Cladding Load Distribution - revisited

[Max-88]

Hi,

Re. distribution on loads on cladding elements - see attached model and screen captures.

The load distribution appear to be different from a straight manual calculation for member 39.

1. Load of 0.9kPa on cladding, orientation local axis z, distributed to adjacent beams at 9m c/c would indicate a udl of 8.1kN/m, orientation local axis z of beam. Robot indicate a udl of 8.2kN/m on global z and -0.44kN/m on global y. That adds up to 8.21kN/m, orientation local z of beam.

2. The distribution of surface load on roof cladding to adjacent beam 39 is not accurate when superimposing the many generated loads.

Can anyone shed light why there are so many generated loads, some superimposed, from one load on cladding. 

3. The symbol of load on cladding appear to be parallel to global z rather than local z. Load label/text is correct. - No screen capture on that, limit of 3 attached files.

4. I would like to avoid splitting the beam at locations where load changes as per previous advice from Stefano, to keep the model clean

I am rather fond of the availability of cladding element and load distribution as it considerably cuts down load modelling time, especially with wind loads to AS/NZS.

cheers

max

[Artur.Kosakowski · Accepted solution]

Hi Max,

1. Load of 0.9kPa on cladding, orientation local axis z, distributed to adjacent beams at 9m c/c would indicate a udl of 8.1kN/m, orientation local axis z of beam. Robot indicate a udl of 8.2kN/m on global z and -0.44kN/m on global y. That adds up to 8.21kN/m, orientation local z of beam.

You are right but at the part when there is a change in the cladding load value the load applied to the bar is slightly smaller. The total value of load should be the same for bar load and cladding load however due to way the distribution algorithm "assigns' load to supporting object it may happen that the distribution is not as you would expect it to be. We are aware of this fact and the improvement of this process is on the wish list for the next versions of RSA. 

2. The distribution of surface load on roof cladding to adjacent beam 39 is not accurate when superimposing the many generated loads.

Can anyone shed light why there are so many generated loads, some superimposed, from one load on cladding. 

Again this is the consequence of the implemented approach for detection of supporting objects and tributary areas discretization. 

3. The symbol of load on cladding appear to be parallel to global z rather than local z. Load label/text is correct. - No screen capture on that, limit of 3 attached files.

This is the standard way such load type are displayed in Robot. The correct orientation for laod applied to a (meshed) panel would be visible after you display loads generated automatically which for cladding is shown as loads on bars instead. See post 8 from http://forums.autodesk.com/t5/robot-structural-analysis/uniform-planar-load-on-contour/m-p/3030154/

4. I would like to avoid splitting the beam at locations where load changes as per previous advice from Stefano, to keep the model clean

You may try to generate more nodes along a bar without dividing it.

[Max-88]

Thanks Artur.

cheers

max