I have a sketch with 149 data points (same plane) that, when connected with individual lines into a closed loop, resemble (loosely) a bananna (vane profile). I'd like to smooth out these lines into a continuous arc with tangency.
I've played around with various commands: spline, fit curve, etc. but since I'm not familiar with the majority of these commands, I'm not getting the results I had hoped for.
I think 'Studio Spline' is a good option to create continuous arc with tangency using different options available in studio spline.
Please refer below documentation which will help you to achieve what you desired.
Since all the points are in the sketch,you can also use sketch constraints to constrain pole locations and end tangency.
It will be good if you can provide some image/part file or may be hand drawn sketch which would explain the expected result.
Here's a few snips of the sketch showing the overall view and both "ends" of the part. With it as-is, it will still produce a good part (I have .010 profile to work with), but I'd like for it to look better.
I'll look into the studio spline some more, but I couldn't get it to work correctly earlier. It seems to work better if there's more than one spline vs. an enclosed loop. Also, studio spline doesn't seem to want to let me select points by themselves unless they're already connected with lines?
Keep in mind that this part is roughly an inch tall, so some data points (especially near the "ends") are only a few tenths apart in one axis or another.
I think you need to ask how precise you need these splines to be.
If you try to create a spline through every point, you will get garbage between points, because you would be including the inacuracy in each measurement point to the Nth degree.
The art of splines is to get the simplest curve that accurately represents what you want.
You could arbitrarily pick points to come up with a spline that visually fits the rest of the points, or you would need to use something like Fit Curve to give a bet fit between the points.
You probably need to use multiple splines, since the surface of the vane is entirely different than the end radii.
Fitting splines to points can turn into a mess very quickly if you're not familiar with the nuances of splines (degree/segmentation) and how those nuances can affect the resulting surfaces. Most of this is dependent on your design/surface requirements and what you're attempting to model.
With all due respect to previous responses and in addition to what @MarkLawry has suggested, you can also try using analytic geometry (lines and/or arcs) to simplify the profile which you're trying to create (you may not have to use all splines in all areas - you might be able to do this will all arcs). You may not be able to fit every single point onto your final curves - but that's sometimes how trying to fit geometry to points will go.
NX 184.108.40.206 MP11 Rev. A
GM TcE v220.127.116.11
GM GPDL v11-A.3.7
Thanks all for the replies. I ended up using 4 different splines; 1 at each "end" and 1 on each side. It got me pretty close to where I wanted to be.
As mentioned above fit representing the top surface points with one spine, then the bottom profile with another spline.
I would then create a conic curve for leading and trailing edges, fitting by eye then deviation analysis measurements.
If you are actually creating airfoil do not create the leading and trailing curves. Enlarge the high and low pressure splines, sweep those to create surfaces then conic blend or conic surface the lead and trailing.