GDL: Ruledshell & Generatrices

[Lingwisyer]

Hi all,

I have a ruled shell where one end is a pill shape and the other is two 180 degree arcs to form a circle with a mid point. From my understanding, you use the generatrice to inform which node connects to which. So for my case I am wanting the first and final nodes of the pill to connect to the start of the circle, line start (1) and line end (4) to arc start (1), and the second and third node connect to the mid point, line end (2) and arc end (3) to arc end (3?). The result I am getting is that everything seems to connect to the starting node of my circle...

	RULEDSHELL mat_Int,	mat_Int,									!Base
		mat_Int,	mat_Int,	mat_Int,	mat_Int,	mat_Int,
		5,	5,		4,
		0,	tShell,	1,	0,
		! transformation matrix
		1,	0,	0,	0,
		0,	1,	0,	0,
		0,	0,	1,	0,
		! Bottom Profile
		xWaste - rWaste,	yWaste,	1,
		xWaste,				yWaste,	900,
		0,					180,	1+4000,
		xWaste,				yWaste,	900,
		0,					180,	1+4000,
		! Top Profile
		-A3/2 + B/2 + xInt,		B3/2 + yInt,	2,
		A3/2 - B/2 + xInt,		B3/2 + yInt,	2,
		B3/2,					-180,			2+2000,
		-A3/2 + B/2 + xInt,		-B3/2 + yInt,	2,
		B3/2,					-180,			2+2000,
		! transformation matrix
		1,	0,	0,	0,
		0,	1,	0,	0,
		0,	0,	1,	0.015,
		! generatrices
		1,	1,
		3,	2,
		3,	3,
		1,	4

Ling.

[runxel]

Ling, that is my understanding of the generatrix as well.

Are you able to give an example with is fully parametrized already so we do not have to guesstimate the correct values for your variables?

[Lingwisyer]

Here is a copy of the shell without the offsets.

[runxel]

Try this, this works for me.

Just make sure that both polylines go into the same direction and CCW.

RULEDSHELL mat_Int, mat_Int, mat_Int, mat_Int, mat_Int, mat_Int, mat_Int,
		5,	5,		4,
		0,	tShell,	0,	0,
		!// transformation matrix
		1,	0,	0,	0,
		0,	1,	0,	0,
		0,	0,	1,	0,
		!// Bottom Profile
		0,	rWaste,	1, 				!#1
		0,	0,		900,			!#2
		0,	180,	1+4000, 		!#3
		0,	0,		900,			!#4
		0,	180,	1+4000,			!#5
		!// Top Profile
		A/2,	B/2,	2,			!#1
		-A/2,	B/2,	2,			!#2
		B/2,	180,	2+2000,		!#3
		A/2,	-B/2,	2,			!#4
		B/2,	180,	2+2000,		!#5
		!// transformation matrix
		1,	0,	0,	0,
		0,	1,	0,	0,
		0,	0,	1,	0.5,
		!// generatrices
		1,	1,
		1,	2,
		3,	3,
		3,	4

[Lingwisyer]

Ah, you rotated the circle 90 degrees. That does work a lot better, but it would appear that my actual issue that the two faces of my shell do not like being so close to each other, so I either get the "thickness" intruding into the hole in the middle, or it spreading out beyond the edges...

I might try using a zero thickness shells to do two SEOs on a poly. 

[Lingwisyer]

I forgot there is no SEO with upward / downward in GDL... and you cannot SWEEPGROUP a non-solid...