Jaime;
Using CAESAR, you cannot PREDICT harmonic response, you can only trouble-shoot by simulating a response that you may encounter in the field.
There are several reasons for this:
1) CAESAR does not calculate the EXACT natural frequency, only approximates based on the mass-spring model. We need an infinite number of nodes to generate the exact natural frequencies.
2) You most likely do not know the EXACT forcing function (i.e. what is the magnitude & forcing frequency)
3) Your mathematical computer-approximation will differ from actual installation. For example, exact restraint locations and stiffness will not be the same in reality as in your computer model. Mass may vary as well - insulation and fluid density EXACTLY correct? Is the mass of attached instrumentation or any other attachments considered? Structural steel can contribute mass, where you may model it as a simple restraint (stiffness, no mass).
4) The actual harmonic response curve is VERY narrow (vertical axis = response, horizontal axis = frequency). This means that even very small alterations of calculated natural frequency and/or forcing frequency can give dramatically different predicted system response.
All of this means that the variety and magnitude of differences between your computer model and the real system are far too great to enable prediction with CAESAR (or any other beam-element computer program).
The Harmonic analysis feature in CAESAR is intended as a tool for evaluating field vibrations that already exist. That is, you can fine-tune your model and forcing function such that the results on the computer simulate the vibration problem in the field. Once you simulate the "shape" (displacement solution) you can answer the question: "Is there a problem?". If it is a problem, you can modify the model through changes in stiffness (restraint) or mass, run the analysis with the same excitation, and see what the new system response is, thereby evaluating possible solutions to the problem.
There are times when changing the forcing function can be a solution (eg. slow the compressor), but this is not typically the preferred solution to such problems.
In advance of construction 'Analog studies' can be performed to examine acoustic and other characteristics of a system to help guide design to avoid vibration problems.
There are also basic guidelines that some follow, such as those provided above.
This may be a good time to review relevant pages of sections 4 & 5 of your pipe stress seminar course notes.
Regards,
Jim.
_________________________
J.