Hx (Rx) Elastic Ground Reaction

[kope]

Hi there. How can I check the Hx (Rx) Elastic Ground Reactions for a bar on elastic ground? (I found out the Ky and Kz in the diagrams for bars dialog but there's no Hx)

thanks. 

[Rafal.Gaweda]

I am afraid it is not possible.
Option is registered for implementation.

[kope]

Once the bar on elastic ground is transfered to the continous footing module, it seems that the calculations are done using only the Kz (vertical) related forces, is that correct? In this case, i cat set to none the Hx rotation on bar on the elastic ground dialog and use a regular linear suport with the same elastic stiffness. I think I'll get tjhe same results, right?

[kope]

seems that cannot be done if the bar elastic ground extends beyond the supported wall. The linear support may be applied only to the panel's edge...

[Rafal.Gaweda]

---

@Anonymous wrote:

Once the bar on elastic ground is transfered to the continous footing module, it seems that the calculations are done using only the Kz (vertical) related forces, is that correct? In this case, i cat set to none the Hx rotation on bar on the elastic ground dialog and use a regular linear suport with the same elastic stiffness. I think I'll get tjhe same results, right?

RC contitnus footing design moduel calculates element on imported internal forces so depending on elasticity defined in FEM model these forces may vary.

Then design is made basing on imported vertical forces and soil layer defined by user in footing module.

[kope]

Right. The internal forces should be compatible with the various boundary condiitions set for the FEM model. Then the beam
design is conducted as for any other beam, irrespective of its type (continuous footing or RC beam)

But, I was reffering to the soil pressure diagram for the continuous footing - my understanding is that only vertical forces are considered,
meaning no eccentricity along Y-direction and no out-of-plane bending (such as that imposed by some supported wall) are included.

(the pressure diagram looks uniform along y direction - i.e. transversally - no matter the rotation around its own centerline
due to any side loading).

Is this correct?

[Tuctas]

That's true Kope, unfortunately the diagram of soil pressure under the foot is always uniform... 😞

[kope]

The deflections calculated by the continuous footing module shouldn't be compatible with those provided by the FEM model assuming both are using the same KZ stiffness?

[Tuctas]

 I suppose they are the same...

 I was talking about the fact that Robot doesn't take into account column’s base moments about longitudinal axis of the footing and ignores torsional/rotation interactions between footing and soil (that would result in non-uniform soil stress distribution in the direction perpendicular to the axis of continuous footing) in the Continuous Footing Design module. If you are interested, take a look at an older post of mine: 

http://forums.autodesk.com/t5/Robot-Structural-Analysis/Spread-footing-vs-Continuous-footing/m-p/394...

[kope]

yes Tuctas, I was refering to the same issue - that the program disregards pressure modification due to Hx rotations.

The deflection is another story 🙂 Meanwhile I found out some other things that i don't understand. Let me summarize

all of them for other readers convenience; they are related more ore less to the elastic ground beam

1. The Hx rotations are disregarded when calculating the soil pressure
2. The deflection calculated by the module is not the same with the FEM model (see the attachement) 
3. Mx diagram is null in case you're assigning any value to the Hx stiffness (see attachement)
4. The question mark appearing nearby the element name within the RC Component inspector remains intact even if you're using the RC element update (such as for instance after adjusting the loading and running the analysis again). You have to add again the member in order to get rid of the question mark.

The displacement calculated by the FEM model - 2cm

The deflection calculated by the RC module -  6cm

Null Mx diagram along the bar in case that HX elastic stiffness is assigned

[Tuctas]

Good summarization Kope...

As concerns point 3), an answer could be: 

Generally, it is not possible to directly display reactions resulting from Hx rotation stiffness.

They can be analysed only indirectly checking the variability of MX torsion moment diagrams – but with some approximation because MX diagrams are displayed assuming linear distribution of reaction resulting from Hx (precise end values in nodes connected by straight line). It is necessary to divide beams in smaller parts to obtain higher precision of the display of MX in such case.

As concerns point 2), i think that is a quite serious issue (that i hadn't noticed untill now...) and i think that should be answered by the support...

[kope]

thanks for your reply. But... as i said - I was not able to get any Mx diagram for the beam once elastic ground is assigned to it.

(I'm using the "Results / Diagrams for bars / NTM tab / check Mx / Apply")

the Mx diagram plots just fine once elastic ground is removed and some other proper boundary conditions are attached.

[Pawel.Pulak]

It is the issue in 2013 and 2014 release related to HX elastic supports along bars:(

The workaround in 2013 and 2014 release is replacing HX elastic ground along bars by nodal elastic supports in rotation distributed along bars.

Regards,

[Tuctas]

  Thank you for the reply Pawel.

  And what about the issue with the deflections..?

[Pawel.Pulak]

In the current release deflections in continuous footings are calculated in the analogous way as in RC beams.

It meams that deflections from FEA are modified to consider reduced cracked stiffness resulting from reinforcement. It is done NOT considering the interaction with soil.

It may result in strong increase of deflections in relation to deflections in FEA - incoherent with applied loads and with the stiffness of soil

The workaround is to use the reduced stiffness of section in FEA "predicting " the loss of stiffness resulting from cracking etc.

See the screen capture below illustrating it for the model from kope:

Some ideas to improve the current behaviour of it are taken into consideration.

Regards,

[kope]

thanks for clarification. there goes my take on this issue.

assuming the displacement profile being correct for a given soil stiffness (simply put - the soil settlement has to be compatible with the applied pressure)

then the right choice would be to transfer the displacement shape instead of forces.

The impact of changing the beam stiffness on displacements is by far masked by the enforced compatibility between vertical load and soil settlement.

And anyway... it is user responsabililty to come up with resonable assumptions on beam stiffness during the modelling stage. I found

for instance quite convenient the possibility to adjust the stiffness on section definition, since there is great variance between codes on 

this matter.

Then, given the soil deformation and adequate boundary conditions (meaning the relevant DOF values at both ends),

the beam may be transfered to the provided reinforcement module.

The internal forces are then recalculated within the RC module since the stiffness degradation can now be accurately determined for a given displacement profile.

this has at least three benefits: 

- the rotations around beam centerline may be now properly accounted for, since displacement shape is the one being transferred

- accurate stiffness degradation may be calculated 

- consistent results to elastic ground definition

[HectorDuenas]

Mr. Kope, for the model you attached is correct the result that Mx is null.  That's because the beam twisted in the elastic media as a rigid body.  If you want torsion moments (Mx) in the beam you need for example to put transversal beams (as in a grid) with bending stiffness that make possible those torsion moments.

Hope this helps

Regards

Hector

[Μηχανικος]

Has this issue been resolved?