Split Physical Members in Risa3D

[Leadbelly]

I was wondering if anyone knew the difference between "Actually Split PHYSICAL Members" and "Just Add Joints to PHYSICAL Members" did to the model. I find that for the same section they produce different results and I am not clear when to use one method over another.

Any thoughts?

 

[JAE]

With a physical member with intermittent joints (i.e. "just add joints to physical members) you really have essentially one member with multiple joints. The program combines the solution and reports all the individual members as one member.

This physical member concept is used for cases where you might have a long member with intermittent connected joints and yet want that long member to be dealt with as one entity - similar to the top chord of a long truss where you want the whole top chord to be treated by RISA as a single member, yet there are occasional joints and connecting members that affect the matrix solution.

If you "actually split physical members" you are breaking the one single member in to multiple two-joint members.

In either case, the results should be the same in terms of the matrix solution (Josh can you confirm?).

Why you are getting different "results" may be that the results are design results that are affected by the program's perceived design parameters such as unbraced length, etc.

 

[JoshPlumSE]

Different code check results are common. This is because of the program defaults for unbraced lengths. If you've got a 10 ft long beam, then default value for the unbraced length is 10 ft. If you've got 10 beams that are each 1ft long then the default value of the unbraced length will be 1 ft for each beam.

That being said, you should get the same basic results for member forces and stresses. There are probably oddball cases where splitting a member will affect the analysis results, but I cannot think of any offhand.

 

[Leadbelly]

I found it to be most affected when the lengths got very long and when the code check on a beam was above 1.-- or 2.-- that splitting a member reduced the code check.

Thank you both for the responses. I had a hunch it had to do with the unbraced length but I assumed that splitting a member and adding a joint would have the same bracing characteristics in RISA 3D. Guess not.