In a previous article, we introduced the risks and design considerations of repurposing an existing natural gas compressor system for hydrogen. That discussion focused on the fundamental wave theory of pressure pulsations and explored the characteristics of a simple bottle as a basic pulsation damping device.
A simple bottle functions by introducing a sudden increase in volume along the flow path, making it the most straightforward pulsation damper. However, more advanced designs incorporate multiple chambers separated by baffle plates, which can be connected through choke tubes for enhanced performance. Additionally, other damping mechanisms include side branches with connected bottles or the use of orifice plates to disrupt specific standing waves.
This article will explore these pulsation damping solutions and how their behavior changes when transitioning from natural gas to hydrogen systems.
Pulsation Bottle With Baffle Plates And Choke Tubes
The physics behind such a bottle relies on the principle of a Helmholtz resonator. This is an idealised principle that includes two large volumes which are connected by a smaller connection where the gas can oscillate in between. To investigate the effects of using such a resonator, the system from the previous case is extended. Another bottle with a slightly larger length of 800 mm is introduced to prevent any resonant interactions between the two bottles. The two bottles are connected with a piece of pipe, representing the choke tube, with a length of 200 mm, as shown in Figure 1.
Figure 1 | A pulsation bottle with two chambers that are connected through a choke tube.
Such a pulsation bottle has a certain frequency, called the Helmholtz frequency (fH), above which all pulsations will be dampened. This pulsation bottle with two chambers connected with a choke tube effectively forms a low-pass filter, reducing the intensity of any excitation above the Helmholtz frequency. In the range around fH the pulsations are amplified and at frequencies much lower than fH the influence on the pulsation is negligible. Normally, the system would be designed in such a way that the first harmonic of the excitation is at least 20% above the Helmholtz frequency. The Helmholtz frequency is given by:
Where a is the wave speed in m/s, A is the choke tube inner diameter surface in m2, L is the length of the choke tube in m and V1 and V2 are the volumes of the two bottles in m3. The Helmholtz frequency scales linearly with the wave speed, which, as discussed in the previous article, is vastly different for hydrogen than for natural gas. For the current example system, the Helmholtz frequency is 31 Hz for natural gas and 104 Hz for hydrogen. Using the transmission loss parameter to determine the damping across the multiple chamber bottle, Figure 2 is obtained. The two vertical lines denote the Helmholtz frequency for the two gasses. Higher transmission loss means more dampening at that frequency. It can indeed be seen that all frequencies above the Helmholtz frequency are damped.
For a well-functioning acoustic damper, the Helmholtz frequency should be at least 20% below the excitation frequency (Eijk 2023). The excitation frequencies are in large part determined by the compressor running speed. When reusing a natural gas system for hydrogen, the compressor is likely replaced and different running speeds are to be expected. The multi-chamber pulsation bottle should have been designed to be used with the specific compressor. Problems could occur when an existing pulsation bottle is re-used with a different process gas, and a compressor that is running at the same speed. The dampening characteristics of the bottle shift with the change in wave speed which could bring the Helmholtz frequency to be above the compressor excitation frequencies, which will then no longer be dampened. When the compressor operates at a frequency lower than fH the pulsation might be amplified, which is highly undesired.
Figure 2 | The transmission loss over the pulsation bottle for natural gas.
Pulsation Bottle On A Side Branch
Another configuration for damping of pulsations consists of a pulsation bottle on a side branch, which is also referred to as a side branch absorber. This has a different acoustic response to the harmonic wave input, as it is not part of the main routing. To model it, the simple bottle example is modified by connecting it to the branch side of a Tee piece in a header line. Obviously, there is a closed end condition at the top of the bottle, as the fluid can only enter and exit through the lower pipe. The model is shown in Figure 3.
Figure 3 | A simple pulsation bottle on a side branch
Similarly to the simple bottle, there are specific acoustic modes that will be effectively damped and other frequencies at which dampening is low. The pulsations are largely damped downstream of the absorber.
Figure 4 | A side branch absorver dampens the pulsations after the branch
The advantage of this configuration is that it is a feasible retrofit option in case the gas composition is changed and different pulsations need to be damped. Depending on the available space and the gas composition, the size of the bottle can be determined to ensure the pulsations remain within their allowable level. By applying a moveable piston at the top of the bottle, a tuneable bottle is created, which has a different volume depending on the piston position. This allows for tuning the dampening frequencies of the absorber to match with the changing process conditions.
Pulsation Damping Through An Orifice
Another common method to reduce pulsations involves the installation of an orifice plate to break the standing pressure wave. The narrow opening in the plate allows for the creation of an additional (nearly) closed end in the pipe of interest. An orifice plate is most successful when it is placed where the flow velocity variations are the highest. With an orifice plate, specific modes can be effectively eliminated. This method can be combined with any of the previously discussed pulsation bottle configurations.
The same standing wave will exist at a different frequency, when the gas is changed from natural gas to hydrogen. Whether this frequency is excited will be determined by the compressor characteristics. It is possible that the same mode, at a higher frequency, is still excited. In this case, the orifice plate will still effectively eliminate this mode. There is also a possibility that the mode is no longer excited, in which case the orifice plate would no longer be required at that location. When different acoustic modes are excited in the system after changing to hydrogen, restriction orifices in different locations could be necessary. When an existing system is to be used with hydrogen instead of natural gas, the necessity of requiring an orifice should be checked again.
If the hole’s diameter is reduced, the pulsations are further damped, but this is generally not necessary. Reducing the orifice diameter also increases the pressure drop over the orifice plate, hence influencing the pressure at the outlet of the system and reducing the system’s efficiency.
Conclusions
Pressure damping devices come in different shapes and sizes. In the previous article we only touched upon the use of a simple bottle, which is basically a sudden volume increase in the middle of the flow path. This article explored more advanced alternatives. We investigated a multi-chamber bottle with baffle plates connected via a choke tube, the effects of a side branch absorber, and the use of an orifice plate. Multi-chamber pulsation bottles, using baffle plates, are highly effective dampers. The connection between chambers through a choke tube mimics a Helmholtz resonator: above the Helmholtz frequenzy fH all pulsations are damped. Since this frequency is dependent on gas wave speed, a detailed pulsation analysis is recommended when the gas mixture changes. The use of a side branch absorber has similar damping characteristics as the simple bottle discussed before. Its power lies in the ability that it can be installed and sized accordingly depending on the available space around the compressor equipment. Its damping capabilities can even be enhanced more by having a variable volume, as it can be adjusted based on the composition of the gas mixture.
Another option to dampen pulsations is the installation of an orifice plate in the pipe of interest. It is a powerful option, because it can be combined with any of the pulsation dampers we discussed before. An orifice plate is most efficient when it is placed at locations where the flow velocity variations are maximum, as it forces the creation of a new closed end, effectively eliminating acoustic modes. When the gas composition is changed, the location and usage of the orifice needs to be reconsidered.
Dynaflow Research Group can assist you in tackling pulsation challenges. Whether you’re assessing the impact of a new gas mixture or facing unexpected issues, our advanced pulsation software, BOSpulse, enables us to support you at any stage of the design process.
How can DRG support you repurpose natural gas compressors to hydrogen
When considering a switch from natural gas to hydrogen, whether partially or fully, Dynaflow Research Group (DRG) is here to support you with comprehensive pulsation analysis services. We can provide expert guidance at every stage of your project. Whether you want to conduct a preliminary analysis to explore the implications of a different gas mixture for your current system, or if you are already in the midst of a turnaround and ran into unforeseen issues, DRG is ready to advise you and provide a thorough understanding of your observed problems.
With our industry-leading pulsation analysis software BOSpulse, we have the right tools available to ensure your assets keep operating safely. If you rather want to carry out your own pulsation analysis in the future, BOSpulse might be an interesting tool for you as well.
Authors: Thijs Krijger, Senior Engineer and Thijs van Lith, Engineer at Dynaflow Research Group
References
Eijk, André. “Effect on Pulsation and Vibration Mitigation Measures when Adjusting a Natural Gas System to Hydrogen Mixtures.” 13th EFRC CONFERENCE. Zagreb, 2023. 173 – 183
Share this article
Get in touch with one of our team members
