Create a curve from series of theoretical points.

[[Ch...]]

How do I create a curve from theoretical points?   I feel like this should be possible, but I can't figure out how.  I have been doing too much management stuff and even have my own cmm programmer now; please tell me that I'm not losing my touch.   

 

[[Cl...]]

Just a thought. What if you open the curve strategy, click on segment - List. Here you can save the points.

I'm wondering if you can open the point list in Notepad and add the points?

[[Ma...]]

There are so many ways to alter curves.

I am lost at what you want to achieve - as Clarke stated - you can save points from curve, alter it in text editor and then load nominals into new curve.
You can rotate nominals, move them. You can even mirror curve - with and without help of model.

[[Cl...]]

Trying it in simulation now.

Edited April 30

[[Cl...]]

With existing curve. Click on the down arrow on the left. Scroll down to the last point. Right click, copy point.

Paste point. A dialog box will open with the XYZ, IJK values to edit. I just tried it and it worked.

[[Ch...]]

I don't think I'm being clear about the goal.  There is no existing curve.  The points are not in a straight line and can't be created with curve.  It's hard to tell in the picture, but the theoretical points are all in the same plane, while the original points are not.   I could make it using FreeForm, but I can't report radius's as they change along the compressor wheel profile.   

They are individual points taken along the edge of a compressor wheel.   Then I create theoretical points and use formulas to put all the points in the same plane.  

In a perfect world I could recall those theoretical points into a curve.  But it won't let me.  

 

[[Ma...]]

Until monday i can not try that, but you should be able to recall points into curve. Just don't use theoretical - use normal points with strategy, where formulas will be in nominals. This should let you recalculate curve.

Otherwise you would need SW for models and make that curve from model.

[[Cl...]]

Assuming you have the XYZ/IJK of each point? They can be entered individually into a curve feature. Or am I still not getting it?

 

Edited April 30

[[To...]]

Is this close to what you're looking for?

[[Ow...]]

It's been awhile since I've done it but, basically type your nominals into different columns of an excel document and then save the document as a (Tab Deliminated).txt file and then read that file into curve like Tom defined above.

In the 2D curve example illustration I've attached, first row is X, 2nd Y, 3rd row Z.

If it is a 3D curve, you'll need the Unom, Vnom,wnom's  vectors as well.(I,J,K).

Edited April 30

[[Ow...]]

I tried to edit my original post to include that you have to have nominals in a curve to recall points into it, it won't recall points to create nominals.

[[Ma...]]

Please sign in to view this quote.

You can recalculate so nominals can be created. But i think it won't work on theoretical points - it has to be normal points. Recalculating without actuals will make nominal curve from nominals - with actuals will create nominal curve from actuals.

It works the same as recalling measured points from standard features.