Composite beam in STAAD

[Bill2001]

Hi,
I am trying to model a composite beam (steel beam + concrete slab).
For this purpose I modeled my composite beam as a steel beam and plate elements.
I applied then an offset distance 0.95 m to the steel beams.
Now I see that the steel beam gets a very large axial force and the Mz moment at the mid span is very small!!!
I am really confused the way STAAD calculates section forces.
Do you have any clarification on this?
Do you have any sample for a composite beam used in STAAD (built-in Composite Deck)? Could this feature used on user defined beams such TAPERED?
I have uploaded my input file.

Thank you
Bill

 

[SteveGregory]

Instead of modeling a beam with plate elements, check the help file index for Composite Beams.

 

[Bill2001]

Thanks for your hint, but what do you think about the section force computed by STAAD?

I mean small bending moment and large axial force.
Unfortunately, I don't have any other software to verify the results.
Suppose this is a bridge deck, what shall we do if the section forces are not correct.
Steel beams fail when I apply a check code due to large axial forces.
It doesn't make sense at all.
Does STAAD calculate the section forces correctly?
Do you know how I can get arround this situation?
Thanks
Bill

 

[SteveGregory]

It sounds like you may be modeling a floor with plate elements and these elements are loading the beams up with axial loads. Instead of plates, consider using the master-slave spec to model a rigid diaphragm.

 

[Bill2001]

Thanks for your reply.
I did as you suggested.
I still get axial force.
What do you think?
I have attached my modified model.
Bill

 

[SteveGregory]

The axial stress is the result from the negative temperature load. Your model has a concrete beam floating in space that should be deleted and a steel tapered beam. The composite beam design works with standard shapes not tapered beams.

The slave master commands only apply to a floor diaphragm.

 

[Bill2001]

Thanks SteveGregory,
I don't know how I can make a rigid floor diaphragm.
Just wonder if you mind helping me on this?
Again thanks a lot.
Bill

 

[SteveGregory]

Do you want to design a single beam or a whole floor of beams? The only purpose of modeling a rigid diaphragm is to distribute lateral loads for a whole floor to your lateral load resisting system.

If this is a single beam, I suggest hand calculations.

 

[Bill2001]

Thanks again SteveGregory,
Do you know how I can get rid of the axial force in the beam?
I have used the EXPLICIT definition method in STAAD to create a composite beam (CM.std)
The steel beam and concrete slab act monolithically and as expected for load case 1, max Mz=5000 kNm, but it doesn’t work for the temperature load (Shrinkage).

How can I get around this problem so that I get Mz=5000 kNm in my previous model?
Bill

 

[SteveGregory]

Call tech support

 

[edward1]

I have used STAADPRO to model composite steel girders with concrete deck bridges and I have found best results are with slab modelled as elements and girders as tapered steel beams. I have used concrete beams to connect the centroid of deck to the centroid of the girder with dimensions of this beam to be the width of top flange of girder and depth equal space of element.

 

[Bill2001]

Hi edward1,
Thanks for your reply.
I am glad to hear that you have successfully modeled a composite beam in STAAD. I was about to give it up and try other software for this problem.
Have you applied MEMEBER OFFSET command to steel girders?
I have done the same, but the max mid span moment is not correct due to huge axial force in the girders!
Could you please if possible upload your model so that I can see what you have done to overcome this problem?

Thanks
Bill

 

[edward1]

Bill,
The model you have is correct. If you look at the bottom flange stresses in the girder they will be close to the bottom flange stresses of a composite beam design such as Digital Canal's Composite Beam Design. You probably need this type program which calculates more details about the composite beam design than STAADPRO.
Ed

 

[Bill2001]

edward1,
Thanks for your help indeed.
I must try this software.
Bill

 

[edward1]

Bill,
I stumbled on the way to keep axial forces low using MEMBER OFFSET to model composite beams. You need to model the slab in the XY plane with the beams offset using LOCAL command. Be sure to use the correct BETA angle for the right beam orientation. All gravity loads will be in the Z direction.
Edward

 

[edward1]

I have included a model of a curved girder composite steel bridge that was also run on MDX with very close results. Notice the MEMBER OFFSETS and BETA angle used. This bridge was constructed in 2006 and was fitted with instruments to get natural frequencies and stresses from test trucks.

 

[Bill2001]

edward1
Thanks for your reply and please fogive me if I am late in responding you.
I am not sure if one can specify the member offset distance in conjunction with the beta command. Please see attached pictutre.

LOCAL means that the distances f1, f2, f3 are in the same member coordinate system that would result if the member were not offset and BETA = 0.0
What that means is that when you specify a LOCAL offset, the program is applying those offset values to the unrotated position of the member (BETA=0 position).
What do you think?
Bill

 

[edward1]

Bill, If you model elements in Global XY plane, then use BETA 90 and LOCAL OFFSETS START and END 0 -# 0. That works correctly for me. If you click on SYMBOLS and LABELS, then STRUCTURE, and FULL SECTION you will see the orientation and offset of the beam.
Ed

 

[Bill2001]

edward1
Thanks for your help!
I did the same to my model. All axial forces are gone now. Could you please take a look at my modified model see if you find any irregularities in my modified model.
Please not that I used PERFORM ROTATION X 90 to rotate my original model, then I specified the offset distance as shown below:

MEMBER OFFSET
1 TO 20 START LOCAL 0 -0.7 0
1 TO 20 END LOCAL 0 -0.7 0
The BETA is specified as 90 (BETA 90)
Thanks again.
Bill

 

[edward1]

Bill,
Just change the support at joint 105 to 21.
Ed