The Phoenix Gold Cyclone..

And the two Pac Sci servo motors.

These motors are significantly larger than even the plastic housing that contains the Cyclone's servo motor.

Nate (Luvdabass on CarAudioForum) is a machinist, and he played a critical role in making this project even feasible.

He flatted the larger output shafts on the motors, and then fabricated some metal coupling clamps that fit and secure onto the motor's output shafts, which has a screw-lock slot on the other end that receives a 3/16" x 3/4" steel bar.

The moving vanes will be laminated across this steel bar.

Here's the motor coupled to the steel shaft...

With another powerful subwoofer motor there in the background...

I was initially going to build an enclosure, and have the motors hanging off each side of the enclosure. I realized, however, that this limits how wide I can build the enclosure, and still fit it in my trunk.

I discovered that the motors fit EXACTLY inside of 4" PVC pipe.

What I have done is to build housings for the motors, out of 4" PVC pipe, 3/4" clear plexiglas, and 1/4" blue plexiglas, to which the motors are mounted. The housings are held together with a combination of special Acrylic bonding cement and predrilled screws.

These housings will be built into the side panels of the enclosure, allowing the motors to be inboard.

The vanes will extend over these housings to the outside walls of the enclosure, allowing me to maximize both enclosure and vane width.

These pictures show the complete assembly so far, with the motors inside the housings, coupling clamps mounted to the motors, with the center shaft secured between the motors.

This is the correct final width of the assembly.

Note the small flatted output shafts extending out the back of each motor.. it is to this that my subwoofer's suspension will be attached, providing me easy access if tweaking suspension compliance is desired.

I made a posterboard template of the moving vane, so that when I cut the foam and Kevlar, I will have consistent pieces.

You can now see why the motors are enclosed in housings.. They will be mounted inboard, in the enclosure, the vanes sweeping over the outer surface of the housings.

It was important to get this sized properly, so I used masking tape to mock it up for verification.

If you are trying to visualize this, picture two of these panels, one on either side of the center vane.

There will be a solid piece of Kevlar stretching across both sides of both vane units, laminating and bonding them to the center shaft.

Note that the vane extends only to within just over 3/4" from the backs of the motor housings.
This is to accomodate for the 3/4" thick walls of the future enclosure. I propped up two scraps of 3/4" MDF just to illustrate.

The backs of the housings will be flush with the outside of the enclosure, allowing for access (not to mention, I currently envision the suspension will attach to the two smaller flatted shafts extending out the rear of the motors).

Here are the actual vinyl structural foam vanes cut, and set in place exactly how they will be positioned when laminated to the center shaft, sandwiched and bonded between two Kevlar layers.

This foam is so lightweight, if you hold it in the air and drop it, it settles to the ground as slowly as a sheet of paper.

Here is the suspension concept for the servo sub:

A cam unit mounted on the rear output shaft (which is conveniently flatted) will be located between two lengths of spring steel.

These lengths of spring steel are secured to the enclosure by two end clamps, maintaining pressure on the cam unit.

As the motor turns back and forth, the cam unit moves from it's "at rest" vertical position, pushing the spring steel rods apart, which provide a resistive force, naturally pushing the cam back to a vertical position...
Thereby returning the subwoofer to a centered "at rest" position.

Note that in this way, suspension compliance can be varied by either varying the thickness of the spring steel, or the distance between the two end clamps (moving them in towards the center would stiffen the suspension).
The Kms (or Cms, depending on how you think of it) curve can be varied by the actual shape of the cam unit.

This is what you need to perform vacuum bagging:

Top: breather/bleeder - cottony absorbant material, to wick up the excess resin.

Middle: Peel-ply- Resin bleeds through the performations into the breather/bleeder. Next is the bagging material, bag sealant tape, package of Kevlar cloth, and the epoxy resin and hardner.

Bottom: A vacuum gauge and T-fitting, a vacuum venturi, and vacuum valve. The venturi uses air from a shop air compressor to create vacuum, while the vacuum valve is what protrudes through the bag.

Here is a dry layout of the vane assembly, just to roughly cut out pieces of Kevlar that were large enough.

One experience cutting Kevlar is enough to make you understand why bulletproof vests are made of them - regular scissors simply do not cut it (literally).
I observed that companies make specialty Kevlar cutting shears for $60-$200, but avoided having to make that purchase after locating some tips on how to modify a $15 pair of Fiskar's 9.5" curved blade heavy duty shears to perform like Kevlar cutting machines.

I initially weighed the steel center shaft, giving me a baseline weight.

In the scope of the rotary subwoofer, this weight is almost (yet not completely) negligable, being the center of the rotating axis.

Primarily, I measured this so I could later calculate what the actual lamination for this 36" wide vane ended up weighing.
Initial weight: 1 lb exactly for the center shaft.

Two full days later, I opened the bag, and peeled the breather/bleeder and peel-ply from the fully cured vane assembly.

The cuts that were made to the foam prior to laminating were not done for visual impact, but rather for strength. The way the Kevlar contours and mates through these holes adds rigidity and strength, similar to the way that an I-beam of 1/2" steel is stronger than a flat panel of 1/2" steel.

At this point the vane unit is laminated, but still needs the excess Kevlar trimmed from it.

The total weight after laminating (including the steel center shaft weighed initially): less than 2 pounds.
Not bad for a 36" wide, 12" long, 1/4" thick rigid panel....

Particularly considering that the swinging weight will end up being only around 3/4 lb after the excess is trimmed from the vane unit (again, discounting the steel center shaft located at the rotational axis).

Here is a picture of the fully trimmed vane unit, fully assembled!

I may laminate one more strip of Kevlar across the center shaft, just overlapping each half of the vane slightly, just to add a little more strength to this area, I expect it to be a pretty high stress area.

At this point, the raw "driver" is complete, sans suspension (which will be dealt with last).

Now I just needed to make the curved sweep-surfaces that the outer edges of the moving vane would brush up against.

I decided that the best way to make a "perfect every time" exact arc shaped surface was to make a jig for my bandsaw, and cut out numerous exact copies of the same perfect arc shape.

The jig was simply a metal pin in a wooden panel to establish the fulcrum. I drilled holes at 1" intervals and simply kept moving the pin to the next hole.

...And I ended up cutting well over 100 of these little arcs (better safe than sorry), although I "only" needed about 96 of them to make two 36" long panels out of.

Just for size reference, just wanted to show my pile of MDF arcs is larger than my 9515 and Brahma together... let's hope that's good foreshadowing!

OK, yes, the pile is just closer to the camera.. details, details...

I first clamped and glued them together in groups of six, and let them cure.
Then, I glued the cured sections together to end up with two 36 inch long curved panels.

While not appearing as consistent from this backside view, the important part is the other side...
This technique allowed me to fabricate two long panels with a perfectly mating curve on the inside of the panel, which is important for proper operation.

The next step was to glue the formica surface to them, since I want a perfectly smooth surface for the vane to sweep against, with as little airleakage as possible.

Like the Phoenix Gold Cyclone, there isn't any need for anything other than it's construction tolerances to prevent leakage... so let's hope I've got all my measurements on!

This is a very significant milestone, and possibly the first good picture that I have to really appreciably communicate the scale of this monster.

Each curved sweep surface panel is just a hair over 36", making this entire assembly just under 38" wide (or tall, as you are looking at it here).

The holes in each of the side panels will contain the motor housings (as pictured in the 4/20/2003 set) and the vane will fit exactly between the curved panels.

If you are having trouble picturing how the end product will come together, these pictures should help!

The motor housings were secured to each end panel using wood screws about every inch around the inside of the 4" diameter PVC housing.

I then did a mock-up, with all the parts in place... motors, vane unit, etc.

One very important element of this project that doesn't yet exist is the dividing wall that will separate the two "phases" of energy as this subwoofer operates. This necessary dividing wall is going to extend along the vertical pencil line on this picture, right up to the center shaft, bisecting the enclosure.
I may need to get a little clever, getting the wall to couple against that center shaft without impeding it's motion...

Here's a more distant shot of the mocked up driver unit.

The suspension's cam slides on the end of the motor shaft protruding out the rear of each motor, and the Delrin leaf spring it will press against will be secured to the large side panels.

One main purpose for mocking this all together is to determine my clearances. I found, after I mocked it all up, that I needed to take a belt sander to the edges of the vane in a couple areas, to bring them down slightly and allow for motion without rubbing.

I added two MDF braces down both curved panels.

This serves dual purposes:
1) This adds rigidity to the curved panel
2) This is necessary to separate the top half's front and rear chambers, and the bottom half's front and rear chambers

Remember, this is initially planned as a bandpass box, and each half of the vane needs to be treated as it's own subwoofer... so that means two bandpass boxes, hence the four chambers in my illustration way above up there.

Here's a picture of the first cut on the suspension.
Nothing fancy, figure I'd try one "spring" per side to begin with.
You can see the Delrin cam that I built, now attached to the flatted rear output shaft on the back of the motor, which will operate against the flat spring pressing against it.

And If the Delrin doesn't make the ideal spring, I have a fiberglass 1/4" rod that I can substitute, it may have better damping and spring properties, who knows!