Skip to main content
T
TracyWillis10-Marble
10-Marble
October 20, 2009
Question

Control O.A.L. Of An Equation Curve

  • October 20, 2009
  • 8 replies
  • 2109 views

I need to generate an equation curve and be able to set and control the length of the actual curve. Making the equation curve to form the helical “corkscrew” isn’t a problem.I just can’t figure out how to control the “R” value of the curve based on my total length.

Is this doable?

Thanks

TW

      8 replies

      C
      CM101-Visitor
      1-Visitor
      October 20, 2009
      You can put a point on the curve with the desired length from the correct end, then put a datum thru the point normal to the curve, then trim the curve.
      L
      lwh1-Visitor
      1-Visitor
      October 21, 2009
      If the curve is analytical, why not calculate the radius from the required length in a part relation and use it in the curve equation?
      T
      TracyWillis10-MarbleAuthor
      10-Marble
      October 30, 2009
      Summary

      Thanks to Terry ! Very helpful info.

      The solution: Create a parameter named "RADIUS" real number with an arbitrary initial value. In the equation for your curve, set the radius equal to the "RADIUS" parameter. Create an analysis feature for the curve and save the length of the curve as a parameter value. Analysis > Measure > Length > Select your curve use the drop-down to create it as a Feature, and name it "curve_length" click on the Feature tab to ensure the parameter is checked to be created, and has the name "LENGTH" Use a Feasibility Study to find the value of R to give a desired curve length: Analysis > Feasibility/Optimization >select the Feasibility radio button under Design Constraints, select "Add", and then select "LENGTH:CURVE_LENGTH" if not already selected, then set the middle box to "=", and select the "Set" button & enter the desired value for the max length of the curve and hit OK, then Cancel from the Design Constraint box. Under Design Variables, select "Add Parameter", and select "RADIUS:whatever_your_part_is_named” enter Min/Max values for the radius value range that will be allowed. You may need to also tighten the default convergence value on the options tab > Preferences > Run tab: then hit Compute:

      Couple things to consider: Make sure that it is possible for Z & T to generate a curve larger than the length set within feasibility. That is also true for the Minimum and Maximum R values. You may need to tweak the min / max R values in feasibility if the curve is coming up short. If everything seems to be working fine but the curve is still slightly short check the convergence setting. If that value is too large it will “clip” the curve and can be shorter than the desired length you entered.


      I didn’t try Frank’s suggestion, I had already use the above process.

      I tried Harald’s suggestion; I was able to trim a sketch curve but not the equation datum curve.

      Thanks – and Stuff.

      Have a great weekend everyone.

      tw
      C
      CM101-Visitor
      1-Visitor
      October 30, 2009

      If I understand this correctly, you want a helical curve of a certain length, and you want both the ends to be coincident to a plane thru the axis of the helix (one turn)? Somewhat like a helical lock washer? You're going to use this curve as a trajectory, I assume? If so, I have an easier solution for you. Shoot me an e-mail.

      Frank

      In Reply to Tracy Willis:

      Summary

      Thanks to Terry ! Very helpful info.

      The solution: Create a parameter named "RADIUS" real number with an arbitrary initial value. In the equation for your curve, set the radius equal to the "RADIUS" parameter. Create an analysis feature for the curve and save the length of the curve as a parameter value. Analysis > Measure > Length > Select your curve use the drop-down to create it as a Feature, and name it "curve_length" click on the Feature tab to ensure the parameter is checked to be created, and has the name "LENGTH" Use a Feasibility Study to find the value of R to give a desired curve length: Analysis > Feasibility/Optimization >select the Feasibility radio button under Design Constraints, select "Add", and then select "LENGTH:CURVE_LENGTH" if not already selected, then set the middle box to "=", and select the "Set" button & enter the desired value for the max length of the curve and hit OK, then Cancel from the Design Constraint box. Under Design Variables, select "Add Parameter", and select "RADIUS:whatever_your_part_is_named” enter Min/Max values for the radius value range that will be allowed. You may need to also tighten the default convergence value on the options tab > Preferences > Run tab: then hit Compute:

      Couple things to consider: Make sure that it is possible for Z & T to generate a curve larger than the length set within feasibility. That is also true for the Minimum and Maximum R values. You may need to tweak the min / max R values in feasibility if the curve is coming up short. If everything seems to be working fine but the curve is still slightly short check the convergence setting. If that value is too large it will “clip” the curve and can be shorter than the desired length you entered.


      I didn’t try Frank’s suggestion, I had already use the above process.
      I tried Harald’s suggestion; I was able to trim a sketch curve but not the equation datum curve.
      Thanks – and Stuff.

      Have a great weekend everyone.

      tw

      C
      CM101-Visitor
      1-Visitor
      October 30, 2009
      If I understand this correctly, you want a helical curve of a certain length, and you want both the ends to be coincident to a plane thru the axis of the helix (one turn)? Somewhat like a helical lock washer? You're going to use this curve as a trajectory, I assume?

      Anyways, I modeled the above in a much simpler fashion. If it's a constant radius and you don't need to assign a graph function to it in any way (like needing ends normal to the plane or variable radii or pitch) you don't need to use an equation curve. In my model with the absolute accuracy set to .0001, the curve is accurate to .0000001 inch. The only drawback is Pro/E's usual inability to finish the regen cycle, so you must regen twice. If this is what you wanted, it's far less monkey business than the other method.

      Frank

      C
      CM101-Visitor
      1-Visitor
      October 30, 2009

      Oh, and I forgot to mention, making multiple turns is as simple as changing the section relation for the cylinder surface as follows for 2 turns:

      sd3 = rsd2/2

      For 3 turns, make that /3, etc.

      Have fun!

      Frank

      C
      CM101-Visitor
      1-Visitor
      November 3, 2009

      Anyone play with this yet?

      T
      TracyWillis10-MarbleAuthor
      10-Marble
      November 4, 2009
      Hey Frank,

      I briefly looked at the 2nd model you sent me but I haven't had a chance
      to really dig into. We had to move on for the moment because the
      summary I posted actually did work for us. It may have not been ideal
      but we were able to clear the hurdle and keep the project moving. I
      really like the way you controlled the pitch of the helix, never seen
      that approach before. I will not share the file and destroy it as you
      requested once I am finished learning from it.

      I may post another summary after I tear into this again next week.



      Tracy Willis

      Designer / Drafter



      Cook Urological, Inc.

      11OO West Morgan Street

      Spencer, IN 47460

      (812) 829-4891

      (812) 829-1801 (fax)



      Confidentiality Note: The information contained in this e-mail is
      strictly confidential and privileged information which is intended for
      the use of the above addressee(s) only. All other use is strictly
      prohibited. If you are not the intended recipient, any review,
      distribution or copying of this document is strictly prohibited. If you
      have received this e-mail in error, please notify the sender immediately
      and delete the document from all computer systems, or notify Cook
      Urological at (812) 829-4891.


      Terms & ConditionsAccessibility statement