I'm trying to model a hose that curves through 3 dimensions. In the
past, I have created a sweep, using the blue spline as the trajectory
but this only produces flat (2 dimensional) curves.
I created some points at varying X,Y and Z positions and through these
points, I made a datum curve that was close enough to what I wanted.
But the sweep won't recognize the datum curve as a trajectory.
Can someone help me with this or point me to a tutorial on how to do
The Sweep feature won't follow a spline. You could do it with straight
sections, single or multiple radiuses. If all you really need is to make a
circle follow the spline, use 'Insert>Advanced>Pipe'. You can specify solid
or hollow, OD and wall thickness. It kinda looks like a hose.
You cannot sweep along a spline.
Choose the best two views that describe the path and draw curves on
planes, make sure that the ends are coincident and that the curves do
not self intersect. Create a curve by intersecting these two and you'll
be able to sweep along it. If your run is complicated you may have to
make it in several parts.
Whilst you can't sweep along a spline, (splines don't have orientation)
you can create a multi-trajectory sweep along 2 splines, so if you
create a second spline parallel to the first, you can make the feature
as you describe it.
Hence, whilst 'pipe' will effectively make a rond section sweep along
a spline, if you want other shapes you'll then need to extrude a
surface through the pipe centreline, and use it's intersection with the
pipe as your second trajectory.
Can't try this on WF2 as I'm on WF3 but...
Are you using Insert > Sweep? I think using VSS will allow you to use curve
thru points. [I'm led to believe this should be the default function now for
a constant section sweep - unless you have nontangent chain or want to use
the fill option]
Have you got ISDX? A 'free' Style curve through your points should also be
Create a surface using the curve as a boundary and then use the surface edge
as your traj?
WF3 will use a curve thru points as a traj on a insert > sweep
Good question, Peter; lot's of good ideas guys. One of the suggestions, the
VSS still puzzles me. It's not that it's not possible. It just seemed, when
I first thought about it, that there's no good way to create the second
trajectory curve so that it controls the section size and shape as intended
and without kinking or selfintersecting. And we are talking about a 3d curve
that can wiggle all over the place, changing directions repeatedly, even
doubling back on itself. So a simple copy/offset wouldn't do. Any ideas? If
you can do it, I'd like to see it.
You can make the second curve, but the only way I know is to create a
pipe along the initial spline (which technically you aren't doing, but
if you put it through the same points with the same constraints the
centreline should be the same ) then extruding a surface through the
pipe, with the centre spline as sketch, intersecting the pipe surfaces
and the extruded surfaces to get a second curve offset from the first
by the radius of the pipe, cutting out the pipe, and using the two
I have only done this once, looked at the resulting model, thought
'what a mess, my intended section is more or less perpendicular to a
plane anyway, I'll let that drive the direction and accept the model
Maybe I've missed something in the discussion but you don't have to have two
trajectory in the VSS function, surely Peter only wants one traj?
Theres also an Option on the Dashboard - Constant Section Sweep
Additionally, unless there's something driving d(radius) it remains constant.
(The Constant Section option is interesting, allows the use of additional
trajectories that create conflicting or unsolvable constraint(?) and allows the
use of projected section curves.)
- - - - -
Wondering about the difference in users' perceptions re sweep functions; is VSS
relatively new to Foundation? Perhaps it's because I'm a newcomer to Pro/E
(started with WF), but seems to me that VSS is the 'one size fits all' sweep
function, the rest (older?) serve occasional or 'special purpose' needs. I've
gotten the impression from some older users that they think VSS is "over kill",
etc. and it's puzzled me.
I guess, Sean, that you know a good bit more about the VSS in WF2 than I do.
Thanks for the tip. And, yeah, works very cool. Well... oh, no, not in WF2
(or at least I can't confirm). I did it in the only thing I have available
for messin' around ~ the SE of WF3. So, it works great there.
Come to think about it, this doesn't accomplish anything beyond the Pipe
feature, if one wishes to do only a round hose. You can, with the new VSS,
create any shape to sweep and with only the central 'trajectory'. Well, I
guess I exaggerated: an elipse follows the normals and constraints based on
the twists in the 'curve thru points' foundation of the section. IOW, Pro/e
picks the sketch plane and orientation and your elipse WILL follow the
vertical/horizontal orientation. Or maybe this is where that second
trajectory comes in, where you want to get beyond the limitations of this
default orientation. And, Sean, you didn't just do an artful dance around
how to construct this second 3d section guide trajectory by bringing up the
single trajectory VSS, did you!?! I'd like to be able to tell that second
trajectory to twist, but stay equidistant from origin or to stay in a
normal plane but increase in distance, meaning the section expands. I don't
know how to do any of these with the resources available.
Part of VSS' beauty is that you don't need the second trajectory. (Or if you
need one for another function, create it with VSS.)
Place an angular dimension relative x or y reference. Drive the angle as a
function of trajpar.
Drive the dimension as a function of trajpar.
I think VSS tends to intimidate, sometimes, because if you don't understand how
it drives section constraints it will be frustrating. Once the lamp starts to
flicker, though, it becomes apparent it's an incredibly versatile function.
I get it tyhat you need only a single trajectory for a 'VSS', but then why
is THIS a VSS if you can't vary the section? On, yes, you may not vary the
section with a single trajectory. That was what the second and third
trajectories were for, to vary the section along those trajectories.
What are you referencing with this angle? Sorry, we're talking about an
elipse. When sketched, it goes n/s or e/w, along the orientation lines
(dashed vertical/horizontal) of the sketch plane. I've tried centerlines and
other modifying references for an elipse. Nothing changes its orientation.
Tranpar can control only what's controllable; the ellipse shape (or rather,
orientation) is not controllable. If you think so, please demonstrate.
I do not lack the flicker, merely the wattage to reveal the detail.
Is there a good reason to not use VSS if you want a constant
section? It is "the" sweep function with dashboard interface
(at least up to WF2, WF3?), dirt simple to set up, etc.
The (pictured) volute portion is a VSS using a single arc as
trajectory, driving the section (four curves) with trajpar.
(I ~think~ the entire shape could have been done as a
single VSS, a circle traj and evalgraph(s), but that wasn't
the object of the game.)
True. The sketcher ellipse is (?) a special case,
but there's a way to get the shape...
_ Start a VSS, selecting a single trajectory.
Let's use a circle.
_ Create a pair of perpendicular centerlines coincident to
origin point, at some angular displacement relative the
X and Y sketcher references.
_ On those centerlines, place four points to define Rx, Ry
(plus and minus, symmetric about the origin).
_ Connect the four points with four conic arcs.
_ Constrain each conic arc tangent direction perpendicular
to the centerlines.
_ Tools -> Relations:
__ The dim controlling centerline angle = 0 + trajpar * 360.
__ Rho for each conic arc = sqrt(2)-1, OK.
_ Complete the feature.
You've got a torid with rotating elliptical section. The only
difference between that and using a sketcher ellipse will be
two additional seam (two sided) edges on the Y axis. (It takes
less time to do than to describe.)
I believe there are a lot of VSS tutorials floating about.
is the only one that I could remember the location.
Web search for
pro/e "variable section sweep" | vss tutorial
turns of a lot of hits.
Yeah, sounds familiar. They, as the biggest users of Pro/e on the planet,
had some respect for it and for what it took to be a proficient user. I was
there as a contractor, the direct hires got the training. At least they went
through a tooling apprenticeship program to learn stuff about how a product
gets made. That was the best thing about Cat. At least I had the tooling
apprenticeship and the formal training on Pro/e before I got there.
Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here.
All logos and trade names are the property of their respective owners.