Surface-to-surface tangent mate? 1

[Theophilus]

Does anyone know if it's possible to mate two surfaces as tangent if both surfaces are irregular? I have an IGES file of an air foil I'm importing (irregular) and I need to make a fixture assembly to do some work on it. I'd like contacts to the air foil at particular points (for maximized accuracy) with a revolved part (similar to a tire).

I haven't figured out how to do this in SW2001+. Can it be done in 2003?


Jeff Mowry
DesignHaus Industrial Design

http://www.designhaus-i-d.com

 

[StarrRider]

Theophilus

I am not sure if this is what you want or not, but you can make any cylindrical part tangent to a completely enclosed surface like a slot. This was one of the mates added in 2001 Plus.
I just made 2 parts. The first was a rectangle with 1 side changed to a construction line. I added 2 lines to make a point and then added a 1" radius to all of the corners and extruded. The second was a revolved box (like your wheel).
In the assembly, I added a mate so that both parts were on the same plane. Then I selected all of the inner surfaces on the pointed box and the outside of the wheel and hit mate. Make sure you do a preview because invariably the wheel will want to mate backwards to what you want.
Selecting Move allows the wheel to appear to roll around the pointed track.

Lee

 

[StarrRider]

Theophilus

I forgot to mention that the first mate is important. SW will generate an error if you try to add the second mate first, at least it will if it has to change the orientation of the wheel.

The mate that is created is called a CamMateTangent.

Lee

 

[Theophilus]

StarrRider,

Thanks for the tip on the Cam Mate Tangent. I'll try it out. Do you know if this is limited to cylindrical surfaces to something straight (flat)?

In my case, one surface is the compound/irregular surface of an air foil and the other is essentially a revolved circle. This way, the contact between the two parts will be made at a single, calculated point (predictable and precise).

If there is no way to mate these, perhaps I can position my fixture "tires" generally where I want them, then collide them into the air foil and maintain their position at the collision point. Anyone know if this will yield accurate results if I have SW stop the movement at the collision point?

Thanks for the help.


Jeff Mowry
DesignHaus Industrial Design

http://www.designhaus-i-d.com

 

[StarrRider]

Theophilus

I think it is limited to cylindrical surfaces (or cams) to a flat path. However, I don’t think that should be a limitation for you.
I may be all wet here, but you could add a plane in your assembly and extrude a profile where your air foil crosses it. Make the extrusion nice and large in the first place so that you can easily select all of the surfaces. Make your wheel flat instead of round but add a controlling dimension for it’s width. After the Cam Mate is added and the wheel is rolling around like you want, change the width of the extrusion to .0001 and do the same thing with the wheel. It may not be a point but it is awfully close.

Lee

 

[TheTick]

You may try generating datum planes in each of you respective components that are tangent to the surfaces.
1.)use 3Dsketch to place a point at the desired location of tangency on the surface.
2.)place an axis thru the point and normal to the surface.
3.)place a datum plane thru the point and normal to the axis

 

[JNR]

If I understand your problem correctly, I think it is more an issue of math logic. I am not surprised that it is missing from SW mating capabilities. A "cylindrical" surface (ie: not necessarily circular in cross-section but extruded on an axis a-la SW "cam" surface) has a straight line edge when sectioned parallel to the axis. That makes it possible to relatively easily compute the tangency. Even a tapered cylinder (cone) might stand some chance of success. For any general case 3D surface this is not true and it would take some effort to even check for this condition. Add to this the assumtion that the edge was a straight line and not curved in one or more planes! Thus in the general case of two warped surfaces one could only expect them to be tangent a one given location on each. Even this would require other constraints on their orientation, otherwise there woud be an almost infinite number of solutions. You are asking for the surface tangency to match at all points along the edges of both surfaces, which may be quite common in your application, but not in the general case. Yours are only that way because you deliberately contructed them like that. Maybe you should be looking at mating the two edges and using some of the surface merging tools to create the right conditions or deriving some surface normal information and utilizing that. If the surfaces are constructed from sketches normal to the edge try this. Add a tangent construction line in the sketches at each end (or a perpendicular line, etc.). Show the sketch temporarily and mate the two sketch lines. It's brute force, but I've done simlar things. It should also updeat correctly if you change the sketch/surface definintion.

 

[StarrRider]

Theophilus

I just went back and reread this entire thread. The Cam Mate Tangent idea would allow you to view your tool but it would not provide you with a path for the tool to follow. I do too much Scan reading, sorry.
Wouldn’t a simple sketch offset provide you with the tool path that you need?

Lee

 

[Theophilus]

Thanks to all who have responded. I think the intent for what I was needing this sort of mate for was a little vague in my original post.

From what I can tell (haven't tried it yet), TheTick probably nailed the solution to my intended need. My fixture/jig needs to make contact tangent to the air foil surface at several specified points, but sketch cross sections always yield splines--and though I may be able to find tangency to the spline in 2D, I truly need tangency to the surface in 3D at my specified point. The axis/plane trick will probably allow this. Thanks! I didn't know about axis orientation of this sort before and therefore wouldn't have considered using an axis for this purpose.


Jeff Mowry
DesignHaus Industrial Design

http://www.designhaus-i-d.com