Inconsistent results for deflection of RC slab

[Anonymous]

Hello all.

I am new to Autodesk Robot and this forum. I have been trying to learn the basics of Robot by modelling a simply supported slab and exploring all the different functionalities within Robot, and I have come across a "problem" I can't find a good explanation for. The issue is with regards to the deflection of the slab.

- What is the reason for the massive difference in deflections when I am in the "Slab - provided reinforcement" window using the "Plate and Shell Reinforcement" to display "deflection map" from "FE results" when I am comparing results from "For required reinforcement" and "For provided reinforcement after verification" when the two quantities of reinforcement are more or less equal?

- The same massive difference in deflections can be observed from the "Geometry" window using the "Panel Cuts" dialog box. Displacements in z-direction from the "Detailed tab" is much lower than Deflection from the "SLS" tab.

Also, how can I do calculations from the "Plate and Shell Reinforcement" dialog box from the "Slab - required reinforcement" window, inside the "Slab - provided reinforcement" window?

What is the difference between the calculation mentioned above and the one that is carried out by the "Calculations" dialog box in the "Slab - provided reinforcement" window?

Best regards, and thanks in advance for the help.

[Krzysztof_Wasik]

Hi @Anonymous 

Displacement presented in deflection maps is calculated for elastic material.

Deflection in Required reinforcement is calculated for cracked and reinforced slab section (theoretical reinforcement area is taken into account).

Deflection calculated in Provided reinforcement is calculated in similar way like in required reinforcement but takes into account real reinforcement (reinforcement bars distributed in slab, their area is usually bigger that theoretical reinforcement)

Please refer to webinar Reinforcement calculation

[Anonymous]

Thank you for the answer.

I was thinking the explanation was something along those lines, but I still have a hard time believing that is the only thing going on because of the vast difference in deflection between "Deflection calculated in Provided reinforcement" and "Deflection calculated in Required reinforcement" when the amount of reinforcement that is provided is not that much more than what is required (in my example)...

It seems to me I am missing something else.

[Krzysztof_Wasik]

Hi @Anonymous 

When I compare results for deflection (saved in your model), deflection in Provided reinforcement (max 128)

is similar like in required reinforcement

So please explain what you mean by vast difference between deflection in those modules

[Anonymous]

Hi Krzysztof,

Attached you will find the two results that baffel me.

providedrequired

Best regards.

[Krzysztof_Wasik]

Hi @Anonymous 

You have probably verified deflection with "Stiffness update FE" method.

Then you should compare it to similar verification type in required reinforcement module.

In that case in both provided and required reinforcement you should determine SLS QPR combination for deflection verification.

I supose that you did not chosen QPR load combination.

[Anonymous]

Hi Krzyzstof,

To be honest with you, I don't really know which limit states are considered for the different deflections. I have yet to familiarize myself enough with Robot that I feel I have total control over what is going with all the different calculations/results etc. I hope to get there eventually.

But it seems to me both results are comparing apples to apples, ie. same method for evaluation of stiffness and same limit state load case.

Best regards.

[Krzysztof_Wasik]

Hi @Anonymous 

Simpified method calculates deflection envelope (for all QPR combinations) but uses simplified (rough) stiffness estimation method. This method does not need combination selection while verification. Setting is taken from

calculation stage

Stiffness update FE method needs QPR SLS  combination selection, because in first step FE stiffness for each finite element is determined, then stiffness matrix is updated, then static analysis for selected SLS QPR Combination and for modified stiffness matrix is calculated.

For EN based codes QPR combination selection is required for deflection verification.

So if you compare results for Simplified method with  Stiffness update FE method, you should (for FE update method) choose QPR combination which gives  maximal elastic deflection in simple static analysis.

In your case (one load case only) it does not matter which QPR combination will be chosen 

[Anonymous]

Hi Krzysztof,

I think I understand where you are going, and it answers why simplified method and FE method yields the same answer in my case.

What I don't understand is that within the same stiffness method, the required reinforcement results in a deflection far greater than the deflection caused by provided reinforcement. To my eyes, these two amouts of reinforcement is not that different, yet they yield very different results.

Is this because of some error in the model, is it because I use the program wrong, is it because of an error in Robots engine, or is it simply that my expectations/understanding of the significance of the extra reinforcement (prov - req)  that is off?

Best regards.

[Krzysztof_Wasik · Accepted solution]

Hi @Anonymous 

If you compare simplified calculation to FE stiffness update you compare two different methods (historical simplified to more detailed FE stiffness update)

If you do not choose QPR combination for FE stiffness update method results are incorrect.

Comparing results for FE stiffnes update I get the following

required reinforcement deflection for SLS QPR1 - I get 138 mm deflection with bottom cracking 0,26-0,37 mm

provided reinforcement deflection for SLS QPR1 - I get 111 mm deflection with bottom cracking 0,22-0,28 mm (due to crack reduction by redundant provided reinforcement

Almost all bottom part of slab is cracked , crack width reduction matters in deflection calculation (FE stiffness update) so slab is very sensitive for provided reinforcement area.

When I modify reinforcement reducing redundant reinforcement deflection and crack width increases

Example

In my opinion it seems to be reasonable software behavior.

[Anonymous]

Thank you. Your example made it clear to me where I got things wrong.

---

If you do not choose QPR combination for FE stiffness update method results are incorrect.

---

Is there a reason selection of none-QPR SLS combinations is available? If it is bound to give incorrect results, I reckon removing that option would be the better course.

Thanks again.