Applies To
	Product(s):	AutoPIPE,
	Version(s):	2004, XM, & V8i
	Area:	Modeling
	Original Author:	Bentley Technical Support Group

Problem:

Depending on the methodology for modelling a beam to beam T intersection results in a difference in natural frequency results.  Displacements also vary but are within a margin of error. Is there any reason Why are the results are not exactly the same between following modelling methods?

Beam A:                                                                                                 Beam B:

                         

Beam C: (made of 2 beams, C and C-1)

      

Solution:

 Based on some observations from reviewing a model with the aforementioned beams and interpreting the online help:

1. 1. The ‘Rigid length’ is deducted from the geometrical length of the beam for the analysis. With shortened length, the mass and stiffness are different (compared to full length).

   2. Instead of using the ‘Rigid length’ at each end of the beam, one could physically create different beams corresponding to those rigid lengths at each end and mark them as ‘Offset’ beams. As a result, the ‘offset’ beam length is treated as rigid length.

   3. Note that static analysis is affected by differences in stiffness properties due to different lengths and different material properties. Shorter length means smaller mass which is adds to the differences in mode frequency.

   4. The stiffness in the anchor and stiffness in the beam is different. (see Beam properties, and Tools> Model options> Edit> Anchor Stiffness.
   5. As seen in AutoPIPE's online help:
      
      

See Also

Beam Structure modeling

Bentley AutoPIPE