yep, its mathematics usually
Tulip and Concept on my web site are both derived from mathematical curves, area of section is done on a basic graphing calculator. then you can determine the depth of the box required to get the required volume.
The other approach is to simplify the shape down to rectangles and triangles, so long as the final box is basically just a prism the calcs are generally easy enough,
Some of the 3D shapes like "Stacks" can be done using larger formulas for truncated cones, ellipsoids etc.
Tulip and Concept on my web site are both derived from mathematical curves, area of section is done on a basic graphing calculator. then you can determine the depth of the box required to get the required volume.
The other approach is to simplify the shape down to rectangles and triangles, so long as the final box is basically just a prism the calcs are generally easy enough,
Some of the 3D shapes like "Stacks" can be done using larger formulas for truncated cones, ellipsoids etc.
Stuart A said:
You can use a drawing package like Raka does or come pretty close with simplifying the shape and estimating like Andy suggests.
I took Andy's approach with eliptical subwoofer enclosures. I took the area of a circle with the diameter of the large length of the elipse, another circle with the diameter of the short length and took the average. Close enough.
With egg shapes I drew an oval over the egg large enough that there was as much area of egg overlapping oval as there was oval overlapping egg (clear as mud, right?). Afterwards I put it into a CAD program and let that calculate the area precisely. My overlapping estimate was quite close.
A third way is to post a drawing on this forum. I'm sure someone will carve it up in the right way and come back to you with an easy formula.
Steve
Hum, you could always put a half of the internal structure cutout on a graph, and depending on its geometry make a formula (
Then use the formula to subdivide the geometry in rectangles wich are Y times Delta X. The closer you get to a infinity of subdivisions wich is not hard once you have a formula (when delta X tends to 0) the closer you will get to the real surface of the cutout.
That gives you the half surface of a 2d cutout, then multiply with 2 times the height.
You could even make a excel spreadsheet to do all that for you, and even integrate a third dimension into the whole formula for more complicated designs..
I guess thats the hard way.....
Anyway, its the first thing I came up with.
to Ilianh
That is precisely the way "Tulip" was calculated.
I can't currently find the origainl fromula for the curve, but here is the computer drawn curve using the formula.
Obviously, once you have the formula, you can do what you like with it !!! now getting the formula, that is a different matter
http://users.tpg.com.au/users/gradds/designs/tulip.JPG
That is precisely the way "Tulip" was calculated.
I can't currently find the origainl fromula for the curve, but here is the computer drawn curve using the formula.
Obviously, once you have the formula, you can do what you like with it !!! now getting the formula, that is a different matter
http://users.tpg.com.au/users/gradds/designs/tulip.JPG
Translam certainly is an exciting construction technique.
However, the one drawback is the potential waste of material.
For example, Keith Kidder's project is certainly impressive but there's a huge amount of waste inside each rib and even on the outside of the curves.
What you could do here to radically cut down on waste and cost is to break each rib at a line thru the middle of the front and rear baffles. Then arrange them in your CAD software like spoons in a drawer and have the CNC cutter work with that. Yes it's a little more work during assembly (although with guide holes and rods alignment won't present a problem) but it would sure eliminate lots of wasted mdf/ply and money.
Another thing: if you use guide holes and alignment rods as you're building up the layers it might pay to use some threaded rod. That way when you reach your last layer you can put a flange/washer on the rods and torque them down with bolts like an integral clamp while the glue dries. Of course, there's no need to remove the rod, just let it become part of the cabinet.
Mos
However, the one drawback is the potential waste of material.
For example, Keith Kidder's project is certainly impressive but there's a huge amount of waste inside each rib and even on the outside of the curves.
What you could do here to radically cut down on waste and cost is to break each rib at a line thru the middle of the front and rear baffles. Then arrange them in your CAD software like spoons in a drawer and have the CNC cutter work with that. Yes it's a little more work during assembly (although with guide holes and rods alignment won't present a problem) but it would sure eliminate lots of wasted mdf/ply and money.
Another thing: if you use guide holes and alignment rods as you're building up the layers it might pay to use some threaded rod. That way when you reach your last layer you can put a flange/washer on the rods and torque them down with bolts like an integral clamp while the glue dries. Of course, there's no need to remove the rod, just let it become part of the cabinet.
Mos
travis said:
Corian (not sure about the spelling but it's heavy acrylic stuff that they use for expensive kitchens) has good potential for speaker cabinets. Acoustically it's very inert and it can be cut and shaped quite easily.
Corian comes in sheets of 1/2" thickness so, if the speaker was larger than 'little' you might want to back it with something. You could probably go to someone in the kitchen business that uses the stuff and get off-cuts very inexpensively.
For bass units it could be excellent. If the cabinet has to handle the mids as well, I would be tempted to 'soften' the inside with silicon or similar (let it dry thoroughly before adding the drive units).
Just a suggestion.
Steve
SAND CHAMBER...
This layered lamination construction technique would seem ideally suited to creating an external sand chamber, connected to the inner box by just a few tabs on a few laminates. Vibrating sand translates cabnet energy into heat. Sand can be added when the speaker is in the optimal listening position, and vacuumed out through a bottom hole when you need to move.
TAD's $45K Model 1 is constructed from over 50 slices of Baltic Birch plywood. There are several excellent pictures on the web of the laminates. Permanent threaded steel rod is used for alignment and strength.
This layered lamination construction technique would seem ideally suited to creating an external sand chamber, connected to the inner box by just a few tabs on a few laminates. Vibrating sand translates cabnet energy into heat. Sand can be added when the speaker is in the optimal listening position, and vacuumed out through a bottom hole when you need to move.
TAD's $45K Model 1 is constructed from over 50 slices of Baltic Birch plywood. There are several excellent pictures on the web of the laminates. Permanent threaded steel rod is used for alignment and strength.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.