Jump to content
AutoDesSys Forums
kmwhitt

Simple Two Source Sweep

Recommended Posts

I would appreciate it if someone could give me pointers on performing what should be a simple two source sweep.  In the first image you see the two sources and path being used.  In the second image you see the best I can do using centroid alignment.  The other alignments don't seem to work.  The result seen in the second image is not acceptable as the shapes on either side of the "channel" are not symmetrical (easier to see in full shaded - third image).

 

In past versions of FMZ I would have approached this using faceted sources.  This way I knew how many points needed to be on each source and how they should align.  How is this supposed to be accomplished now that everything is smooth?  For example, should I be adding more control points to the second source shape?  If so, how many and where are they to be located?

 

Something else that troubles me is that I don't seem to be able to use "as positioned" for the 2-source sweep.  This would save me lots of time as I plan on creating several sweeps with various sources to outline some geometry.  It looks as though I will have to move each generated sweep into proper position if I can even get them to work.

post-10687-0-64008600-1458030211_thumb.jpg

post-10687-0-62606700-1458030216_thumb.jpg

post-10687-0-28140700-1458030221_thumb.jpg

2SOURCESWEEP.fzb.zip

Share this post


Link to post
Share on other sites

hi,

it is very easy to get what you want, when you use the right tool -

but it is not so easy to find out what the right tool is....

 

in this case try "Perpendicular Loft" with Caps on.

 

 

vva

 

Share this post


Link to post
Share on other sites

Hi VVA -

 

I explored this tool, however, my path has to turn a corner and maintain a flat 90 degrees on the bottom - not like what I get with the perp loft (attached).  This should be possible using a 2-source sweep, but it doesn't seem to work.  That said, I guess I could perform an additional axial sweep and union these two shapes together.  Thanks for the suggestion!

 

Kevin

post-10687-0-89635800-1458038960_thumb.jpg

Share this post


Link to post
Share on other sites

Hi Kevin,

 

Yes, this should be easy to do with the Perp Loft as VVA says, and it should also work as a 2 path sweep as you suggest.  Thanks for your report, we will get this corrected.

Share this post


Link to post
Share on other sites

I was able to get the shape I wanted using the perp loft, however, now that the geometry has differing topology, I am unable to boolean union the entire piece together.  Can anyone direct me on how I get the attached to be a solid - all in one piece?  I suspect were the 2-source sweep working properly, this would give me exactly what I need as the topology would match up.  As it is not working correctly, I will need to find another way...

post-10687-0-77268300-1458054110_thumb.jpg

perp loft.zip

Share this post


Link to post
Share on other sites

Hi Kevin,

 

One approach we were going to suggest would be to create the objects as surfaces whose edges touch each other, and then Stitch the results together. Once if forms a closed volume, it will then become a solid:

 

post-5-0-73959300-1458055504_thumb.jpg

 

However, as we were doing a quick check of this, we noticed that there are some mis-aligned objects with gaps in the larger back object:

 

post-5-0-04713400-1458055505_thumb.jpg  post-5-0-30808900-1458055505_thumb.jpg

 

Not sure if that was intended, but that is preventing the Boolean from completing properly.  Did you build it this way on purpose?

Share this post


Link to post
Share on other sites

No, the gaps were not intentional.  Thanks for pointing these out.  Your suggestion regarding the surfaces, is this deriving surfaces from the existing geometry or starting over from scratch?  If starting over from scratch, how would it be possible to get the transition from the dented trim to the smooth?  Are you suggesting another perp loft?

Share this post


Link to post
Share on other sites

Hi Kevin,

 

Most of what you have is fine.  Most likely if you just fix the "bump out alignment" then you will be able to Union or Stitch the rest...

 

Does that help?

Share this post


Link to post
Share on other sites

The gap I found was in the large piece - not the trim pieces.  The large piece aside - when I try unioning the two profiles by themselves, the boolean does not take.  Are you finding a gap between the trim pieces also?  I have attached a stripped down version of the file - just the two objects.  I see no gaps between these two pieces and the operation will not execute.  If you do see a gap, please point it out.  Thanks.

two trim pieces.zip

Share this post


Link to post
Share on other sites

Hi Kevin,

 

These were not the parts we were showing.  For these, there is a complex "smooth / very nearly flat / NURBS face" here:

 

post-5-0-30965100-1458062719_thumb.jpg

 

Delete that and 3 others:

 

post-5-0-01099900-1458062719_thumb.jpg

 

Then Stitch the two surfaces together.

 

Another aspect to the problem is if you check a Top View and zoom in very close to the upper right corner, you will again see that parts are misaligned:

 

post-5-0-72848800-1458062718_thumb.jpg

 

Perhaps you should work with the Point Snap enabled more frequently (if you are turning it off)?

 

And / or perhaps you would benefit from a training class?  (Please contact sales@formz.com if you are interested in one of our custom 1:1 training classes...)

Share this post


Link to post
Share on other sites

I appreciate the feedback.  Of course, I use object snaps.  This form was created using a sweep, so if there was misalignment it was due to that operation and the perpendicular loft.

Share this post


Link to post
Share on other sites

Hi Kevin,

 

Neither of those objects are currently of Object Type: Sweep, so we can't really investigate how they were created.  If you can post the parametric object, or its sources, we can look into this further...

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×