Hi,

I have to perform an harmonic analysis of a piping system attached to a reciprocating compressor. The system is very wide; I'm trying to apply harmonic forces due to pressure pulsation at every bend, tee and reducer but when I run the caesar model the fatal error 834 occurs:
"Counting generated harmonic nodal forces, there are more than 100 separate loads. The current maximum is 100 set in DYN.FOR."

Is it possible increase the maximum number of nodal forces that one can apply (over 100)?

Thank you.

No you cannot increase this value.

Typically for a pulsation problem you're trying to match observed field observations. Is this your situation?

If so, then you estimate the loading by applying a single load (force and frequency) axially in one of your pipe elements.

Pressure pulsation may produce a force imbalance along each elbow-elbow pair. But all these loads are not in phase. Have you established the phase relationship between each set?

As reguard to the situation above，how can we take the phase into consideration？

Here:

Thank you Mr Ay and Mr Diehl for your answers,

If possible I want to explain better my situation.

Like I have just said, I have to perform an harmonic analysis of a piping system attached to a reciprocating compressor (used for natural gas in Oil&Gas industry). The aim of my simulation (that is performed with Caesar II) is to verify the overall piping system according to the "Design Approach 3" recommended by API 618 (fifth edition, pages 54-62).

I have previously performed an acoustic analysis with a cfd 1D code to evaluate the pressure pulsations inside each pipe of the system and I have also performed a modal analysis to evaluate the mechanical natural frequencies of the system (by Caesar II).

Now I'm traying to simulate the forced mechanical response of the system by imposing on it all the necessary shaking forces.
As described in the RECOMMENDED PRACTICE API 688 (April 2012, pages 9-14), it is necessary to apply the harmonic shaking forces at every geometric discontinuity of the system such as elbows, reducers, tees and capped ends.
As widely shown in literature, in general, the magnitude of the harmonic force is F = (1/2)*dP*A (where "dP" is the peak-to-peak pressure pulsation at that point and "A" is the area of the pipe section), the phase angle in degree is phi= 360*f*(L/c) (where "L" is the distance between the point of interest and the point of the system at which is set phi=0, "c" is the speed of sound of the fluid inside pipes and "f" is the frequency of the pressure pulsation) and the force is applied along the pipe axis (and so for each bend there are two separate loads applied with different directions and the same value of amplitude and phase).

It's clear that for some kind of piping system, such as for example natural gas compression packages used in O&G industry, the number
of geometric discontinuities could be very high and so the number of forces to be applied. In conclusion, in order to perform a rigorous calculation as described in the API standards, in the software is necessary to set a number of harmonic forces significantly greater than one hundred!

About this kind of simulations that are required by API 618 and API 688 and that could be performed (in theory) by the Caesar II software...
is it possible that in a next version of Caesar the maximum number of separate harmonic loads that can be set will be increased?

In addition to this I have another question related to the harmonic analysis. I think that is strange that the software allow to make an harmonic analysis over a frequencies range (specifying the frequency increment in the excitation frequency tab) despite the presence of phased forces. Indeed the phase is a function of the frequency and so if the frequency changes over a range also the phase of each force must change.

Thank you very much for your attention.

Best regards.

I will note your request for more harmonic load sets in CAESAR II. Thank you.
Clearly the your use of CAESAR II to address the API 618 approach exceeds current capabilities.
If there are full anchors internal to your system your harmonic study could be successfully performed on those independent subsets.
I agree that, in your application, changing phase angles will be associated with changing frequencies. Our intent though was to allow you to sweep through a range of frequencies of some single load (e.g., mechanical harmonic motion at a nozzle connection). This would allow you to investigate maximum response where the true stiffness and mass of your system are not specifically known.