Hello,

I have an harmonic study to perform: turbine alimentation with points to solicited on X, Y and Z (80 separate loads).

The calculation is running correctly.

1/ Efforts on solicited supports:

My first question is the following:
How could I obtain the resulting forces on the supports (some of the supports are among the solicited points)? For those supports I obtain such efforts:”*********************” -> huge resultants whereas this is not the case.

Idea:
We could consider a huge mass ex: 1000 tones and jointed via (Cnodes+”0”displacement) to all solicited supports blocking the same direction than the supports.
If we apply the amplitude displacement to this masse:
**We can solicit more than 100 points in the 3 direction… (quid fatal error #835 and CaesarII limitation and previous question on the forum)
**We could obtain resultant on solicited supports (I think that yes but what do you thing of this way to do it?)

Other Ideas witch "relatively" works (not accurate and depends on the considered segment length) is to consider the sum of upstream and downstream element forces (sum because the element force sense is opposite). What do you think of it… ?

2/Solicited supports and gap:
Concerning the solicited supports, some of them integer gaps…

For example:
10 mm gap in X direction, 0 mm gap in Y direction.
For harmonic study along X, 9.8mm are taken by my base case (operating case with temperature) and my harmonic solicitation amplitude is 0.17mm.

**Basically the resultant for this support in X direction is “-1 N ” that in to say 0 N -> no contact (logical because 9.8+1.7<10)
**In the Y direction, no gap, the support resultant is “*******************”->huge

One of the input data is friction coefficient, it mets us think that the harmonic calculation is none-linear calculation.
But is it really the case: if there is any piping vibration amplification, do the pipe is “able” to go in contact with the support:
[9.8mm+1.7mm+Xmm (induce by amplification)] >10mm, //
NB: 0.03mm is very small there is no change there is any contact

In this context, how Caesar II manage contact during harmonic calculation?? Is there a chance I obtain “reliable” output data for support (solicited or not) resultant in direction with gaps?

Thank you in advance.
Best regards
MH

You cannot have non-linear boundary conditions (gaps, +Ys) in a dynamic analysis. These restraints will be linearized based on the load case you select on the Dynamic Control Parameter dialog (Static Load Case for non-Linear Restraints). Gaps will be either close (no gap) or the restraint will be removed. +Ys will either be changed to Ys, or removed.

Your large restraint load "**************" is entirely possible, depending on what you specified in your harmonic input. For example, assume you have an anchor at node 100, and in the harmonic input you move node 100 by 0.1 inches (simulating some sort of vibration). Force = stiffness * displacement, so 0.1 in * 1.0E12 lb/in = 1.0E11 lbs. This value is too large to print in the output field, hence the stars.

Hello,

I understand CaesarII output: Force = stiffness * displacement= infinit.

The matter is that the stiffness between the support and "the earth" in nearly infinite in the software. So if i move the support i obtain an infinite force.

In the "real word": it is "the earth" (eventually concreat slab) that move and the support is linked to "the earth" so it moves too.

That is why I had the following proposition:

We could consider a huge mass ex: 1000 tones (virtual "earth")and jointed via (Cnodes+”0”displacement) to all solicited supports blocking the same direction than the supports (the supports will be delated)
If we apply the amplitude displacement to this big masse:
**We can solicit more than 100 points in the 3 direction… (quid fatal error #835 and CaesarII limitation and previous question on the forum)
**We could obtain resultant on solicited supports

I have to make a try but i would like to have your opinion before on this idea?


Thank you in advance
Best regards
MH

I'm not sure what you are asking here; but here are two points you may consider...
1) Yes, the restraint report shows "************" but the internal forces & moments report shows the load that teh pipe "sees" and this is the load on the support that you seek.
2) If you want several restraints to have the same harmonic load, just specify the same CNode for all these restraints. Think of the CNode as your slab. You will have to disable "Connect Geometry through CNodes" for the plot to show properly.

Hello,

You are right: for resultant on solicited points -> difference between "General element force" upstream and downstream at the concerned point would do the job.

Concerning the "virtual earth idea" it can works in principle.
This approach is used with other software but not flexibility one (general finite element software) it is hard to "adapt" it here. But would allow to solicit as much point as we like. Would the result be right, I don’t know, I think it is more complicated than it seems.

Anyway, I thank you for your help and your time.

Best regards.

MH