In a pressurized water reactor the high pressure system vent lines from the pressurizer and reactor are routed to a common header that can be emptied to the refueling water storage tank or a drain tank. During plant testing the valves are operated in the following sequence: the pressurizer isolation valve is opened to pressurize the common header, the pressurizer isolation valve is closed, then the drain tank isolation valve is opened. This sequence of valve operation verifies that the valves open and close properly — opening the pressurizer isolation valve allows steam to enter the common header and is verified by pressure indication via a pressure transducer, and opening the drain tank isolation valve decreases the pressure in the common header and verifies that the pressurizer isolation valve closed properly. During this sequence of valve actuation, the other solenoid valves in the system are subject to transient steam pressures. During one test sequence an isolation valve to the refueling water storage tank indicated that it was not closed for a period of several seconds. Since there is only one pressure transducer in the common header, a systemlevel analysis was performed to obtain a more detailed understanding of the transient pressures in the common header, and how that might affect solenoid valve performance.

This paper explores new analysis techniques and mitigation concepts developed to extend the current state of the art acoustic induced vibrations (AIV) analyses. These new methods are intended to provide more accurate evaluations of this phenomenon in an attempt to solve AIV problems found in blowdown and piping systems. Current screening methods for AIV are based on pass/fail data with minimal or undesired options for reducing the likelihood of failure for AIV events. Computational fluid dynamics simulations and finite element analysis in combination with lab testing of novel mitigation options using accelerometers, dynamic pressure transducers, and strain gages were performed to better understand the phenomenon and develop possible solutions to reduce the impact of AIV on piping systems. Results of the testing and analyses performed at the Southwest Research Institute (SwRI) indicate that there is a possible correlation with acoustic modes, structural modes, and elevated stresses during AIV events. Minor reductions in dynamic pressure fluctuations throughout piping during AIV events can be made by changes in valve geometry and piping configurations. Results of CFD modeling and analysis demonstrate that computational analysis can be used to evaluate mitigation strategies and suggest that the use of a dampener as a mitigation technique may be successful in reducing the amplitudes of dynamic pressure waves in piping systems caused by AIV events.

Damages and failures in industrial plants are often related to vibration issues. Reciprocating compressors are typically affected by vibration phenomena due to the very nature of reciprocating motion as alternating forces and pressure pulsations are direct and inevitable consequences of reciprocation. Many preventive technical measures are undertaken in the detailed engineering design to avoid high levels of vibration, e.g. properly designed foundations, mass balancing, volume bottles, restriction orifices and piping supports. Nevertheless, vibration problems may still arise after a machine is installed in a plant and is started up, and often the vibration is not the result of a poor detailed design of the compressor itself but may depends on the piping and supports layout in the plant. Considering the extremely high pressures involved in the LDPE process (discharge pressure is generally between 160 and 350 Mpa), especially in tubular reactor plants, safety is a key consideration, and avoiding vibrations and consequential piping ruptures is essential for optimal and safe plant operation. In a tubular reactor polyethylene plant, high piping vibrations were present on the Hypercompressor piping from the first machine start-up. Despite immediate analysis and small modifications to a few pipe supports, some areas continued to be subject to this phenomenon, potentially leading to ruptures, welding failures and hazardous gas leakages. Therefore, the end user decided to involve an independent third party consultant. During the site survey, the piping system was fully analysed to investigate the nature and the causes of the high vibrations, and it was decided that both vibration and pulsation measurements had to be performed, to obtain a complete and realistic picture of the phenomenon. While vibration measurement could be performed as a standard procedure on this kind of machine, pulsation measurement was a challenging operation since dynamic pressure transducers could not be used at such high pressures (above 100 MPa). Thus, an experimental technique was used. The pulsation measurements were performed using strain gage sensors that dynamically detected the circumferential deformation of the pipes. Information about the internal pressure was derived from the pipe deformation through the well-known theory of cylinders under internal pressure and in this way the pulsation measurements could be compared to the acoustical analysis performed during the detailed engineering phase. The analysis highlighted acoustic resonances that were not present in the project analysis, mainly due to an incorrect evaluation of the thermodynamic properties of ethylene gas, which changes significantly when the gas is subject to the high pressures at which the Hypercompressor works. Moreover, the vibration measurements were compared to the pulsations at some key points and to the mechanical natural frequency of the relevant piping segment, identifying also areas subject to mechanical resonance. After understanding the root cause of these vibrations, effective and low-impact countermeasures were recommended and implemented in a few key points, leading to a drastic reduction of the vibrations, below the limit values recommended by the standards.

An analytical model was developed by Sim to calculate the two-phase damping ratio for upward two-phase flow perpendicular to horizontal tube bundles. To verify the model, the present experiment is performed with a typical normal square array of cylinders subjected to the two-phase flow of air-water in the tube bundles. The diameter of cylinder is 18mm and the pitch ratio to diameter is 1.35. Using a pressure transducer and data acquisition system, pressure loss along the flow direction in the tube bundles is measured to evaluate the two-phase Euler number and the two-phase friction multiplier. The drag force along the flow direction on a tube is measured to calculate the drag coefficient and the two-phase damping ratio. The experimental results of the two-phase damping ratios are compared with the analytical results given by Sim’s model for homogeneous two-phase flow. It was found that, as increasing the mass flux, the drag force and the drag coefficients given by experimental test are close to the results calculated by the homogeneous model. As a result, the damping ratio can be evaluated by the homogeneous model for bubbly flow of sufficiently large mass flux.

Low Density Polyethylene (LDPE) production involves reactors operating at up to 3,000 bar pressure or even higher. The final stage of gas compression is done with a plunger-type machine called a “Hyper Compressor”. A “packing stack” for pressure sealing consists of multiple bronze packing rings and oil lubrication injection points; which are positioned in place by numerous packing cups. The distribution of pressure drop in a new cylinder and the changes in sealing behavior over its life are critical to design, but have long been prohibitively difficult to verify experimentally. Four cylinders were extensively instrumented; two in each of two successive test campaigns over four years’ time. Strain gages were located at the outside diameter of each packing cup to infer internal dynamic pressure across the packing stack. Included also in the high-speed data acquisition were strain gages on numerous additional components, thermocouples on packing cup faces, a pressure transducer in the low-pressure lubrication line, and existing eddy-current proximity probes on the plunger rods. Continuous monitoring was performed from the initial installation to final removal of each cylinder. Analysis of this high speed data along with long-term trends led to several conclusions regarding the design and operation of the Hyper cylinder. Testing also revealed various unusual events that can be further studied to better understand some of the unique behaviors that take place in the highly complex cylinder design of Hyper compressors.

The flow-induced pulsations in corrugated pipes result from a feedback loop between an acoustic resonator and the noise amplification at each shear layer in the axisymmetric cavities forming the corrugations. The quality factor of the resonator is determined by the reflection coefficients at the ends of the corrugated pipe and by the damping in the pipe. In this work, the damping of acoustic waves in a set of smooth and corrugated pipes is measured by a direct method. For these measurements, the tested pipes are placed between two measuring pipes equipped with flush-mounted pressure transducers to allow reconstruction of the acoustic waves with the two-microphone method. Loudspeakers are used to generate acoustic waves, and anechoic terminations allow near reflection-free conditions. The tests are done in air without flow and with flow velocities up to 60 m/s. The results for the corrugated pipes allow to investigate the influence of the corrugations on the damping. Two cases are considered: Without flow, the visco-thermal damping is increased for the corrugated pipe compared to a smooth pipe of same diameter. When there is a flow in the pipe, the damping results depend strongly on the flow velocity. At certain frequencies which depend on the flow velocity, the damping increases or decreases. The regions of increase or decrease are shown to be at constant ranges of Strouhal numbers. The decrease of the damping can in some cases be such that acoustic waves are amplified through the corrugated pipe. This corresponds to the acoustic source behavior of the shear layers. The measured amplification is compared to the computed source term due to the axisymmetric cavities.

A new experimental device was developed to observe and measure the dynamic generation of supercritical CO 2 (scCO 2 ) in a closed chamber. The temperature and pressure were measured locally using a thin thermocouple and a pressure transducer, respectively. The Rayleigh scattering in the chamber was visualized using a high-speed video camera. The liquid CO 2 was heated by a ceramic heater located at the top or bottom of the chamber. Using this new experimental device, the color variation due to Rayleigh scattering was observed. The temperature profile was stable and scCO 2 was generated slowly within a few seconds when heating occurred at the top. For heating at the bottom, scCO 2 was created faster within 1 s. Additionally, natural convection, turbulence, and a phenomenon resembling boiling were observed. Numerical simulations of the scCO 2 creation in a chamber were also performed using the COMSOL Multiphysics application along with the PROPATH package to obtain the thermophysical properties of CO 2 . The simulation results showed that scCO 2 creation for heating at the top was stable due to the gas-like properties of the scCO 2 near the heater. For heating at the bottom, the density distribution initially depended on the temperature distribution. However, as natural convection developed, the flow in the chamber was perturbed and the density distribution depended on both the temperature distribution and the density fluctuation caused by the convection vortices. The same tendency was observed in the experimental results. Using PROPATH, the density variation in the scCO 2 generation was estimated from the measured temperature and pressure. The P-T diagram and density variation were compared in each experiment.

The pressure time-history associated with a propagating detonation within a pipe can be reasonably well described using a one-dimensional model. These models describe the advancement of the detonation front and the expansion wave behind the detonation front, over which the pressure decays. The length of the expansion wave, and the corresponding pressure decay period, is a function of the distance the detonation wave has traveled. One-dimensional detonation models typically assume an isentropic expansion behind the detonation front. The consequence of this assumption is that the expansion wave will grow infinitively long in an infinitely long pipe. The loss of energy due to heat transfer, however, has been observed to result in a natural limit in the length of the expansion wave, and corresponding pressure decay. The implication of this limit on piping design for gaseous detonations is that there is a limit on the pressure-impulse that is applied to the system. When a detonation propagates through a piping system, the detonation pressure produces unbalanced loads at the bends, tees, reducers, and closed-ends. These transient loads are a function of both the peak pressure and the details of the pressure decay. In long piping systems, the decay period of the expansion wave can become sufficiently long that very high loads are imposed on the system. Crediting a natural limit of these loads may have significant implications for the design of the pipe and its supports. Although the existence of an expansion wave limit has been previously observed, consensus on its length does not currently exist in the literature. This paper describes detonation tests that were performed in both 2-inch and 4-inch 36.6m long pipes using mixtures of hydrogen and nitrous oxide. Pressure transducers were installed at periodic locations along the pipe. The slope of the initial pressure decay of the detonation was used to characterize its shape. The slope of the initial pressure decay was found to exponentially approach a limiting steady state value, and this limiting value was achieved after the detonation had propagated approximately 120 pipe diameters.

There are pipes with square or rectangular cross sectional shape rather than more common circular shape. Moreover, a part of the walls can be made of different materials or have different wall thicknesses. To estimate pressure wave propagation in such a channel, we examined water hammer phenomenon and measured wave speeds in rectangular tubes assembled with different wall materials and thicknesses by strain gages and pressure transducers. We newly proposed theoretical wave speeds based on the classical theory and the junction conditions of two walls: fixed, simply supported, and mixed supported. Circumferential strains on the wall can be estimated from the different forms derived by the different junction conditions. We confirmed that theoretical predictions with the simply supported conditions agreed with experimental wave speed in compound tubes, while wave speeds in ready-made square tubes with the same wall-thickness were close to those by the fixed conditions.

The phenomenon of condensation-induced waterhammer in an ammonia refrigeration system was investigated experimentally. Waterhammer was generated by introducing warm ammonia gas over static subcooled ammonia liquid placed in a horizontal 146.3 mm diameter carbon steel pipe approximately 6.0 m long. By means of fast response piezoelectric pressure transducers and a high speed data acquisition system rapid dynamic pressures were recorded whenever a condensation-induced event occurred. Employing top-mounted diaphragm pressure transducers to sense gas pressure the speed of liquid slugs propagating along the pipe was determined. The occurrence of condensation-induced waterhammer depended upon three major variables; namely, (1) initial liquid depth, (2) liquid temperature, and (3) mass flow rate of warm gas. For given liquid depth and temperature, once the warm gas threshold conditions were exceeded, shocks occurred with greater magnitude as the mass flow rate of warm gas input was increased. With adequate subcooling condensation-induced waterhammer occurred for initial liquid depths ranging from 25% to 95% of internal pipe diameter. The threshold mass flow rate of warm gas necessary to initiate waterhammer was greater as the initial liquid depth was lowered. For numerous tests with sufficient gas mass flow rate a condensation-induced shock occurred at the end cap of the test pipe, generating an acoustic (pressure) wave that propagated at a modified speed of sound toward the gas-liquid interface at the back of the slug. Peak shock pressures could be correlated with the gas mass flow rate for various initial liquid depth and suction pressures and temperatures. As the initial liquid temperature and pressure were increased the hydraulic shocks became lower because of smaller subcooling.

Hydraulic piping systems are sensitive to excessive acoustic emissions which can give rise to vibrations or to failure of mechanical elements due to fatigue. Common excitation sources are centrifugal pumps due to the periodic interaction of the impeller with the volute tongue. They radiate pressure fluctuations into the connected circuit at the blade-passing frequency which are reflected in the circuit. The aim of the present investigation was the experimental characterization of the perturbations induced in a piping network as a function of the acoustic impedance of the circuit using fast-response pressure transducers. Three transducers were placed along the discharge pipe to decompose the pressure signal into the radiated and reflected acoustic wave, with amplitude and phase. The speed of sound in the water pipelines was determined experimentally. Results of impedance at the pump tongue were compared with a theoretical approach, using a Transfer Matrix Analysis, where each pipe element is represented by a 2×2-matrix, relating the acoustic pressure and velocity fluctuations at the two ports. Attenuation was considered by using complex wavenumbers. The acoustic impedance was changed by using different rotation speeds and by using a cavity which works as a harmonic oscillator whose resonance frequency can be changed. The results presented for different operating points show the influence of changing impedance on the pressure perturbations due to pump-circuit acoustic coupling. The pump was built with a transparent impeller and volute to conduct Particle Image Velocimetry (PIV) measurements in a plane perpendicular to the pump rotation axis. Phase-averaged velocity fields were obtained at different blade positions in the zone around the tongue. Vorticity fields were derived from it and turbulence is shown by representing in-plane turbulent kinetic energy (TKE). Strong vorticity and turbulence occurred in the volute channel behind the blade and near the tongue tip.

Detonations and deflagration-to-detonation transition (DDT) are experimentally studied in horizontal pipes which are partially filled with water. The gas layer above the water is stoichiometric hydrogen-oxygen at 1 bar. For detonation cases, ignition and transition occur outside of the water-filled section. For DDT cases, ignition and transition occur over the surface of the water. Pressure and hoop strain are measured incrementally along the pipe, with pressure transducers located both above and below the water. The detonation wave produces an oblique shock train in the water, and the curvature of the pipe is seen to focus the shocks at the bottom, resulting in peak pressures that are 4–6 times higher than the peak detonation pressure. Such pressure amplification is observed for water depths of 0.25, 0.5, 0.75, 0.87, and 0.92 pipe diameters. For a water depth of 0.5 diameters, pressure is also recorded at several circumferential locations in order to measure the shock focusing phenomenon. Peak hoop strains are found to decrease with increasing water depth, and transition to detonation is seen to occur for water depths as high as 0.92 pipe diameters.

Pressure vessel burst (PVB) is a class of explosion for which there are hazards at virtually all chemical processing facilities. PVBs present both airblast and fragmentation hazards. Blast prediction methods specific to PVBs were first developed in the 1970s and revised blast curves were published in 1995. The published blast curves were developed for spherical vessel bursts, whereas most pressure vessels in use in industry are cylindrical. Blast effects around a bursting cylindrical vessel are not uniform as with a spherical vessel. The blast to the side of a cylindrical vessel is stronger than off the ends, creating non-circular pressure contours. The directional effects diminish with distance as the expanding shock wave approaches a spherical shape. A correlation was developed in the 1970s to account for directional effects using high explosive test data, the best available resource at the time. Like all test programs, pressure transducers extended to limited distances from the explosive charge, yet the data are often extrapolated to a far greater distance. This paper presents the results of recent work on directional effects specific to bursting cylindrical pressure vessels and provides new correlations for blast overpressure and impulse for a range of vessel geometries and burst conditions. The results can be used to predict the airblast hazards from cylindrical PVBs over the range of standoff distances for which directional effects exist.

The phenomenon of fluidelastic instability forms a major limitation on the performance of tube and shell heat exchangers. It is believed that fluidelastic instability is attributed to two main mechanisms; the first is called the “Damping Mechanism”, while the second is called the “Stiffness Mechanism”. It is established in the literature that in order to model the damping controlled fluidelastic instability, a finite time delay between tube vibration and fluid response has to be introduced. Experimental investigation of the time delay between structural motion and the induced fluid forces is detailed in the present study. A parallel triangular tube array consisting of seven rows and six columns of aluminum tubes is built with a pitch ratio of 1.54. Hot-wire measurements of the interstitial flow perturbations are recorded while monitoring the tube vibrations in the lift and drag directions. Pressure transducers are installed inside the instrumented tubes to monitor the fluid forces. The phase lag between tube vibration and flow perturbation is obtained at different locations in the array. The effect of tube frequency, turbulence level, location of measurements, and mean gap velocity on the relative phase values is investigated. It is found that there are two well-defined regions of phase trends along the flow channel. It is concluded from this study that the time delay between tube vibration and downstream flow perturbation is associated with the vorticity convection downstream, while the time delay for upstream perturbations is associated with the effect of flow separation and vorticity generation which is propagated upstream from the vibrating tube.

The aerodynamic excitation of ducted cavity diametral modes, which are inherently antisymmetric acoustic modes, by the oscillation of the axisymmetric free shear layer gives rise to complex flow-sound interaction mechanisms, in which the acoustic diametral modes do not possess a preferred azimuthal orientation. The azimuthal behaviour of this self-excitation mechanism is investigated experimentally. The study is performed for axisymmetric shallow cavities in a duct for a range of cavity length to depth ratio of L/d = 1 to 4, and for Mach numbers up to 0.4. Three pressure transducers flush mounted to the cavity floor are used to determine the acoustic mode amplitude and orientation. The excited acoustic modes are classified into spinning, partially spinning and stationary diametral modes. An analytical model based on the superposition of two orthogonal modes with 90° temporal phase shift is developed to reproduce the spinning and the partially spinning diametral modes. The developed model clarifies the observed complex behaviour of the azimuthal modes.

There are reported cases where a pressure transducer was installed at the bottom end of a liquid-filled instrument tube for measuring pressure in a recirculation pipe. For example, in one case, a small diameter instrument tube extended a long distance downward from a power plant recirculation pipe to the transducer. A continuous pressure oscillation was observed, which was not consistent with any other measurements in the recirculation system. However, it was found from paper-and-pencil analyses that such an oscillation could be traced to a resonant frequency in the system if the vertical liquid column rested on a postulated trapped gas bubble, which could have remained after venting. The relative compressibility of the liquid column is also compared with the gas bubble compressibility to show when acoustics of the liquid column are important.

The world’s largest district steam co-generation plant was successfully brought on line following major modifications in the spring of 2005. During the initial “shake-down” operation of the high pressure steam send-out system, piping vibrations were experienced, which disrupted flow meters and other equipment. Operation of one of five identical flow control valves, used in parallel to regulate steam flow to the street distribution system, resulted in the high vibration levels. To investigate the cause of the vibrations, a project was undertaken, which included testing of the piping system utilizing dynamic pressure transducers and accelerometers and engineering analysis of the acoustics and piping structural dynamics. Multiple combinations of open and closed valves were investigated, and the likely root cause was identified as flow-induced vibration originating at a tee in the system. A piping layout modification was designed. The modification consisted of rerouting the piping downstream of one flow control valve to bypass the source of the flow-induced vibration. Following the piping modification, further tests were conducted that showed the reconfiguration successfully mitigated the vibration.

The effect of an internal turbulent bubbly flow on vibrations of a channel wall is investigated in this paper both experimentally and theoretically. Vibrations of an isolated channel wall and associated wall pressure fluctuations are measured using several accelerometers and pressure transducers along streamwise direction under various gas void fractions and characteristic bubble diameters. A waveguide theory based mathematical model, i.e. a solution to the 3D Helmholtz Equation in an infinite long channel, and the physical properties of bubbles is developed to predict the spectral frequencies of the vibration and the wall pressure fluctuation, the corresponding attenuation coefficients of spectral peak and propagated phase speeds. Results show that compared with the same flow without bubbles, the presence of bubbles substantially enhances the power spectral density of the channel wall vibrations and pressure wall fluctuations in the 250–1200 Hz by up to 27 dB and 26 dB, respectively, and increases their overall rms values by up to 14.1 times and 12.7 times, respectively. In the lower frequency range than the resonant frequency of individual bubble, i.e. 250–1200 Hz range, both vibrations and spectral frequencies increase substantially with increasing void fraction and slightly with increasing bubble diameter. The origin for enhanced vibrations and wall pressure fluctuations is demonstrated to be the excitation of the streamwise propagated acoustic pressure waves, which are created by the initial energy generated during bubble formations. The measured magnitudes and trends of the frequency of the spectral peaks, their attenuation coefficients and phase velocities are well predicated by the model. All the three variables decrease as the void fraction or bubble diameter increase. But the effect of void fraction is much stronger than that of bubble diameter. For the same void fraction and bubble diameter, the peaks at higher spectral frequencies decay faster.

If a pressure vessel containing a pressure liquefied gas (PLG) is exposed to a severe fire, there is a chance the tank will rupture and possibility even explode. It has been shown that the energy stored in the liquid phase is important in the outcome of such an accident. In order to better understand tank failure and energy storage, it is desired to understand the complex thermo-hydraulic response of the lading prior to a thermally induced rupture. In the summers of 2000 and 2001, a series of controlled fire tests were conducted on horizontal 1890 litre (500 US gallon) propane tanks. The test tanks were instrumented with pressure transducers, lading and wall thermocouples, and an instrumented flow nozzle in place of a pressure relief valve (PRV). A computer controlled PRV was used to control pressure, while high momentum, liquid propane utility torches were used to heat the tank. PRV blowdown and fire conditions were varied in this test series while all other input parameters were held constant. It was found that the lading response and energy storage varied according to the fire conditions and PRV operation. The location and quantity of the burners affected the thermal stratification within the liquid, and the swelling and frothing at the liquid/vapour interface. The blowdown of the PRV affected average tank pressure, average liquid temperature and time to destratify. PRV operation caused enhanced convective cooling on the vapour space wall. The degree of blowdown dictated the pressure drop, and thus the liquid flashing and subsequent swelling were affected. This paper will discuss these thermohydraulic responses and their role in the tank failure.

Fracture experiments were performed on thin-walled and preflawed aluminum 6061-T6 tubes. Flaws were machined as external axial surface notches. The tubes were 1) statically loaded with oil, 2) statically loaded with nitrogen, and 3) dynamically loaded with gaseous detonations traveling at 2.4 km/s. The experiments were controlled so that comparisons could be made on sets of specimens with the same material, tube and flaw geometry, nominal loading amplitude, and flange supports, with the only difference being the dynamics of the loading. It was found that there is a significant difference in crack propagation behavior for the three types of loading. In this paper, fracture behavior will be discussed along with the fluid dynamics involved. The tubes were also instrumented with pressure transducers, crack detection gages, and strain gages so that data on loading, crack propagating speeds, and strain history can be compared.