[kodyaudette]

Yes, it's a radial clearance issue. Hmm, I could see the glass microspheres and loctite working reasonably well. I would have to support the rest of the rotor in the right orientation, and maybe slowly turn the motor while it sets to ensure that it's centered.

The wire may be easier to source (not sure where to look for microspheres, but I have Amazon Prime, so that's a good place to start). I'm estimating I would need quite thin wire, as the first fix I tried was wrapping with some electrical tape. I couldn't get more than one layer on before it would start to bunch up.

Off to do some figuring. Thanks for the ideas, I hadn't considered either!

 

[noah katz]

The idea is that if the microspheres are close to the right dia, that will do the centering.

TAP Plastics has some.

 

[kodyaudette]

Okay, so I had the first successful test of the unit today. The rotor shaft wasn't 100% aligned, but pretty close, enough that the VC wouldn't rub under operation, so I went ahead and epoxies the swash plate to the VC. I strapped everything down and applied wood blocks so it couldnt move. Because I have no basement or attic, I have to use the doorway to my garage as the baffle for this device. So I cut some 3/4" plywood to the size of my doorway, applied some adhesive foam lining to seal it, then just pressure fit the plywood in there for now. I cut a 13.5" circular opening for the fan, enough to leave a 1/4" gap on all sides.I placed the device up on concrete blocks to be the right height for the opening. I ran speaker wire from my amplifier to my garage, connected it to the VC, and turned on the motor. Went back inside and pulled up a tone generator. From previous tests, I know that my system won't output anything lower than about 2 hz, I get zero blade movement, no matter the amplitude. Anything above that, I get a response. I haven't tuned this channel for the rotary subwoofer yet. Anyway, I started by sending it some low frequency signals. Everything seemed to be working, but I didn't want to push the volume too much yet, especially with a 6 hz sine wave. Slowly raised the frequency up to about 25hz I could definitely hear a nice clean signal (on top of the noisy fan itself of course) after a while, I heard the fan noise building and then suddenly the blades stopped spinning. It looks like the screw between the rotor shaft and motor shaft broke. So, initial test was successful, but I need to retool how those connect to each other.

 

[noah katz]

> I need to retool how those connect to each other.

Split taper H bushing is robust, which is similar to a collet.

 

[LastButNotLeast]

I strapped everything down and applied wood blocks so it couldnt move.

MacGyver ain't got nothin' on you!

Looks like fun, but I'd still stand back if I were you.

If I were me, WAY back.
Michael

 

[BllDo]

Bump, any updates?

@zettelsm @kodyaudette

 

[kodyaudette]

@BllDo Sorry for the lack of updates.


So, my prototype functioned well, created low frequency content as evidenced by this video:

But:


A) there was a lot of operation noise as the cone slid back and forth, from the electric motor, general minor misalignments and whatnot. There is definitely a reason that these are used with a large baffle. I have the materials to build a baffle when I make the next iteration. My particular application is complicated by the fact that I basically need to be able to roll it in and out of place, as it fits into a doorway.


and


B) It rattled itself to bits! Haha. The main shaft for the helicopter rotor that I bought was a 5mm shaft and it was tied to the motor shaft via a 2mm diameter screw right through the shaft, the forces involved were simply too much for the 3mm of material that was left at that point and it eventually broke. Also, the control rods had plastic rings that connected to the balls of the rotors and these had the tendency to pop off under dynamic loads. It seemed to operate very well from DC up to about 20hz. After that things started to sound kinda shifty. Definitely better to mate this design with some already very capable subs.


So, my current status and course of action:


I'm in the process of finishing up building an acoustically transparent projection screen and screen wall, behind which all of my front speakers will live. As soon as I sell a pair of Definitive Technologies speakers I have on CL, I will be building a pair of 14cf Dayton Ultimax 15" ported subs that model with the capability of doing 122db at 18hz and a decent rolloff below that. Once I get these built and setup and tune my two other 10" subs for nearfield tactile impact, I think the stage will be set to rebuild this design in a more robust manner and allow it to operate within its wheelhouse.

 

[BllDo]

Thanks for the update. That's very interesting. I'm curious how your blades worked out? I really like the design of those.

I just noticed that @chris hudlin posted another video series at the end of last year for another build, lots of good information. Thanks for those that have shared your detail in this thread.

Edit: Are you still using the single phase motor or did you end up needing a three phase?

 

[kodyaudette]

So far as I can tell, the blades are working great. No flexing that I can see. They are somewhat heavier than the blades Chris was using, so I would assume they would have a tendency to keep their momentum under heavier loads , but that may have also contributed to the shear forces on the rotor shaft. I didn't notice the blades acting strangely or acting up, but if they did, I would assume that it would be at higher input frequencies where these systems exhibit their limitations anyway (20+ hz, etc).


So I think the blades are fine. Also as far as initial tests go, the single phase motor was plenty powerful and didn't noticeably slow down under load. I do think that it will be important to figure out a way to gradually ramp up to operating RPMs, so as to not overly load the rotor shaft. Though I didn't have the opportunity to actually measure the amplitude of the ULF output I was getting before everything went sideways. So I couldn't tell you if I was getting 85db at 8hz or 110db. All I can tell you is that I felt a noticeable air texture change and my projector started hopping around. Haha. At higher frequencies you could definitely hear the output, but the operating noise got louder as well. Once I get the next generation built one of my first priorities will be to get some good, quantifiable measurements.


I haven't checked in on Chris in a while, I'll have to see what he has posted.

 

[mtnboydl]

Thanks for all of your updates! I'm not savvy enough to tackle for now, but maybe when the time comes I will just pay one of you guys to do it!

 

[popalock]

hey guys, the guide is complete!
the link: http://www.filedropper.com/diysubwooferguidebychrishudlin
the file is 65mb, rather large for a word document, but that cause of all the pics. its 20 pages long but wont take long to read. enjoy

Link to the guide doesn't seem to be working anymore?

 

[BllDo]

Link to the guide doesn't seem to be working anymore?

The file in post 122 is still attached. Not sure if that's the same one.

 

[asarose247]

just checked it out

original video still available on you tube

 

[popalock]

The file in post 122 is still attached. Not sure if that's the same one.

Not sure? If it is, it appears as though a lot of the important pictures are missing...

 

[asarose247]

not sure why i saved this but maybe . . .

 

[CherylJosie]

This fine construct might as well be made from unobtanium IF obtaining a good usable to its design capacity ULF signal is elusive

Only if the subwoofer is going to be placed on the ceiling.

It's a fan. How well does a fan work when you mount it in a sealed box?

It is mechanically impossible to seal a fan. I scratched my head over this one for a few minutes... whaaaa?

You won't even come close to balanced blades if you try fabricating your own. It will quickly tear itself apart.

the owner very kindly ran my five blades through his sheet metal roller.

I am curious what the balance is like after they have been rolled. Did they deform (flow/elongate) significantly from being rolled? Are they all similarly straight and presenting an equal aerodynamic load to the motor shaft?

 

[CherylJosie]

...Bruce was very up-front about the necessity of dealing with fan noise by means of a sound-damping
chamber between the unit and the listening room. With the 3' X 4' X 5' chamber

At one time Bruce Thigpen was floating the idea of a smaller rotary subwoofer for car stereo use.

So the 3'x4'x5' chamber is the trunk of my SUV... oh wait...

What happens to the ULF s/n when you open the windows? How would anyone hear barely audible ULF over the road noise and vibration even in the best case?

Ill-conceived project IMO. Silk purse, meet sow's ear. I already have a car that has to be shut off to hear the stereo properly. A better stereo is not going to fix that.

Anyone have thoughts on ways to increase high frequency extension?

You got many good suggestions, but I have one more.

Pay attention to that damping chamber. It represents an acoustic impedance and its primary function is to reduce the high frequency bandwidth so that the fan noise etc. is attenuated. It probably has both an absorptive lining for the turbulence noise and also some tuned Helmholtz to tame the blade resonance (a blade attached at one end is basically a Wurlitzer electric piano reed).

If I am correct, the Helmholtz may be a limiting factor in the usable bandwidth, but good luck figuring out how to model that. You might want to go back to the original source but disclosure would probably compromise proprietary information so you may have to re-invent this thing from basic principles.

Probably the blade resonance and also tuning frequency of the damper is high enough that it is not a problem, but unless you at least attempt to derive it you might end up with something unexpected. I would assume that if the blades are not stiff enough and/or too heavy, their resonant frequency could be a problem that even a damper is not going to fix. Plastic is probably better than metal because it is somewhat self-damping. A metal fan blade is like a Wurlitzer piano reed.

Given that none of these projects are using aerodynamically tailored fans let alone tuned damping chambers or even precision machined moving parts worthy of audio s/n ratios, I suspect that fan noise, motor noise, and rattles are going to be more of an issue than bandwidth. I would suggest that any installation should put the motor on an isolated slab of stone that floats on rubber dampers and have an impressive damping chamber on its output side to compensate for the DIY tuning issues of construction.

I agree with everything you've written -- a strong VC motor will allow pitching the blades with a higher load. Increasing motor rpm will produce more SPL than lower rpm, all other variables held constant. But of course, higher rpm also increases the load on the blades the VC has to handle. This load can be mitigated by ensuring close to equal areas in front and behind the pivot axis of the blades, with a little more trailing area for self-centering effect.

Almost. There are potential stability issues. Aerodynamics can be tricky. 'a little more trailing area' presumes that this problem can be fixed with a slight offset. Depending on the blade characteristics and the tuning of that damper, that might not be the case at all. Stability is a function of phase and magnitude in the best (linear) case. There is spring tension and mechanical slop involved with the lack of precision machining that introduces nonlinearity and could introduce resonant or even chaotic behavior, especially under the influence of turbulence.

You are just going to have to experiment. Without an impressive lab and some physicists you are all basically shooting in the dark. What fun!

 

[Stereodude]

It is mechanically impossible to seal a fan. I scratched my head over this one for a few minutes... whaaaa?

I'm glad you took the time to respond to a post from 26 months ago with a fundamental misunderstanding of what was being discussed.

 

[CherylJosie]

I'm glad you took the time to respond to a post from 26 months ago with a fundamental misunderstanding of what was being discussed.

I think I can hear a dog-whistle from the fan in this 'sealed' rotary subwoofer.

 

[Steve Hanna]

I just stumbled on Chris Hudlin's YouTube videos concerning his DIY rotary subwoofer builds. He's building a few of these for customers now and he seems to be tweaking the design as he builds new units. If you build this, some parts might go out of production and you would need to find suitable alternatives. Not a deal breaker.