dimensionnement des barres acier

[amirdj.gc]

Bonjours,

j'ai fait une comparaison entre deux modeles comme le montre l'image ci dessous, dans le premier modele la poutre principale en IPE200 a été definie comme une poutre continue, dans le deuxieme modele la meme poutre a été dévisée aux intersections avec les solives.

quand je fait une verication des pieces dans le premier modele (poutre continue) les profiles IPE 200 sont incorrecte alors que dans le deuxieme modele les meme poutres (Divisées) sont verifiées est correctes sachant que les meme charges sont appliquées dans les deux modeles,

je veut savoir quel est le resultat le plus exacte et la modelisation la plus correcte.

vous trouvez ci joint le fichier. 

Merci.

[Artur.Kosakowski]

Try to look at the model as if you are about to make hand calculations of the beam. I assume that you would see the beam as being a single element between two columns. If so instead of running code checking for 4 small beams you should create a new one is built from them as shown below:

Such created bar 31 is just the same as bar 8 and this is the recommended approach for steel code checking of beams.

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.

[amirdj.gc]

thank you for responding, but i really didnt understand the diffrence between the two models can tou give me an explanation of this different results?

and as i understood you recomand me to modelise the beam as one element?

[Artur.Kosakowski]

From the static calculation point the models are the same but for steel code checking you assigned the same parameters to different bars. E.g. you defined buckling length as being equal to the length of a bar. If the bar is 4 times shorter than the other this makes difference for e.g. capacity against compression.

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.

[mohamedx61]

Hi every one i would add some thing.in other words,in RSA to be sure what i have understood that ; to verify the beams or colum in the view of the code steal no mater what the code is ( for e.g CM66 or EC3 ...)we shoud define them as one ? to be sure that the verification of all elements or barres are correct.thank you

[Artur.Kosakowski]

This is no YES or NO situation. You have top match the model with the behavior of the element in the structure and the way you want to design it. The 3 above frame will give you the same values of internal forces yet if the same design parameters are assigned to them will have different ratios for code checking. Imagine that you assumed unrestrained lateral length as equal to element length (default for the beam bar type). It is 4 time larger for bar 4 than for bars 8 and 9 and 6 times larger than for bars 13to18. The same if you set the deflection limit as e.g. L/250 where L is the length of the element as defined.

For this particular situation the best way of modeling (so that it is easiest to define 'correct' design parameters) is to use the 2nd approach yet if you create two superbars as I demonstrated previously (form chains 13 to 15 and 16 to 18) you will have exactly the same models for the 2nd and 3rd frame.

For the first frame it would also be possible to assign the correct parameters for the strength resistance check (you can define the intermediate bracings) but the check of the deflection would be not correct anyway.

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.