Awl --
Suppose I have a box, a 6" cube, made from 6 pcs of 1/4" alum, butt-jointed together. No big problem, altho depending on the exact connections, all 6 pcs may not
be identical.
OK, now suppose the top face is not 6x6, but, say, 1x1 (or whatever), so that sed box is now pyramidal.
I would imagine SW would figger all the dimensions and bevels for you (does it?), but is there freeware or a utility that will calc these things out? This is sumpn you could actually just put into a spreadsheet, if you had the right formulas and relationships. Easy, in fact, iffin yer up on the math, far from easy if yer not.
Idears, links? Sheetmetal workers have programs/utilities like this, for making ductwork to match different sized openings, etc, altho thickness of the material itself is less of an issue with them.
--

Existential and Mathematical Angst, cadcam PV'd

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

I'm thinking along the lines of a BeamBoy-type ditty, for angles in solid geometry. Graphics would be nice, but not essential.
--
EA

> --
>
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Existential Angst wrote:

Sheet metal software can be fairly sophisticated. The material thickness has to be taken into account to know the "developed radius" at a bend, which will determine the real, final dimensions of the work after bending. And, of course, most sheet metal stuff is 3D shapes made out of all sorts of planes, tubes, and cones.
The only way I know to make your pyramid with a CAD system would be to draw the 6 pieces, with appropriate thicknesses, then assemble them, then trim them against each other to get the bevels and final dimensions. On lesser systems that can't trim in 3D, you'd actually have to create the bevels yourself. Doesn't sound like a job for cheap and dirty software, though I suspect TurboCad would come closest, once you learned it's ins and outs. At a couple hundred bucks or less, it's hard to beat.
KG
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Existential Angst wrote:

K:
I tried extruding it in Gibbs, and it worked - after a fashion. But you'd have to do more work than I'm willing to do to get the bevels connected and then interrogate the solid to get the angles.
Here's a wireframe. It may not display well on some computers.
http://i233.photobucket.com/albums/ee126/BottleBob_photo/Pyramid1.jpg
--
BottleBob
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

Yeah, the wireframe is certainly the right idea!
Probably the best way to do this is on an edge-by-edge, intersection-by-intersection basis, rather than as a whole solid.
iow, we know what the angle of incline of any face is -- simple trig. So now, just take two faces, tilt them in at this angle, and determine the line of intersection of the two planes. Really just solving simultaneous equations for two planes -- easier said than done of course, but ultimately can be stuck into one or two cells of a spreadsheet!
Then do this for a side and the base, top, etc.
Where's effingBrewer when you need him??? Where is he, anyway??
There's an odd bird out there, an engineer/artist, who developed a very neat ditty for fishmouthing tubing. Very ingenious. I emailed him kudos, have the link somewhere, but not at hand. Hopefully someone like this did a general solid geometry ditty for thick intersecting planes.
--
EA, PV'd

>
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
wrote:

Are you asking for a what is known in woodworking as a compound miter cut?
If so, this may help:
or pick calculator that matches your needs:
http://www.martindalecenter.com/Calculators.html
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Existential Angst wrote:

Some time ago I created a diagram showing how to calculate the angles in pyramids etc. Take a look at this and see if it's of any use.
http://www.keepandshare.com/doc/view.php?id 97953&da=y
--
Regards, Gary Wooding
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
wrote:

No software will do this without more information; the thickness of the material determines the slant height after beveling, for instance.
If you have multiple plates joining, the orientation of each plate can be made into a 'normal vector', and by dot-product of vectors you can determine the angle of the plates' intersection. The bevel is half that angle, of course. Use Cartesian geometry, it all works out. Alas, one DOES need familiarity with vectors in three dimensions, and a bit of software to do the normalizations, dot products, cross products is ... useful. Do Fortran much?
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
whit3rd wrote:

Yawn.
Easier to draw it to scale then measure it. We're polly not building a 787 here...
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

Yeah, but he's basically right, except no fortran nec, just a spreadsheet. 787 or not, it's just as easy to get it absolutely exactly right in a spreadsheet as it is to fudge it.
But, 'til a spreadsheet or utility materializes, fudge it is, and for now, 1/4" = 1" on graph paper is proly my best bet.
--
EA

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Sun, 6 Dec 2009 15:35:12 -0500, "Existential Angst"

This would have been an easy homework problem in descriptive geometry, a freshman engineering course back in the day. You construct three orthogonal views (top, plan and side). Then, using dividers and straight edges, you construct aux views until you get a "true view" of the bevelled joint. It's all graphical, no calculations or math.
This is how about everything was designed, including automobiles and airplanes. The guys that did it were well paid. They often worked on 1:1 drawings, wearing kneepads and holding eyeball accuracy to 1/100 inch.
http://en.wikipedia.org/wiki/Descriptive_geometry