Displacement in RSA is too low

[ksibandaA6TWA]

Hello, 

The anticipated displacements when I run my model are shockingly low in my model for the loads that i have applied and I can't seem to figure out why that is the case. These displacement values are for linear checks only prior to verification of the slab and need to be a lot higher than what is shown. The issue is not immediately obvious to myself and i would appreciate any insights that anyone might have to resolve this issue. 

For Facade loading of 15kN/m  i'd expect a deflection of circa 6-7mm at the tip of a 250mm thick assumed to be taking the load over 1000mm width  but given that the load is actually applied only at the edges and the model is only returning a value of 1mm, something is definitely not correct here:

Kind regards, 

[ksibandaA6TWA]

See file attached below

[EduardoMeruvia]

Hi @ksibandaA6TWA , 

You mentioned that your facade loading is meant to be 15 kN/m. As far as I can see your model only has 6.3kN/m as the facade load. However, even if you increase your load to 15 kN/m you'll be still getting about 3mm of deflection. 

I don't see anything wrong with the deflection results in your model as far as I can see.

For your hand calcs, did you assumed a simply supported beam of a cross-section size 250 x 1000 spanning between the supports (6.8m span)? If you did that, then that is probably why you getting more deflections on your hand calcs compared to your model.  Your slab on that edge its actually NOT just simply supported but acting as a 2-way slab with a semi-cantilever edge. Hence, in reality you have more stiffness aiding the deflection of that edge than assuming a simply supported edge beam condition. 

However, all that aside, if you are to consider the long term deflection of you slab and reduce your concrete gross moment of inertia (Ig) by a calculated "cracked" moment of inertia (Icr = aprox 0.4Ig for slabs), then this would give you a better long term service deflection estimate (total deflection) of your concrete slab under the loads in considerations. This was approximately 7.05mm (see image below, reduing Ig by 0.4) when re-running your model with the updated 15kN/m load. 

[Simau]

@ksibandaA6TWA 

Displacements with the model you sent

[ksibandaA6TWA]

Thank you for taking a look at my model, really appreciate that. Apologies for sending a superseded model, the 6.3kN/m was for another floor. 

Your explanation was logical and makes sense for the portion of the slab I had used as an example. However for the other loadcases, such as the imposed load of 3kN/m2. Surely i should be anticipating much larger displacements internally than the ones currently being calculated by the model, especially for some of the larger spans? The values are typically 1mm or less and that doesn't feel correct straight off the bat. I would be expecting something above 5mm for this loadcase alone prior to looking at SLS combination cases or trying to verify the reinforcement. 

[EduardoMeruvia · Accepted solution]

@ksibandaA6TWA, main reason you are getting "smaller" deflection on your current model is due to the 2-way action of your slab and the use of the GROSS moment of inertia (Ig) adopted on your slab panels. If you were to do a hand calculation on a flat rectangular plate with all edges fixed using typical "Roark Formulas" using also the "Ig" you most likely end up with similar deflection results. These deflection will always be smaller compared to a simplify 'simply supported' assumption. 

However, that deflection check only accounts for the immediate elastic deformation, and in concrete the long term deflection is more critical and should take focus.  You could follow the estimate on long term deflections as per my previous post, which can be use as a initial check. If you want a more through check, you can use Robot's concrete code check and stiffness update (inelastic) method.