Good morning.
I would like to know why "Centroid Deflection", "Element End Deflection" and "Element Angular Rotation" shown in Natural Frequency Calculation Report are different from those indicated in Wind Deflection?.

Also, the "Natural Frequency Calculation Deflection" do not consider the allowable deflection indicated in the PVElite Configuration Window; only wind deflection is compared with it. Example: in a 6 meter ID x 100 meter lenght tower, first one is around 2000 mm. and wind ones are 150 mm.
Thanks.
Carlos.

Hello Carlos

A vertical vessel is basically a cantilever subject to lateral loads (wind etc.). If we imagine that we are able to push the tower by applying a horizontal force at the top, we can appreciate that the top of the tower will deflect horizontally. However, the whole of the tower bends, thus, not only the top deflects, but all the other parts of the tower do too. Now, the top deflects the most. If we not consider the centroid (location of the centre of gravity), we need to know its deflection also. Let us consider again the top of the tower. As it deflects, it also rotates, because the central axis of the vessel is bent (called the elastic line). If a spirit level were set on the top of the vessel, the bubble would move. This is what is meant by the term 'rotation'. If you are familiar with strength of materials terms it would be the dy/dx term in beam theory.

The wind deflects the tower in the direction of the wind direction (vector).

There is another effect from the wind. The wind, in passing around the tower forms little Eddie currents either side of the tower. These eddies momentarily change the pressure either side. This occurs rhythmically. This induces the tower to vibrate at right angles to the direction of the wind. If the shedding of these eddies (known as vortex shedding) occur at or near the natural frequency of vibration of the tower, the oscillations can be quite large. This causes axial stresses to be set up in the tower. We have to consider the effects of the static deflection (bending in the wind), and dynamic deflection (from vortex shedding). The greater of the two would govern the stresses to be considered in the shell.

Thanks Ray, but maybe I had not been able to explain my problem correctly. I understand what are the terms that I mentioned in my topic and no vortex shedding is considered in the example that I am referring.
Coming back to that example, the Centroid Deflection of the "Natural frequency for the operating case (Freese)" of the uppermost node is 2346mm. while the centroid deflection (for the same node) in the "wind deflection calculation for operating case" is 152mm.
Then PVElite compares 152mm. with 503mm. (allowable deflection) and says its acceptable.
What I want to know is:
where does the 2346mm. deflection come from?.

Carlos, could you please send me your file (the .pvi file). My e-mail address is rdelaforce-at-coade.com. I would like to look at the results myself. The '-at-' should be replaced by the @ symbol. I have put the strange address to prevent spam at COADE.

Buenos Dias Carlos,

To understand this it is necessary to be familiar with how the Natural Frequency of a free standing base supported structure is calculated. There is a paper by C.E. Freese that explains this. Basically, the vessel is horizontally cantilevered and the deflections and rotations at various points are computed. Once these values are known along with the material properties, a numerical integration scheme can be used to compute the fundamental period of vibration.

The wind deflection is computed based on the similar properties but the force on the vessel due to the applied wind pressure is also factored in. The results for these calculations are separate and distinct.

In the later versions of PVElite, the natural frequency is determined by the solution of a mass/stiffness matrix. The overcomes the restrictions of the Freese method. However, the Freese method can be invoked if required.

Dear
Is there any change in methodology for calculating natural frequency of vertical vessel..
as for Pv elite-2008 natural freq = 8.20977 while same model in -2009 version natural freq = 8.27330

Can you explain me why this change is happened

Hello dhaval

We often get questions as to why a particular model runs differently in two different versions of PV Elite. When the files are sent to us, we usually find the models are different in each version. Maybe only slightly, but they are different.

We use two methods in PV Elite. They are the eigensolver method, and the Freese method. If the eigensolver is switched off, the Freese method is used.

Are you absolutely sure the models are identical?