Gravity deformations

Sie finden etwas nicht oder wünschen sich Funktionalitäten? Hier können Sie es los werden! /
You are looking for functionalities? Go ahead!

Moderatoren: ccad, mz15, auroraIco, Lehrstuhl

Antworten
Marek
Newbie
Beiträge: 4
Registriert: Do 8. Dez 2016, 00:15

Gravity deformations

Beitrag von Marek »

Hello,
I wanted to know if Z88Aurora can calculate the gravitational deflection of an object under its own weight.
So far, the deflection was done by using the program PLOP:
http://www.davidlewistoronto.com/plop/
http://www.grattavetro.it/dimensionamen ... p/?lang=en
I did some simulations and comparative noticed that there is a very big difference between the results obtained - Z88Aurora max deviation is 11,9nm while PLOP 1,73nm.
http://www.fotosik.pl/zdjecie/usun/QaQ5 ... oWTB1w1BXr
Both objects have the same dimensional parameters.
In Z88Aurora position,, surface load "choose 20 because the weight of the mirror is approx. 2,1kg x gravity 9.81 = 20N ...... I do not know whether it is correctly entered data .....
If this can not be calculated in this way please advice which function to use the results were most similar in comparison to the calculations of deformations under the influence of gravity.
Regards.
Benutzeravatar
SHautsch
Alumni
Beiträge: 380
Registriert: Mo 15. Apr 2013, 11:03

Re: Gravity deformations

Beitrag von SHautsch »

Dear Marek,

Z88Aurora cannot calculate gravitational forces.

As you correctly supposed, using a surface load with the weight of the mirror converted to a force (2,1 kg * 9,81 m/s^2 = 20,60 N) can approximate the effect of gravitation on a body.
For this approach it is essential that the body is of equal thickness, as the weight distribution is directly depending on the thickness of the body. As long as the body has equal thickness, one can say that the weight distribution can be described by the surfaces of the finite elements. These surfaces are used when our algorithms distribute the surface load to the FE nodes.
If your mesh is fine enough to ensure good results and the elements are not too small to produce numerical problems with very small numbers (likely for applications like sensors, etc. where the body is smaller than 1 mm), the results should not be too different.
Of course your structure must be oriented perpendicular to the axis in which your force is acting, and this axis represents the direction of gravity.

It would be easier for us to find the problem if you sent us your project zipped via e-mail (or upload it somewhere and send us the link) to z88aurora-|at|-uni-bayreuth.de. Maybe it is a problem with boundary conditions or the mesh quality.
Unfortunately we don't know PLOP, but the website says it's an optimization tool - did you compare the Z88Aurora FE results with optimized results from PLOP? These should obviously show smaller deflections than the standard model calculated with Z88Aurora...

Kind regards,
SHautsch
Antworten