23-Emerald IV

Question

Blend edge to surfaces

Forum|Forum|12 years ago
November 13, 2013
13 replies
27062 views

I am looking for suggestions on how to blend tangent from a variable radius to a non-planar edge. The perfect command would be something like "blend edge to surfaces". Sort of like the "blend section to surfaces" command, but to an non-planar edge or curve instead of a planar section . Any ideas? Pictures and step file are attached. The yellow lines show where the surface would be in 4 locations. Thanks.

Tom U.

This thread is inactive and closed by the PTC Community Management Team. If you would like to provide a reply and re-open this thread, please notify the moderator and reference the thread. You may also use "Start a topic" button to ask a new question. Please be sure to include what version of the PTC product you are using so another community member knowledgeable about your version may be able to assist.

2_surf_blend.zip

Data Exchange

D

dgschaefer

21-Topaz II

I've attached a Creo 2 model.

[cid:image002.jpg@01CEE058.CD3612E0]

I made a copy of the edge of the edge of the extruded surface (turquoise), and then extended the surface. I then created a surface to surface round, through the curve (copy of the old edge). It wanted to round both the inside and outside, so I had to exclude the inside pieces on the 'Pieces' tab. It then only rounded the ends. I created the middle, longer parts separately after, selecting only an individual curve segment rather than the entire curve as I did at first. It did generate several geom checks, which I didn't investigate.

I did not try to merge it together, that may prove challenging as well. I'd start by merging the 4 round surfaces, then merge it to the upper then to the lower.

Good luck,

--

2_surf_blend-prt.zip

M

MikeBrattoli

12-Amethyst

Tom,

Is this what you're trying to achieve?

[cid:image010.jpg@01CEE05A.85D1CA00][cid:image011.jpg@01CEE05A.85D1CA00]

Mike Brattoli
Moen Incorporated
Global Product Development Process Management
Administrator - PLM
25300 Al Moen Drive
North Olmsted, OH 44070

M

MikeBrattoli

12-Amethyst

I created 4 curves thru points, making each end of each curve tangent to the vertical surfaces (top and bottom) then built a boundary blend curve using the bottom edges of the top surface and the edges of the variable round as my first direction curves with the 4 curves thru points as my second direction curves. I set the boundary blend border conditions to Tangent (top and bottom edges) and got the results below

Hope that helps!

Mike Brattoli
Moen Incorporated
Global Product Development Process Management
Administrator - PLM

[Description: Description: New Moen Logo.tif]

[Description: Description: Description: Description: Description: Description: Description: Description: fb]<">http://www.facebook.com/moen> [Description: Description: Description: Description: Description: Description: Description: Description: Description: t_logo-a.png] <">http://www.twitter.com/moen> [Description: Description: Description: Description: Description: Description: Description: Description: YouTube] <">http://www.youtube.com/moenfaucets> [Description: Description: Description: Description: Description: Description:

P

ptc-2860737

1-Visitor

Hi

I did something similar with boundary blends however, I didn't bother with
the curves. I just used the two surface edge curves (top and bottom) and
set tangency to the adjacent surface. I tried all the way around in one
feature but that didn't seem to work, so then I did the long sides first
followed by the short curved ends after (four blends in total) and with
the shorter blends I used the long blends side edges as the second
direction curves.

Doug Barton
Mechanical Designer
Parker Hannifin Canada
Electronic Controls Division
1305 Clarence Avenue
Winnipeg, MB R3T 1T4 Canada
direct 204 453 3339 x7309
fax 204 452 7156
-
www.parker.com/ecd

1_blankfilename.txt

T

TomU

Author

23-Emerald IV

Actually, in this particular case I do NOT want tangency between the spherical surface and the blend. The radius will be added later. The non-tangent edge should set the angle for the tangent, straight-line blend to the variable radius. Something more like this:

[cid:image010.jpg@01CEE061.887FF5E0] [cid:image012.jpg@01CEE061.887FF5E0]

To achieve what you see above I am "faking" the variable radius with a variable section sweep driven by a graph. The graph represents the radius values at the different control points. This "sort of" works for simple, 4 point variable radii, but I would much rather blend to the actual radius without trying to recreate it with a graph. In the pictures above, the "faked" radius is on the left and the actual radius is on the right. As you can see, the tangency isn't quite perfect. (This might be a function of my graph as well - just a spline through points.)

Tom U.

D

dgschaefer

21-Topaz II

I see now. That's not simple. I've created an approximation, but it's not 100% accurate. Maybe close enough?

I assumed your model was symmetric in both directions and cut it to 1/4 and built that:

[cid:image002.jpg@01CEE065.8DE1B060]

I made 3 sketches through the upper edge and tangent to the lower radius. Two are on the midplanes, the third is on a plane through the arc in the variable fillet. I then made a sketch on the long mid-plane (FRONT) consisting of a 3 point spline through the ends of those sketches and projected it onto the variable radius surfaces. The end of the spline on the RIGHT plane must me constrained to perpendicular to that plane. I then made a boundary blend using those 3 lines, the projected curve and the edge of the upper surface. It's constrained normal at the mid-planes and tangent to the variable radius. I doubt it's a perfect straight section at all points, but it's a close approximation. Adding other intermediate curves and additional spline points might help, but would be difficult to properly orient. Lastly, I merged it all together; it only needs to be mirrored, merged and solidified.

Again, a Creo 2 model is attached.

--

1_surf_blend_2-prt-2.zip

R

rrich

1-Visitor

Here was my solution,
Trim to ¼ like Doug.
[cid:image001.png@01CEE082.924F1670]
Create spline to trim lower surface notice 90 degree dimensions these are important for boundary surface to have proper tangencies.
[cid:image003.png@01CEE082.924F1670][cid:image004.png@01CEE082.924F1670]
Create boundary blend 2 curves in long direction top and bottom. 2 curves in vertical short direction at centers (straight lines)
Straight lines are constrained to be normal to the center datum planes. Bottom long curve is constrained to be tangent to the surface
That was trimmed by the last cut. Edge at the top has no constraints.
[cid:image002.png@01CEE083.1DE9EB10]

I could not use control points to get out an additional surface quilt line. ( Further playing and changing to arclength control and it got better)
[cid:image015.png@01CEE083.74936F40]
Mirrored all around and added a 1" radius.

[cid:image016.png@01CEE083.A5DE4CF0]

Ron Rich
Mechanical Engineer
Keithley Instruments, Inc.
28775 Aurora Road
Solon, OH 44139
440-498-2842

D

dgschaefer

21-Topaz II

Ron,

You and I are on the same page, but I missed the second perpendicular constraint on your spline. Good catch, that cleaned it up a bit more.

If you add the interim straight curve like I did it eliminates the arced transition line that you called out in the boundary blend. There are still two lines there, but they are very close to straight.

[cid:image006.jpg@01CEE086.01C1A330]

--

P

ptc-2860737

1-Visitor

Hi

There is a function in WF4 called "Blend Tangent to Surfaces" I think this
does the trick.

Doug Barton
Mechanical Designer
Parker Hannifin Canada
Electronic Controls Division
1305 Clarence Avenue
Winnipeg, MB R3T 1T4 Canada
direct 204 453 3339 x7309
fax 204 452 7156
-
www.parker.com/ecd

21-Topaz II

Wow, I think that's it exactly. I had no idea that existed.

--