Inverted and drop beam deflection comparison

[PSR_1]

Hello!

Should there be any difference in the amount of deflection expected between drop beam and inverted beam. I have a finite element model of a floor where I am supposed to use an inverted beam. The deflection I get when I change the beam from drop to inverted is significant. It almost varies by half ( the inverted beam deflection being twice as much the drop beam). I can't make sense of it, please help.

The program I am using is RAM concept

 

[HTURKAK]

I do not have experience with RAM concept...Pls provide more info. for modelling of slab and beam..

Acc . to your explanation, the floor is modelled with finite element ( plate bending with membrane or just bending element ?)
How did you model the beam ? with frame elements ? If so, how did you provide that the beam is drop or inverted?

In case drop beam , it will be essentially a T beam and inertia will be almost double of inverted beam..

 

[KootK]

I also cannot speak directly to what RAMConcept is doing. I can think of a good reason to explain the difference in the deflections however, at least from a theoretical & computational perspective for a simple span beam.

1) For a dropped beam, in the cracked state, the presence of the slab enhances the compression block width and shifts the neutral axis upwards, effectively producing a deeper beam.

2) For an inverted beam, in the cracked state, the presence of the slab does not enhance the compression bock width and, therefore, does not shift the neutral axis upwards. This produces a beam that is effectively shallower than would be the case with #1.

 

[rapt]

I would expect more deflection for the inverted (upturn) beam, but not double.

But it would depend on a lot more information than we have been given. Continuity, level of cracking, etc.

 

[PSR_1]

HTURKAK:

They say the following about beams in ram concept "The beams are not one dimensional, linear framing members like they are idealized in other programs" so they aren't frame elements, rather they are plate elements with membrane just like the slab.

The program allows top surface elevation for slab and beam elements to be adjusted allowing slabs steps, inverted or drop beams modeling.

Rapt:

The beam is Simply supported( supported on the weak side of a structural wall which isn't absorbing much negative moment)

The level of cracking is well below the code allowable since the beam is PT. In order to control deflection a significant ( which won't affect the stressing stage state nor isn't over balanced) prestresing material is used.

General:

The beam is 5 times deeper than the slab (the shear stress between flange and web is given due consideration).

The floor level has significant dead and live load.

 

[HTURKAK]

If the beam is prestressed and the slab and the beam idealized with ( bending+ membrane ) shell elements, the delection for inverted case will be more but should not be double of dropped beam.. Check your material properties and model again..

 

[Celt83]

reply with some screenshots of:
- The Mesh input plan showing the beam location and the beam properties
- Design Strip layout in the direction parallel to the beam, design strip properties dialog with the trimming setting for the beam strip.
- Post-tension layout at the beam with the profile points visible

Are you doing one full width strip encompassing the beam and adjacent slab or did you break it up into a column and middle strip region? If a full width strip that can create some weird behavior with the PT since the total cross section centroid will tend to be down in the beam web so all of the slab tendons will be above the centroid in the dropped beam case and below the centroid in the upturned beam case.

As others have already there should be more deflection with an upturned beam. You may be getting some odd behavior from the axial tension/compression being generated by the change in depth at the beam. One thing to be careful of is the stiffness of the beams in general will tend to be overestimated because of how concept treats load transfer across and through the section, look at section 55 in their manual.

My Personal Open Source Structural Applications:

https://github.com/buddyd16/Structural-Engineering

Open Source Structural GitHub Group:

https://github.com/open-struct-engineer