• From: Portugal
  • Registered: 2009-07-09
  • Posts: 10

Topic: Non uniform distributed loads in ATENA engineer sofware.

Hi prof. Cervenka and Ing. Pryl! How are you? Hope great...
It passed already several months that I was in your company in the ATENA course of last summer! Great course! Nice city! Wonderful people!
For all the users of ATENA registered in this forum that want and need to know more about this software I really recommend that summer course!

Well, now I'm writting this message because during the past week I tried to simulate a non uniform load (triangular and trapezoidal loads) in order to simulate an hydraulic load in the ATENA engineer software and I didn´t have success. Another thing is that if I want to modulate the behaviour of the fullfillment of a water tank I need to divide several macroelements that combines with the history load process. In another words, I cannot assign a distributed load in a partial surface of the macroelement. I guess this could be usefull during the construction of the history load.
Can you help me on this?

Best wishes,
Biscaia

Biscaia's Website

Quote Post 1

  • dpryl
  • CC software developer
  • Offline
  • Registered: 2007-07-13
  • Posts: 1,107

Re: Non uniform distributed loads in ATENA engineer sofware.

Dear Hugo,
I am afraid there is so far no easier way for this in ATENA 3D Engineering.

In GiD (ATENA Sci), you can assign a load with linear dependence on the coordinates, e.g, -Z*100+50 or similar.
However, the load is also always applied to the complete surface. To avoid making too many volumes (macroelements) just for the load history, you can assign the load directly to selected finite element surfaces after meshing (please do not forget to re-assign it when you re-mesh).
If you do not have any license for the GiD pre- and postprocessor (btw., the university license is not expensive), you can test it in the demo version which is included in ATENA installation. If the model size limitations of the Demo do not allow to handle you model, you can register it free of charge  for 30 days (kind of "try before you buy").

Regards,
Dobromil

dpryl's Website

Quote Post 2

  • From: Portugal
  • Registered: 2009-07-09
  • Posts: 10

Re: Non uniform distributed loads in ATENA engineer sofware.

Dear Pryl,

Despite I suspected that in GID will work perfectly, I confess that I had some expectations this could be done in the ATENA 3D Eng also. The reason is simple: I don't have the GID license and a demo version has limitations (in time and number of FE) as you know...

Well, I did some simplifications: divided a volume in several volumes and applied uniform load with the mean value between the maximum and minimum loads applied in macroelement. I know it is a coarse manner of modulate a hydraulic load... but is better then nothing...

Another thing Pryl:
If I want to monitor the applied uniform load, how can I monitor that load? Well, I know that each step will correspond to a % of the total load but I wanted to see it in the running window chart.

Best regards,
Biscaia

Biscaia's Website

Quote Post 3

  • dpryl
  • CC software developer
  • Offline
  • Registered: 2007-07-13
  • Posts: 1,107

Re: Non uniform distributed loads in ATENA engineer sofware.

There is one more option - save the .inp file from ATENA 3D, add the load manually there (see the ATENA Input File Format manual, 4.4 Load and Boundary Conditions Definition, esp. tables 100 and 101 and around - you need a BOUNDARY_ELEMENT_LOAD with a function).

You would define a constant load for the surfaces, and apply it in a separate load case. Then, you replace the load by the hydro-statical pressure in the .inp (of course, send me the model+load formula if you need help), and run the modified .inp by File-Open other-Analysis control file.

However, even this way, you have to have the loaded surfaces divided in such a way that every analyzed water level reaches a line between two surfaces.


Regarding the monitoring:
Once you have to schtch to Arc-Length (i.e., if you need to get to the peak or behind it using force loading), you definitely need to monitor the load to know how much is applied (the A-L method changes the load and the "load in one step times number of steps" formula does NOT work any more!).

There is a monitor for "% loads" automatically available in the ATENA-GiD interface, but I am afraid it is not yet available in ATENA 3D.
You can select a single node on the loaded surface for monitoring, and multiply it by the corresponding factor to get the total load.
Or, you may add a small force (e.g., 100x smaller than the real loads applied) in another node, and monitor this force - this way it is easier to find out the multiplication factor for monitoring.

Dobromil

dpryl's Website

Quote Post 4

  • From: Portugal
  • Registered: 2009-07-09
  • Posts: 10

Re: Non uniform distributed loads in ATENA engineer sofware.

Another great tip, Pryl!
Thank you!

Biscaia's Website

Quote Post 5