FEM Flat Slab, hogging steel over columns

[sybie99]

Hi Guys,

I am using a FEM package to analyse a flat slab (shell). Previous packages I have used allowed me to what was called "Peak Smoothing" where the peak moments at supports are reduced. The current package I use does not have this option. I am having trouble interpreting the results as not to put in too much reinforcment over columns for negative moments.

If you look at the attached pdf it shows the bending moments at a column support. The mesh is 0.25m size. The bending is given as kNm/m. So at the centre I have 80kNm/m and just 250mm away the bending is 50kNm/m. As these moments are in kNm/m, can I not use the average moment. Say its 80kNm/m at the centre of the support and 40kNm/m at a point 0.5m either side of the support, would it be okay to use a moment osf 60kN/m when designing my reinforcement over the support for that 1m wide strip?

 

[sybie99]

Not sure file was attached, have added it to ENGINEERING.com

 

[ishvaaag]

The traditional ways of designing flexural reinforcement over supports address it by ensuring that an important part of the reinforcement is directly over the support centerline band, diminishing from there. Some programs (CYPECAD is one of them) closely follow the moment requirement over supports. Others may not follow the requirement as stringently, but it is clear from the traditional practice that averaging over wide zones was not seen as sound practice, since they were approximately following the moment requirement.

The practice of averaging may have some utility at the limit states since whilst failing the tensile forces (if transversely transmissible) will spread through shear lag to the more outwards main reinforcement over support zones. However when averaging it is important to ascertain the more accurately the better that at the service loads actually expected and with the actual reinforcement provided nor crack nor deflection problems are of relevance.

Upon such constraints, where the code permits it, averaging could be entirely justifiable; if the averaging is moderate (differentiating maybe a narrow band over supports) and the safety coefficients are generous, averaging shouldn't result in any particular problem.

 

[sybie99]

Thanks Ishvaaag, that clarifies it a bit, hence the requirement in the British codes to provide 2/3rds of the column strip rebar over a width of half the column strip centred over the columns. Thanks

 

[rowingengineer]

sybie,
You can distribute the reinforcement as per you post above in to the column and middle strips with averages, ensuring that you have enough reo for punching shear in the column zone.

Also you did a print screen with a dual monitor possibly? on your second monitor you have you emails open, maybe you should crop the pdf in the above post to just show column contours.

http://www.nceng.com.au/

"Programming today is a race between software engineers striving to build bigger and better idiot-proof programs, and the Universe trying to produce bigger and better idiots. So far, the Universe is winning."