Topic: modeling temperature changes

Hallo,

I would like to model a rehabilitation concrete layer over an old deteriorated asphalt layer of a highway. In winter (in the morning) the top surface of the concrete layer will be colder than the bottom surface. The deviation of the temperatures of the top and bottom surfaces of the concrete layer from a benchmark temperature (the temperature at casting the concrete) will be -38 C and -32 C respectively. this temperature deviation across the concrete layer thickness can be considered as composed of constant part (-32 C) and linear part (-6 C).

my question is how to model this temperature change across the thickness of the concrete layer?.
Is it possible by Atena Engineering? or should I use Atena Science?

after modeling the temperature change, I would like to subject the concrete layer to traffic loads. How can I combine the effect of the temperature change and the traffic loads together?

best regards

Mohammed

Re: modeling temperature changes

Dear Mohammed, in ATENA Engineering it is possible to apply temperature as load, however, it is only possible to define a constant temperature increase/decrease for each macroelement. If you have a simple-enough geometry, you might be able to divide your asphalt and cocnrete layers each into several macroelements and assign the corresponding temperature to each macroelements.

In ATENA Science, you can define linear temperature gradients. If this is not enough, you can run thermal analysis first and then import the resulting field(s) into static analysis (see the ATENA Science Example Manual, chapter "3. TRANSPORT ANALYSIS").

In any case, remember the loading is incremental in ATENA. This means you can first apply the temperature gradient and then the traffic forces, and this results in both of them acting on the structure.

Re: modeling temperature changes

Dear Sir,

Using GiD I can define the loads increments using the intervals. my question, how can I do that using ATENA Engineering?

best regards
Mohammed

Re: modeling temperature changes

Hello, you first define a few load steps applying the load cases with supports+thermal load. Then, define the steps with the supports+the traffic load.

Re: modeling temperature changes

thank you very much

on the other hand, In ATENA Engineering one can first choose if the macroelement is a plate/shell element or standard. How can I do that in GiD?

best regards

Re: modeling temperature changes

Please read section "5.3.1 Shell Material" of the ATENA-GiD User's Manual.

Re: modeling temperature changes

Hallo Sir,

I would like to use the second option (e.g., to use a linear teperature gradient through the concrete layer thickness), but how can I do that. In GiD - Conditions - Temperature für volume - Linear. there is only three coefficients.
Coeff X
Coeff Y
Coeff Z

where can I enter the temperature gradient?

best regards
Mohammed

Re: modeling temperature changes

Hello, the value from the "Constant" tab is also used (to define the value at coordinate origin). The 3 coefficients define the dependence of the temperature on the 3 coordinate directions. The exact formula is specified in the help text.

Re: modeling temperature changes

Dear sir,

I would like to get the values of the bending moment and axial force due to different loading on this new concrete layer. Could you please tell me how to do that.

best regards
Mohammed

Re: modeling temperature changes

Dear Mohammed, direct support for these is only only implemented in the ATENA Engineering 2D GUE (Moment lines, MNQ). The main reason for this situation is that these values are typically not of much interest within nonlinear analysis (unlike linear design calculations).

If you need these quantities (please explain - maybe there is some other way to get the information of your interest), you need to integrate the corresponding internal forces outside of ATENA, e.g., in a spreadsheet.

Re: modeling temperature changes

thank you

I have applied the load combination at the serviceability limit state and monitored the crack width. I want to compare the value of the crack width from ATENA with the hand calculation, therefore I need the normal tensile force and the bending moment values.

best regards
Mohammed

Re: modeling temperature changes

Dear Mohammed, please do not forget the formulas for hand calculations most likely (99%) have been developed for elastic models, and would not work correctly when applied to the forces and moments from the nonlinear analysis (cracked structure). Therefore, to get proper input for the formulas, I would recommend to run a separate elastic-only calculation with the same loading, or simply calculate the elastic deformations analytically, too.

Re: modeling temperature changes

Dear Dobromil,

thank you.
I bear in mind this issue. I want to have an approximate comparison.
I would like also to design the concrete layer at the ultimate limit state? Could you please help in this regards?

best regards
Mohammed

Re: modeling temperature changes

Dear Mohammed, the most common approach to ULS using nonlinear analysis is applying the design load (multiplied by a security factor) to the model (instead of checking the stresses, moments, etc. as usual for linear models). Frequently, one simply applies the nominal load level and then increases the load up to the peak and the ratio of the max. calculated load to the design nominal load is then compared to the required security factor.
If there are multiple load combinations suspected of being "the worst", the above has to be done for each of them and the worst of them decides.

Re: modeling temperature changes

Dear Dobromil,

I am analysing using the GiD software. The programs (GiD and ATENA Science) have crashed after running for two days, how can I continue the analysis from the last saved analysis step?

best regards
Mohammed

Re: modeling temperature changes

Please see the tab "Restart Calculation from Calculated Step" in teh Problem Data dialog.

Re: modeling temperature changes

Dear Dobromil,

I have checked the formula used to calculate the crack width by hand. It is based on the analysis in the cracked state (i.e. nonlinear analysis).
It is actually not my choice. I have been asked to perform the hand calculation. I ''must'' do it this way.
I am sorry, but the method you have suggested earlier is a little bit tedious. Is there any simpler way to do that?

best regards
Mohammed

Re: modeling temperature changes

Unusal tasks frequently require tedious work (automation is concentrated to the frequent ones). Formulas based on cracked state AND considering moments (except for the design ones simply ignoring all concrete in tension) are not common. Maybe if you post what exactly you plan to calculate, we can find some easier way, but I suspect you need the spreadsheet.

Re: modeling temperature changes

Hello Dobromil,

I am now analysing a strip of the street layers under serviceability limit state in order to check the maximum crack width that results. The strip has dimensions of 50cm*900cm.
I have choosen different mesh sizes (25 cm, 12.5 cm, 5 cm and 1 cm).
It seems that the result depends on the mesh size in a nonsense manner.

for the 25 cm mesh size, the max. crack width is 0.16 mm
for the 15 cm mesh size, the max. crack width is 0.15 mm
for the 5 cm mesh size, the max. crack width is 0.07 mm
for the 1 cm mesh size, the max. crack width is 0.001 mm

Could you please help in identifying the problem?

best regards
Mohammed

Re: modeling temperature changes

Dear Mohammed, withou seeing your model it is impossible to guess what is going on. Before anything else, I recommend to check if the meshes being compared look reasonable, above all, if the edge aspect ratio (i.e., length of the longest edge to the length of the shortest edge of an element) is under about 3:1 for volume elements.

Re: modeling temperature changes

Dear Dobromil,

thank you

I will send you the model per Email, Could you please have a look at it.

best regards
Mohammed

Re: modeling temperature changes

Dear dobromil,

You told me:
If you need these quantities (---), you need to integrate the corresponding internal forces outside of ATENA, e.g., in a spreadsheet.

I have defined monitoring points at the relevant mesh nodes in order to get the values of the forces there to calculate the bending moments. But I have got only ''---'' instead of the values. Could you please help?

regards

Re: modeling temperature changes

Dear Mohammed, I don't know about a way to define monitor for INTERNAL forces in the ATENA-GiD user interface - how have you introduced it?

To output the internal forces for later integration in a spreadsheet, you do not need any monitors. Just output the values through the Output Data dialog of AtenaWin. In your case, it is very easy to filter the internal points you are interested in - I see 2 ways:
A. output the value for all integration points (IPs), output the reference nodel coordinates for all IPs, filter in spreadsheet based on the X (or Y or Z) reference coordinate identifying your cut plane of interest.
or
B. select the IPs in AtenaWin and export the internal forces only for the selected IPs.

Re: modeling temperature changes

Thank you
In the Output Data dialog of AtenaWin, it is possible to output the internal forces for nodes only, not for IPs also?!!

Re: modeling temperature changes

Have you already checked the AtenaWin Manual?
Please remember stresses and other integration point values need to be extrapolated to nodes - similarly to displacements and other nodal values, which are interpolated from nodes into IPs.