M8ta concept, and this new version of it, a few points from my perspective-

• My preference for tweeter height is basically at seated ear level.
• A larger speaker like this is not a quasi near field monitor, in the way that 6-1/2" two ways often are- the size and esthetics usually get selected because they are planned for a bit larger room- so to some extent, it's reasonable to cut a little slack about the vertical separation if it has other benefits.
• And the DA32TX is focused on lower range extension, power handling, and distortion. So there's some flexibility compared to my original Satori two way with the MW19P and DA25TX, and the SMJ version, the Calliope. I have an LR3 crossover concept roughed out from data taken with separate test structures.
• I'm considering evaluating two kinds of crossovers- a somewhat more classic 2nd order, using phase and level matching similar to a Duelund, but I suspect I will end up favoring my home grown LR3. (Linkwitz and Riley did not describe a 3rd order with -6 dB transition, but as I've described elsewhere, it has some neat characteristics if the driver Z-axis behavior can sync up with the desired crossover point. I met Linkwitz at one of the California get togethers.

So, I'm in the midst of re-organizing working space and stashing away the completed new versions Saint-Saëns stuff to expedite a working space for this.

Hopefully nothing will happen like last summer with a succession of medical issues related to my wife and her older daughter gobbled up all my time from late June through October!