Announcement

Collapse
No announcement yet.

Family Contraints: RP vs Geometry

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    Family Contraints: RP vs Geometry

    Is there a general guideline for when to constrain something to a reference plane rather than the geometry itself? I'm currently trying to make some nested windows and keep running into over constraint issues and am wondering if they could be avoided by changing how I'm constraining.
    Revit for newbies - A starting point for RFO


    chad
    BEER: Better, Efficient, Elegant, Repeatable.

    #2
    Yeah: always constrain to a Refplane, never to geometry...
    Martijn de Riet
    Professional Revit Consultant | Revit API Developer
    MdR Advies
    Planta1 Revit Online Consulting

    Comment


      #3
      You can put dimensions driven by parameters between a reference plane and a geometry line without problems. When you modify the parameter the geometry will move, not the ref plane. But if you put another dimension between that geometry line and another geometry line, than you will have a problem So i`ll say the easy rule is to have at list one end of the parametrised dimension linked to a reference plane.

      But...
      Usually is easier to make more reference planes (for bouth ends) so you can concatenate a sckeleton for all the geometry items with only a few defined parameters. If you want smaller number of reference planes (like just to one end of your paremetrised dimension) it is very possible to have to create a lot of new parameters that are calculated values like a sum or a difference of the existing ones. So you have to choose more parameters and few reference planes ore just a few parameters and more reference planes. I prefer the second method it`s faster.

      Comment


        #4
        Originally posted by gaby424 View Post
        You can put dimensions driven by parameters between a reference plane and a geometry line without problems. When you modify the parameter the geometry will move, not the ref plane. But if you put another dimension between that geometry line and another geometry line, than you will have a problem So i`ll say the easy rule is to have at list one end of the parametrised dimension linked to a reference plane.

        But...
        Usually is easier to make more reference planes (for bouth ends) so you can concatenate a sckeleton for all the geometry items with only a few defined parameters. If you want smaller number of reference planes (like just to one end of your paremetrised dimension) it is very possible to have to create a lot of new parameters that are calculated values like a sum or a difference of the existing ones. So you have to choose more parameters and few reference planes ore just a few parameters and more reference planes. I prefer the second method it`s faster.
        Well, imvho option 1 is no good with complex families. Cause this will eventually cause the "overconstrained" issue. Especially with nested components. For instance (out of the top of my head, might be wrong example):

        Let's say you created a glass infill panel for a window. Just used one horizontal and one vertical refplane, constrained the extrusion on two sides and flex the other sides using parametrised dimensions (width and length). Now nest into the parent. This will break (overconstrain) as soon as you want to constrain the panel on four sides to the frame in the host. Or it will break as soon as you want to swap out panels.

        BUT
        If you created both the frame and panel using 4 refplanes defining the width and height you can easily lock them together in the host family.
        Last edited by mdradvies; March 3, 2012, 08:28 PM.
        Martijn de Riet
        Professional Revit Consultant | Revit API Developer
        MdR Advies
        Planta1 Revit Online Consulting

        Comment


          #5
          well i never constrain a nested family on more than 3 RP each one for x y z spatial directions. So in our discution i only constrain just at 2 ref planes from the host by using lock function to the geometry lines where i know i have a RP in the nested family. The length and height dimensions of the nested family are contro11ed by the linked values of the similar parameters that are recreated in the host family. And when i constrain panels that need to be swaped i use a zero dimension locked between a reference plane in the host and the reference origin from the nested fam.

          Why do i lock just with 3 RP? 3RP make a point in space. So is like an one glued point for the nested family. It is nailed trough that point at the host scheleton. To flex it i only use linked parameters. Not need another reference planes to risk overconstrained errors.
          Last edited by gaby424; March 3, 2012, 09:34 PM.

          Comment

          Related Topics

          Collapse

          Working...
          X