Phone: +31 (0)85 058 0046
E-mail: infoaanvraag@dynaflow.com
E-mail: infoaanvraag@dynaflow.com
ABOUT US
HEAD OFFICE
Laan van Oversteen 20
6th floor
2289 CX Rijswijk
The Netherlands
The objective of the API 618 pulsation study is to prevent large pressure pulsation levels. Excessive pulsations in a piping system can reduce the service life of a compressor, produce significant noise and also produce large shaking forces which can lead to pipe failure.
The pulsation characteristics of a piping system depend on factors such as:
Two approaches can be employed:
as described in the API 618 2024 edition.
From a project execution perspective, four (or five) different phases will be distinguished:
We always start the project by modelling the relevant piping system using our own pulsation software package BOSpulse. The compressor(s) are modelled in detail as well as any pressure reducing components such as dampeners/accumulators or orifice plates. All downstream piping through which the pulsations will travel is modelled. The modelling of downstream piping can end at for example a closed valve, injection points into a gas filled pipe or when reaching a large diameter vessel.
A number of simulation cases are set-up based on the compressor RPM and the different operating conditions. Simulation cases will include a change of compressor RPM from +- 10%. This is done to account for any geometrical uncertainty in the model as well as uncertainty in the gas properties and does not represent an actual change in compressor RPM.
Preferably, the mechanical model is built during this phase as well. However, this might not always be possible at the start of the project as this requires more information about the system such as support locations. Optionally, the structural steel is also modelled.
The simulation cases as identified during the previous phase are analyzed using BOSpulse. The results are interpreted and are compared to the allowable values as described in the API 618. The following checks are performed:
If API 618 criteria are not met, we will advise with the client about possible solutions. These solutions vary per system but can include:
During this phase, the shaking forces on the system are assessed. A mechanical model is first built (if not already done in Phase 1) and then analyzed. If the structural steel is not included in the mechanical model all supports are assumed stiff relative to the piping. It is up to the client to ensure that this assumption is valid for all supports.
The review following design approach 2 entails doing a mechanical natural frequency calculation. The review results in a table of various pipe sizes that indicates the maximum allowable span between pipe supports, based on the maximum compressor operating speed. The spans are designed to provide enough separation of the mechanical eigenmodes from the acoustic shaking force frequencies.
This phase is entered if the system is designed according to design approach 3. When the excitation frequency separation margin and the shaking force amplitude guidelines for the piping system cannot be met, a forced mechanical response analysis of the system shall be performed. The goal of the mechanical analysis is to determine the maximum cyclic stress amplitude that can be produced by the shaking forces resulting from the pressure pulsations.
This is done by selecting the maximum shaking forces as calculated during the pulsation analysis and applying in a harmonic stress solver to the mechanical model. DRG uses a coupling between BOSpulse and ANSYS to perform the mechanical analysis, but it can also be performed using a stress software program such as CAESAR II. The maximum cyclic stress amplitude is compared to the endurance limit for the material in order to prevent fatigue failure.
If the stress amplitude exceeds the endurance limit of the material, we will advise the client about possible solutions to resolve this issue. Recommendations include:
In agreement with the client, the reporting will be presented in the form of an elaborate presentation or a written report. The reporting will include the model details, input parameters, hypothesis, a description of the simulation cases and the results. If applicable, the recommended changes will be described including the consequences of these changes with respect to the problems identified. These recommendations for improvement are always an integral part of our deliverables.
The following additional information is required:
Laan van Oversteen 20
6th floor
2289 CX Rijswijk
The Netherlands
© Dynaflow Research Group BV
