Constructing a sphere...

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi all,

I am designing a set of small BR bookshelf speakers using a full range driver, and I'm looking into how to go about constructing spherical cabinets for them. My initial thoughts is printing out a shell using a part printer I have at school (can print parts up to 30cm*30cm*30cm IIRC), but the printer material is not suited for speaker enclosure.. So I can either dress a thin shell with layers of fiberglass, or I can use it to cast a silicon mold. The latter would perhaps provide the better result, but I have never tried something like this.. Any tips of techniques or molding material from anyone who has good ideas or has gone down this route in the past?

Any feedback will be appreciated..
 

Attachments

  • untitled-2.jpg
    untitled-2.jpg
    90.7 KB · Views: 754
sreten said:
Hi,

how about fibreglass over a cheap plastic football ?

Or using the inside of the football as a mould ?

:)/sreten.

I think styrofoam spheres will be the thing. I plan to intersect it with two planes (cut it) one on the bottom side, and one at the speaker mount. After that, I'll cover them in epoxy and fiberglass or something similar... continue in layers for about a cm, then sand and smooth it, using a planar sander on the bottom side, and finally drill holes for port and driver, and pour acetone in it...

Now, if I could only find styrofoam spheres of the right size... 23-25cm diameter..

EDIT: I mocked up the translation here I think... I wasn't referring to styrofoam, but something we call "isopor" here in norway.. Clue translates it to "expanded polyester". The stuff the chips you put in boxes when shipping triodes are made of..
 
xstephanx said:
one more vote for one of those cheesy soccerballs you used to get when you were a kid to play in the pool with. usually smooth plastic with a soccer ball pattern printed on it. after you coudl just deflate it to take it out.


Problem will be the planar surfaces... How would you guys do that with a soccer ball..?
 

Attachments

  • untitled-1.jpg
    untitled-1.jpg
    29.5 KB · Views: 545
It would be doable with the ball... it could be constructed as a sphere, then sanded down the planar surfaces and reinforced with more fiberglass from the inside after the driver hole is opened.. I might give it a try if I can't find what I need in any of the local hobby stores.
 
Now as I understand this project, it does not have a tweeter, but if you are thinking of adding a normal tweeter with a ~90 mm flat and round faceplate in a sphere, there will be a corner on the sphere near the edge of the faceplate. This corner appears at the same distance from the dome throughout all 360 degree directions. This in turn leads to a baffle step that is not very nice.

This is less of a problem for the woofer, where almost all of the non-spherical part (the driver cone) is radiating sound. In the tweeter case, the dome covers just a little part of the faceplate.

Sorry if that was not part of your question.
 
Ah, the turkey gets cold over here faster than a rabbit gets ****** :hot: I think the spheres introduces a whole new set of problems.. Not so much the construction itself, which I believe will be feasible enough, but doing the inside right, as Dave pointed out in the 2nd posting.. Port placement, etc.

Also, since I'm designing for someone else (a birthday gift to my sister), I don't have the opportunity to tweak and listen over a long period, especially not in the apartment where she will be using them, so I believe it is safer to take the rectangular frontplate with all its flaws, and correct them as best can be done. As yet another point, the spheres started to look kind of stupid as I increased volume to my needs, with the tiny FR125S driver looking way to small for the fat spheres.. And I wouldn't sacrifice performance to make'em look good either. In other words; my spheres is canceled.. for now at least.

I'll move this discussion back to the fullrange forum where it came from , thanks for the tips though, I'll definitly build some spheres on another occasion.

Svante; The intention was (still is but in a regular box) a fullrange driver (CSS FR125S) with no tweeter. Interresting remark you made though. I'll make sure I never try that stunt with a tweeter :cannotbe:


Anyways, I believe something like this will work far better than what I can do with the inside of a sphere.. :
 

Attachments

  • untitled-2.jpg
    untitled-2.jpg
    53.9 KB · Views: 423
planet10 said:
There is an extensive thread here about a spherical woofer.

Keep in mind that the outside of a sphere is great for minimal diffraction, but quite ugly inside for resonance.

dave


This is my first post to a forum, any forum ever so please forgive my ignorance. I have two questions for Dave.
1. Where is the “extensive thread” you refer to? I don’t understand how to find or follow it.
2. Please explain why a sphere is “ugly inside for resonance.” I thought they were hands down the best choice for a speaker enclosure inside because they defeated reflected waves.

Robert
 
SY said:


The symmetry of a sphere makes it easy for standing waves to form. They'll be distributed differently than in a rectangular box, but they'll be strong and distinct. If you've got some math background, look up "spherical harmonics."

I'm no expert but doesn't that only apply if the source is at the focal point?
Here are a few measurements of a 4.5" woofer in a 2pi anechoic chamber relative to a free air rear load:

1) mounted to the surface of the sphere (not in the focal point)
An externally hosted image should be here but it was not working when we last tested it.


2) mounted at the focal point (used a 1/2 sphere as the enclosure)
An externally hosted image should be here but it was not working when we last tested it.


3) mounted at the focal point with loose fiberglass stuffing (2" thick)
An externally hosted image should be here but it was not working when we last tested it.



Feel free to draw your own conclusions from this... :D
 
Excellent measurements! I don't mean to be greedy here but could you do the same thing in a cubic box (just surface mounted) for comparison?

Were these near field or far field? I wonder if the twice-as-big variations in the focal point mounted one were diffraction from the edges of the hemisphere and unrelated to internal modes?
 
LOL... we're on the same wavelength :) ( no pun intended )

In the same test session, I measured the 4.5" driver with an 8" square enclosure and a 4.5" ID tube..both without and with insulation. Mic distance in all measurements is 1 meter and the front of the woofer
was still mounted on a flat surface exposed to the inside of the chamber so there were no diffraction related effects. The different enclosures were brought up and held behind the driver.

An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.