Is there a way to make a line tangent to two splines?

[Art5]

I posted about this in an existing 'Feature Request" thread, and was advised that making a new thread might be a better route to go. I'm well aware that the Fusion team has chosen not to implement the ability to create tangent constraints in reference to splines (even though other Autodesk products have this capability) but I have need for this on a regular basis. I routinely have to project geometry from existing part data into a sketch and then constrain other sketch geometry to the projected geometry. I've searched the forums and found multiple threads where others have a need for this as well. Several 'work arounds' have been posted but as far as I've been able to find none of these work arounds actually work, as they essentially require you to guess where you think the point of tangency might be rather than pinpointing the exact location of tangency. I've attached a sketch from a recent project that highlights what I need to do. 

As always, any and all help is greatly appreciated.

Thanks,

Art

[etfrench]

Your file requires us to guess where and what you want tangent to what.  What are the required tolerances?

[TheCADWhisperer]

@etfrench 

My guess is...

[Art5]

As usual, @TheCADWhisperer is one hundred percent correct!

As far as tolerances, for this particular purpose if it was within about 0.001" that should be acceptable for what I need to do. Obviously if it was right, that would be ideal. If I draw two arcs or circles and a line in a sketch, it's very simple to constrain that line tangent to both arcs or circles, and it will be exactly right. I'm still struggling to understand why this isn't the case for splines. Seems like a major oversight. 

If I could dimension the line from the splines (I can use the Measure tool to measure the 'minimum' distance from the line to a spline), it would seem that I could use the Fix constraint on the spline and I could then adjust both dimensions to the same value and do an offset of that value to put a line where I need it.

[TrippyLighting]

I wonder what you are trying to achieve with the imported DXF data.

I am not trying to be abrasive, but some basic inspection reveals that the end points of the  two curves are not even coincident. 0.005mm is not a lot in the realm of physical artifacts, but for a CAD software that is wide open like a barn door.
Even if you connect the curved tangentially, it will not create a proper sketch profile for extrusion or modeling in general. 

I completely support the request to be able to make lines tangent to spline curves without having to use convoluted workarounds or approximations, some of which I developed myself 😉

Any other CAD software I use has that ability. SolidWorks, Autodesk Inventor, ZW3D.

[Art5]

"I wonder what you are trying to achieve with the imported DXF data."

I'm glad you asked! This dxf was exported from the file I'm working on in Fusion. The sketch will ultimately serve as a machining boundary (I would have gotten around to making the end points coincident if I hadn't gotten stuck trying to get the line tangent to the splines). Our machines are only accurate to about 0.0005", so 0.005 mm (0.00019") would be acceptable for my application, aside from the fact that I want the chain to be continuous. 

[Art5]

Well, here we are again. I'm working on a new project, and have yet again run into this problem. I have two splines that I've projected from part data. I need to make a line tangent to both of them. Super frustrating that I'm unable to simply apply a couple of tangent constraints. As Trippy has stated, "Any other CAD software I use has that ability." 

Wondering if anybody has a good method to achieve this. Thanks in advance. 

[davebYYPCU]

In your previous thread, you rejected the work around using trimmed arcs, 

don’t hold your breath waiting for Autodesk to rescue you.

[Art5]

As stated MULTIPLE times in the other thread, the work around you're referring to isn't acceptable due to the fact that it doesn't pinpoint the actual correct location of tangency. Since you seem to really be having trouble comprehending the problem with the work around I'll reiterate. It requires you to GUESS where you THINK the point of tangency MIGHT be. Words like 'guess' and 'might' clearly aren't indicative of a solution that is CORRECT. While I appreciate your attempt to "solve" this, there's no need to pop in just to be a jerk, Dave. That doesn't accomplish anything productive. I have to do this regularly, I need a solution for this issue that actually works. I don't want to have to post on the forums every time I need to do this and hope that someone here can pull it into another software and provide me with the "fixed" sketch. I need a method that works that I can do on my own.

[davebYYPCU]

Show me how the line is not tangent to both splines.

[Art5]

Sure thing. If you look closely at this picture, you can see that when I try to trim the line and the spline to one another, they trim at different locations. This shows that the method you used to achieve this hasn't pinpointed the correct location of tangency. In the other thread you used the phrase "good enough for gov't work". Again, perhaps for what you do this method is "good enough", but for what I do I need it to be right, not "close enough".

 

[davebYYPCU]

Did you trim against the spline or the rectangle?  My spline was green (fixed)

[Art5]

I assume you mean the construction lines, that's the only rectangle I found. I deleted the rectangle prior to trimming. The spline turned from green to blue when I trimmed it. My guess is that the trim operation removed the fix constraint which is why it's no longer green. 

[davebYYPCU]

The tangent line was constrained Colinear to the rectangle’s long side. (Both splines)

It’s midpoint was coincident to the spline, remove constraints, and the tangent line is now unconstrained,

Trimming to a midpoint, is an unknown function.  You will need a break line off the midpoint for that test.  
Trim both sides and the line will disappear if accurate, and if not, measure it (- I get 0.0058xxxx of a mm.)

[Art5]

The spline didn't turn from green to blue until the trim operation. 

I also get the same measurement. You asked me to show how the line is not tangent to both splines. I believe this demonstrates that the line is not tangent to both splines; regardless of how far it's off, it's still off. I'll be happy to admit that it's close, but still not actually correct. 

[bwalker145]

My workaround has been to build a spline on the projected geometry, and use a collinear constraint between the two tangent handles to define the relationship. It's unfortunately another approximation, but has worked well for my projects.

+1 to wanting an actual tangent to spline function.

[bwalker145]

@Art5 

I just tried another workaround that as far as I can tell, develops a true tangent relationship with a spline feature.

I created the top component, and used two "Tangent Relationship" constraints under the Assemble tab. It was a little finicky, I had to roughly position the top component before applying the constraints. Might have worked a little better if I modified the bottom surface of the bar into two separate coplanar surfaces.

Sample size of 1 so we'll see how it works in other situations, but this might be a viable option.

[bwalker145]

@Art5 

I tried this last method on your file and think it worked well. Take a look, should be a true tangent-to-splines line.