Dear All,

can i know is there any option in CAESAR to get the output of amplitude at a particular node number.

( Or )from the displacements dx, dy, dz can we find out the amplitude?

Adding to that question,

1. for my system there is no slug, seismic or any dynamic loads, i just did modal analysis, so i got frequencies what abt amplitude. can i take the displacements in static run as amplitude at a particular frequency.

2. I Think amplitude is max of displacement, but in which direction i have to take that displacement value either X or Y or Z??

A Modal Analysis only yields the frequencies and mode shapes. There are no actual system displacements or stresses - until you apply a load and perform a full dynamic analysis.

Thanks Rich, I agree with you, BUT

PROBLEM 1. if i perform full dynamic analysis, then maximum displacement is amplitude? IS THIS STATEMENT IS CORRECT.

if it is correct which maximum I have to consider as amplitude either DX DY or DZ


PROBLEM 2. For design of ACTUATOR, vendor asking the frequency and amplitude values, and for that system i dont have any dynamic input so what is the solution.
Maximum I can get frequency from modal run what about amplitude.

1) What type of dynamic analysis are you refering to? You would have to consider all directions X, Y, Z.

2) Actuator for what? As to the frequency, there are an infinite number of natural frequencies for a piping system. Are you sure you're after a system natural frequency, or perhaps the frequency of some dynamic (loading) event?

Thanks for ur reply sir,

1) if I consider dynamic loading is slug, then from ur statement what I understand is at a particular node in x direction max displacement, one amplitude and in y direction max displacement second amplitude and same as in z. So at a particular node it has 3 amplitudes in 3directions (or) we have to consider max of these 3values ?. please clarify,

2)actuator is for valve operating , and iam not doing any dynamic loading for that system, iam just run the modal analysis to get the natural frequecies of the system( iam not considering about freq after 33hz), that is clear, now vendor is asking both freq and amplitude values. For the design of actuator for valve. What is the solution for this problem?

Pls help , iam new in this topics

1. A peaceful solution would be to consider sqrt[Dx^2+Dy^2+Dz^2] as "amplitude".

2. If I were you, I would give Vendor a table with f, Dx,Dy,Dz, and sqrt[Dx^2+Dy^2+Dz^2] for the points considered, maybe Vendor wants to perform a generalized Fourier model; anyway it seems that Vendor knows better what he wants. Why stop at 33 Hz (this is a limit coming up from other topic), give Vendor more and see what Vendor says...
I'm new too in this kind of topics, where the Valve Actuator is like the hammer of Thor. May the mystery of Vendor Actuator be with us!

thanks mariog for ur response,


1. ok that's a good idea.

2. BUT if i run only modal analaysis,(freq i will get) output displacements are not actual then which displacements sqrt[Dx^2+Dy^2+Dz^2] i have to take?, STATIC analysis output displacements???

Dear All please respond and share ur valuable knowledge.

Of course they are not "actual" values because you haven't defined the "actual" phenomenon giving such values.

Have no idea on what your Vender asked for, maybe for a frequency response analysis of feedback control system. Why not try to clarify with Vendor?

Dear Mariog,

1. iam only doing static analysis and next modal run, now if he wants to do "frequency response analysis" then what i have to do for amplitude????


2. This is the actual mail i got.... On/Off Ball Valve (Novated item) Vendor has asked for Project data Like Frequency and Amplitude of Vibration (If any) for finalizing the Design for Actuator. This data is also required for identifying any ground support / Structure support required for Actuator and Accumulators of the Valves. from this mail can we decide which amplitude i have to give or any further information required??

You are not understnding the requirement. Modal analysis is not going to help, neither any dynamic analysis is going to help. Try to support the pipe upstream and downstrem of the valve to make rigid and tell the vendor negligible vibration from pipe.

Dear pku,

Thanks for ur response

Good suggestion in some cases. but every time Is it possible to make rigid near by valve piping . I think Thats y in that mail also, he written IF ANY.

Now if it is not possible to make the pipe rigid. Then?

Durga,

Vendor does not say "if pipe is rigid".
Vendor says " Vibration (If any)--> than announce me".

You would say "we made stiff supports around and piping arrangement around is quite rigid, vibration is not an concern".

Or you would say "we've been obliged to put there spring supports, arrangement is not stiff, is possible to have vibrations due to [...] and the parameters of vibration are [...]".
In this case you must fill in the [...] fields- that's all.
How? Here is your engineering contribution.

Sometimes is "not simple". They are cases when vibrations can be self-excited - associated with flow hydrodynamic instabilities (the system vibrates under no periodical external forces) and you cannot be 100% sure the system vibrates or not unless you have references on similar cases, anyway (IMO) you cannot describe accurately such vibrations by math.

I think you have enough elements to take a decision which is your way.

Dear Mariog,

Yes even i mean the same thing, IF ANY means -- if any vibrations are there send amplitude vs freq values.

The problem is solved, i made the nearby valve routing rigid and no question of freq and amplitude.

Thanks Mariog, PKU, and Richard for this help

For my knowledge sake, still iam in confusion abt these three statements

Can you please tell me.

1. if i run only static analysis, how to find amplitude value ?
2. if i run only modal analysis how to find amplitude value ?
3.if i run only dynamic analysis, how to find amplitude value, i think Max displacement in x,y,z ?

Dear Durga, Mariog,

Throwing our recent experience for such kind of problem.

In our case the vendor asked the same,but in different way, the actuator is in Horizontal direction, 24", length is 3000MM, weight is also very high, & location is near to equipment nozzle Hence we model the actuator in CAESAR as rigid by it's weight & length. check the sagging & surprising we had to put the Spring Support to sustain the weight of actuator & Guide also to prevent vibration.Prepared Special Pipe Support by concerning with client & Valve Vendor.

Also, just near to the valve, drain of small branch vibrate,so,we provide gusseting that drain location on both sides.

I suggest to Provide DX,DY,DZ values to vendor at the last node of valve Actuator.(STATIC ANALYSIS + MODE SHAPE-Graphics)

In your case, is it vertical or horizontal direction, if the length,weight are high, check the actual condition & requirement from vendor as mariog said in his post.

may be it can help you.

Regards
SND

Dear SND,

In my case also same situation, 24" horizontal , huge weight, that's y we also model the actuator in CAESAR as rigid by it's weight and length.

what I learn from the above discussions......

Summary:

1.If any piping system having only static analysis case, for the design of ACTUATOR--- Provide DX,DY,DZ values to vendor at the last node of valve Actuator.(STATIC ANALYSIS + MODE SHAPE-Graphics. -- From SND

2.If any piping system having dynamic analysis also, we can get the actual displacements values---- Provide a table with f, Dx,Dy,Dz, and sqrt[Dx^2+Dy^2+Dz^2] for the points considered. -- From Marig

3. Try to support the pipe upstream and downstrem of the valve to make rigid and tell the vendor negligible vibration from pipe. -- From PKU

Thanks for sharing ur valuable inforamtions.. in the above statements if any one find mistake/wrong/any suggestions ur welcome....

As my "contribution":

- try to define accurately the questions you post, otherwise it is possible to get a wrong answer;
- you can have "actual" values when define "actual events"; a modal analysis can get only a basis of calculation (let's say eigenvalues and eigenvectors- basis in the vector space); what I suggested there was to transfer to vendor (if it is the case!) that basis of calculation when Vendor is not able to understand that you cannot have actual results without defining the vibrating phenomenon.

Firsty thanks for ur contribution Mariog,

- I accepted may be my question is not accurate/clear Because intially i don't have complete information, & for me it is first time interacting with that vendors, even this area is also new. so what i thought is I will get some knowledge on this area from the experienced people.

Thanks once again and i will try to follow ur suggestions in my next Post.

Durga,

Re-read my first post above. You can't get amplitudes from a modal analysis. Remember a modal analysis is the free I damped vibration of the system. This yields an infinite number of independent modes. The true response (which will yield amplitudes) is obtained from the contribution of all modes to a specified loading. No loading, no actual response.

Dear Richard,

Can You Please check these statements, what I understand from this Discussions..

1.If any piping system having only static analysis case, for the design of ACTUATOR--- Provide DX,DY,DZ values to vendor at the last node of valve Actuator FROM THE STATIC ANALYSIS RESULTS.(STATIC ANALYSIS + MODE SHAPE-Graphics). BUT NOT from the modal analysis because "There are not actual system displacements or stresses."

2.If any piping system having dynamic analysis also, we can get the actual displacements values---- Provide a table with f, Dx,Dy,Dz, and sqrt[Dx^2+Dy^2+Dz^2] for the points considered FROM THE DYNAMIC ANALYSIS RESULTS Because we are applying the actual loads we will get the actual displacements.

3. LAST OPTION, Try to support the pipe upstream and downstrem of the valve to make rigid and tell the vendor negligible vibration from pipe.

Yes to all three questions.

You can't find amplitude from static analysis or modal analysis (1 or 2 above). But you can see amplitude (response)in the Caesar output when you conduct dynamic analysis such as Hamonic or time history.