Dynamic/Equivalent Static Analysis of Dynamic Events (Notes Prepared by S Koley)

A piping system may respond far differently to a dynamic load than it would to a static load of the same magnitude. Static loads are those which are applied slowly enough that the system has time to react and internally distribute the loads, thus remaining in equilibrium. In equilibrium, all forces and moments are resolved (i.e., the sum of the forces and moments are zero), and the pipe does not move. With a dynamic loada load which changes quickly with timethe piping system may not have time to internally distribute the loads, so forces and moments are not always resolvedresulting in unbalanced loads, and therefore pipe movement. Since the sum of forces and moments are not necessarily equal to zero, the internally induced loads can be differenteither higher or lowerthan the applied loads. For this reason, different analysis methods must be used to determine response of a system when subjected to dynamic loads. CAESAR II provides several methods for analyzing different types of dynamic loadings, which help optimize the trade-off of accuracy vs. computing requirementsthese include harmonic solution, response spectrum method, and time history analysis. The force vs. time profiles of the dynamic loads most often encountered during the design of piping is usually one of three types Random, Impulse and Harmonic. Each of these load profiles has a preferred solution method as well. These profiles, and the load types identified with them, are described below.

Random
With this type of profile, the load changes direction and/or magnitude unpredictably with time, although there may be predominant characteristics within the load profile. Loads with random force/time profiles are best solved using the Spectrum method. Major types of loads with random time profiles are: Wind: The wind possesses kinetic energy by virtue of the velocity and mass of the moving air. If an obstacle is placed in the path of the wind so that the moving air is stopped or is deflected, then all or part of the kinetic energy of the wind is transformed into the potential energy of pressure. A piping system which is located outdoors is usually designed to withstand the maximum wind velocity expected during the system operating life. Dynamic Pressure. The intensity of wind pressure depends on the shape of the obstacle, the angle of incidence of the wind, and the velocity and density of the air. For standard air (density of the air = 0.07651 lb/ft3, temperature = 59 0 F), the expression for the wind dynamic pressure could be adapted from Bernoullis equation for fluid flow as follows Where, p = dynamic pressure, lb/ft2 V = basic wind speed, mi/h CD = drag coefficient, dimensionless For the case of piping under wind loading, then Eq. can be rewritten as Where, F = linear dynamic pressure loading on projected pipe length, lb/ft D = pipe diameter, including insulation, in Static analysis Wind loadings, even though they may have predominant directions and average velocities over a given time, are subject to gusting, i.e., sudden changes in direction and velocity. As the observed time period lengthens, the observed number of changes increases in an unpredictable manner as well, eventually encompassing nearly all directions and a wide range of velocities. The piping wind loading analysis is usually performed by a static method. In the analysis, the wind loading F is modeled as a uniform load acting over the projected length of the pipe, parallel to the direction of the wind. Two horizontal directions of wind loads (north-south and east-west) are included in the analysis. The design loads are based on the worst case of the two directions. For load combination, the wind and the earthquake are assumed to not happen at the same time. A
safety factor, the gust response factor G, should also be considered in the analysis. This factor is used to account for the fluctuating nature of wind and its interaction with structures. To take the account of shape of the body obstructing the fluid a Shape factor of between 0.5 to 0.7 is assumed for piping systems.

Page 1 of 8

Dynamic/Equivalent Static Analysis of Dynamic Events (Notes Prepared by S Koley)

Earthquake: Seismic (earthquake) loadings are caused by the introduction of random motion (accelerations, velocities, and displacements) of the ground and corresponding inertia loads (the mass of the system times the acceleration) into a structure through the structure-to-ground anchorage. The random ground motion is actually the sum of an infinite number of individual harmonic (cyclic) ground motions. Two earthquakes may be similar in terms of predominant direction (along a fault, for example), predominant harmonic frequencies (if certain of the underlying cyclic motions tend to dominate), and maximum ground motion, but their exact behavior at any given time may be quite different and unpredictable. Seismic analysis Static Analysis. Static analysis may be used to evaluate power piping systems in nuclear power plants. It is performed by analyzing a
piping for the statically applied uniform load equivalent to the site-dependent earthquake accelerations in each of the three orthogonal directions. All rigid restraints snubbers supporting the pipe in the direction of the earthquake acceleration included in the analysis. The total seismic effect is obtained by combining the of the three directions. The minimum earthquake force for structures described in ANSI A58.1 one form of static seismic analysis. The code recommends that a lateral force will be assumed to act no concurrently in the direction of each of the axes of the structure in accordance with the formula:

Where V _=lateral seismic force, lb Z =numerical coefficient, dependent upon the earthquake zone I = occupancy importance factor, usually between 1.0 and 1.5 K =horizontal force factor, dependent upon the arrangement of lateral force-resisting elements, usually between 0.67 and 2.50 C =1/(15T1/2) but not to exceed 0.12 T = fundamental period of structure, s S = soil factor, dependent upon the soil profile type, usually between 1.0 and 1.5 W = total dead weight of structure, lb

Dynamic Analysis (Modal Response Spectrum Method). The seismic response spectrum is a plot of the maximum acceleration response of a number of idealized single-degree-of freedom oscillators attached to the floor (structure) with certain damping. These response spectra are based on design response spectra and specified maximum ground accelerations of the plant site. Usually, a series of curves with different damping values for operating and design basis earthquakes for each orthogonal direction are generated. In the modal response spectrum analysis, the piping system is idealized as lumped masses connected by massless elastic members. The lumped masses are carefully located to adequately represent the dynamic properties of the piping system. After the stiffness and mass matrix of the mathematical model are calculated, the natural frequencies of the piping system and corresponding mode shapes for all significant modes of vibration are also determined using the following equation:

Where K = stiffness matrix Wn = natural circular frequency for the nth mode M =mass matrix n = mode shape matrix for the nth mode The modal spectral acceleration taken from the appropriate response spectrum is then used to find the maximum response of each mode:

Where San = Spectral acceleration value for the nth mode D = earthquake direction coefficient tn = transpose of the nth mode shape Mn =generalized mass of the nth mode Yn =generalized coordinate for the nth mode

Page 2 of 8

Dynamic/Equivalent Static Analysis of Dynamic Events (Notes Prepared by S Koley)

Using the maximum generalized coordinate for each mode, the maximum displacements, the effective inertia forces, the effective acceleration, and the internal forces and moments associated with each mode are calculated as follows: Where, Xn =displacement matrix due to nth mode Fn =effective inertia force matrix due to nth mode an =effective acceleration matrix due to nth mode M-1 =the inverse of mass matrix Ln =internal force and moment matrix due to nth mode b =force transformation matrix These modal components are then combined by the appropriate method to obtain the total displacements, accelerations, forces, and moments for each point in the piping system.

Impulse
With this type of profile, the load magnitude ramps up from zero to some value, remains relatively constant for a time, and then ramps down to zero again. For rapid ramping times, this type of profile resembles a rectangle. Loads with impulse force/time profiles are best solved using the Time History or Force Spectrum methods. Major types of loads with impulse time profiles are Relief valve: When system pressure reaches a dangerous level, relief valves are set to open in order to vent fluid and reduce the internal pressure. Venting through the valve causes a jet force to act on the piping system; this force ramps up to its full value, from zero, over the opening time of the valve. The relief valve remains open (and the jet force remains relatively constant) until sufficient fluid is vented to relief the over-pressure situation. The valve then closes, ramping down the jet force over the closing time of the valve.There are two types of pressure relief valve discharge, namely, open discharge and closed discharge. Open Discharge. A typical open discharge is the transient due to discharging of fluid from a header line to the atmosphere through relief valves or safety valves. When the header line pressure reaches the valve set point, the valve opens and decompression waves will travel both upstream and downstream of the valve. This flow transient sets up pressure imbalances along each pipe segment (a straight run of pipe bounded by elbows). The transient forces can be calculated by a computerized method. while the reaction force at the valve exit due to steady-state flow is determined relatively easily. Closed Discharge. In a closed-discharge system, the fluid is transmitted to its terminal receiver through continuous discharge piping. A typical closed discharge is the transient induced by a sudden opening of the relief and safety valves located on top of the pressurizer in a plant. A water seal, which is maintained upstream of each valve to minimize leakage, driven by this high discharge pressure, generates a transient thrust force at each pipe segment. Here some back pressure will act on the valve which will reduce this transient thrust force and due to that additional analysis for discharge force is not carried out in normal practice. Static Analysis. The static method of open discharge described in Appendix II of ASME B31.1 can be summarized as follows: The reaction force F due to steady-state flow following the opening of the valve may be computed by:

where F = reaction force at exit, lbf W = mass flow rate, lbm/s V = exit velocity, ft/s g =gravitational constant= 32.2 lbm.ft/lbf. s2 P =static pressure at exit, psia Pa = atmospheric pressure, psia A =exit area, in2

Page 3 of 8

Dynamic/Equivalent Static Analysis of Dynamic Events (Notes Prepared by S Koley)

The dynamic load factor (DLF) is used to account for the increased load caused by the sudden application of the discharge load. The DLF value will range between 1.1 and 2.0, depending on the time history of the applied load and the natural frequency of the piping. Most of the cases a DLF = 2.0 is used as a conservative approach. DynamicAnalysis (Time History). This analysis technique provides a true simulation of the dynamic system response as a function of time due to the imposed load, and is usually more accurate than the corresponding response spectrum method. The objective of the analysis is to determine if the dynamic response of the system under the time history matches the expected results (i.e. as predicted by dynamic theory). For example: 1. A load applied instantaneously to an un-damped system should yield twice the displacement of the same load applied statically. Additionally the system should cycle between plus and minus this displacement forever. 2. The same load applied to an under-damped system should show close to the same initial displacement, but further cycling decays the displacement magnitude 3. The same load applied to a critically-damped system should produce a response which goes through one cycle before stabilizing at the static displacement. 4. The same load applied to an over-damped system should produce a response which does not cycle at all. Rather the displacement slowly attains the static magnitude. For Open relief valve system, force-Vs-time response is generated. A typical Force-Vs-time profile for a relief valve is given below: Suppose node 5 is the discharge point of a pressure safety valve and the discharge line is open to atmosphere at node 15 after a bend at 10. Now, PSV will pop up, a startup shock wave passes through a single elbow system. Nodes in the piping model are 5, 10, and 15. The system is shown as follows: As the wave starts off between 5 and 10 there is an initial dynamic axial load on the anchor at 5. When the shock wave hits the elbow at 10, the axial load in the 5-10 element balances the initial imbalance at node 5, and there becomes an axial imbalance in the 1015 element. This shock load will be modeled as two completely separate impacts on the piping system The first is the dynamic anchor load at 5. (If 5 is a flexible anchor then this load may cause dynamic displacements of the piping system and 5 will just be subject to the dynamic time history pulse due to the shock.) Assume the anchor at 5 is a flexible vessel nozzle. The second shock load is the unbalanced dynamic pressure load in the 1015 element that exists until the shock reaches the node 15. Friction in the line resisting movement of the shock wave is considerable. In the time the wave leaves the anchor at 5 until it encounters the bend at 10 there is a 50% drop in the pulse strength as shown in the following plot.

Page 4 of 8

Dynamic/Equivalent Static Analysis of Dynamic Events (Notes Prepared by S Koley)

Slug: Most piping systems are designed to handle single-phase fluids (i.e., those which are uniformly liquid or gas). Under certain circumstances, however, the fluid may have multiple phases. For example, slurry systems transport solid materials in liquids, and gases may condense, creating pockets of liquid in otherwise gaseous media. Systems carrying multi-phase fluids are susceptible to slug flow. In slug flow condition a liquid slug can hit the pipe elbow and consequently increase the load on elbow. This load is time dependent. This load lasts till the slug traverses the elbow and then suddenly drops to zero again. The duration of slug is calculated based on the length of slug & stream velocity. This results in short duration impulsive loads on pipes. Thus the piping system is required to be designed for the dynamic loading. In analysis the acceleration effects may be neglected as their effect is assumed to be secondary. It may be assumed that the slug is formed across the full cross-section of the pipe for the maximum impact. This configuration is least probable for vertically downward flow, as no hold-up is possible for accumulation of liquid and eventual formation slug. Hence slug forces in elbows for vertically downward flow lines not to be considered. Static Analysis. For static analysis Force at elbow is calculated and a DLF of 2.0 is multiplied with that. Force on the elbow due to fluid flow to be calculated as, FAXIAL = V2 A (1- Cos ) FORTHOGONAL = V2 A Sin Where, FAXIAL Force on elbow in axial direction along with direction of fluid flow. (N) FORTHOGONAL Force on elbow in outwards direction of the bend at the plane of bends (N) - Density of liquid (Kg/ m3) A Internal cross-sectional area of pipe in (m2) V Fluid velocity (m/Sec) - Angle of bends (Degree)

Here it is assumed that slug is 100% through out the cross section. The actual force can be assumed lower depending on the process data of percentage of slug formation through the cross section.

Page 5 of 8

Dynamic/Equivalent Static Analysis of Dynamic Events (Notes Prepared by S Koley)

Dynamic analysis( Force Spectrum Method): To carry out dynamic analysis Force-Vs-time profile is generated. To generate node-wise time Vs force spectrum it is assumed that the elbow is subjected to force due to liquid slug and drops to a smaller value based on the density of gas, after some duration after the liquid traverses the elbow. The profile depends on some parameters as below: a) b) c) d) Slug length (m) Gas length (m) Slug duration (millis) Slug periodicity(millis) = liquid volume fraction length (m) (elbow to elbow) = vapor volume fraction length (m) (elbow to elbow) = slug length(m)1000 / fluid velocity(m/s) = {slug length(m) + gas length(m)}1000 / fluid velocity(m/s)

A typical Force-Vs-time profile for a Slug loaded system is given below:

Force

Slug force

Time 1 2

Slug periodicity (Tp)

Slug duration(Td)

1:- time to taken by the force down to zero. 2:- time taken by the force to get its peak value.

Page 6 of 8

Dynamic/Equivalent Static Analysis of Dynamic Events (Notes Prepared by S Koley)

Fluid Hammer: When the flow of fluid through a system is suddenly halted at one point, through valve closure or a pump trip, the fluid in the remainder of the system cannot be stopped instantaneously as well. As fluid continues to flow into the area of stoppage (upstream of the valve or pump), the fluid compresses, causing a high pressure situation at that point. Likewise, on the other side of the restriction, the fluid moves away from the stoppage point, creating a low pressure (vacuum) situation at that location. Fluid at the next elbow or closure along the pipeline is still at the original operating pressure, resulting in an unbalanced pressure force acting on the valve seat or the elbow. The fluid continues to flow, compressing (or decompressing) fluid further away from the point of flow stoppage, thus causing the leading edge of the pressure pulse to move through the line. As the pulse moves past the first elbow, the pressure is now equalized at each end of the pipe run, leading to a balanced (i.e., zero) pressure load on the first pipe leg. However the unbalanced pressure, by passing the elbow, has now shifted to the second leg. The unbalanced pressure load will continue to rise and fall in sequential legs as the pressure pulse travels back to the source (or forward to the sink). The ramp up time of the profile roughly coincides with the elapsed time from full flow to low flow, such as the closing time of the valve or trip time of the pump. Since the leading edge of the pressure pulse is not expected to change as the pulse travels through the system, the ramp down time is the same. The duration of the load from initiation through the beginning of the down ramp is equal to the time required for the pressure pulse to travel the length of the pipe leg. These unbalanced forces are usually generated through a two-step computerized calculation. The fluid system is modeled as an assemblage of control volumes (e.g., piping volumes or steam generator) interconnected by junctions (e.g., valves, pump, or break). Piping fluid flow data, such as flow area, friction losses, valve closing-opening time, feed pump characteristics, or break characteristics, together with fluid initial conditions (flow rate, pressure, temperature, and mixture quality) are supplied as input to a thermal hydraulic finite difference computer program. Using this input information and a built-in steam table (fluid thermodynamic state), the first step solves the three equations of conservation (mass, momentum, and energy) at each time step for fluid properties such as pressure, velocity, internal energy, and mixture quality. A typical stop valve closure time history and its associated dynamic pressure time history are shown in Fig. B4.13. The second step utilizes a postprocessor. This postprocessor then accepts the output information from the first step and computes the unbalanced forces in piping segments by applying the momentum theorem. Static Analysis. Static analysis is simple and saves computer time. It is used when the unbalanced forces are small and the total transient time is long. In the analysis, the peak values of the time-history fluid forcing functions at pipe segments are applied statically to the piping. The piping stress, deflections, and support nozzle loads are then calculated by the computer program. To obtain a conservative result for the static analysis, care must be taken in applying a proper dynamic load factor to the unbalanced forces. To calculate the unbalanced force first magnitude of the pressure wave is calculated as :

dP = c dV
where: dP - the pressure rise due to the pumps instantaneous stopping - the fluid density c - the speed of sound in the fluid. dV - the change in velocity of the fluid The speed of sound in the fluid can be estimated from c = [Ef / (+ (Ef / E) (d/t) )]1/2 Where: Ef - is the bulk modulus of the fluid E - is the modulus of elasticity of the pipe d - is the pipe mean diameter t - is the pipe wall thickness Unbalanced load on the piping system due to the time it takes the pressure wave to pass successive elbow-elbow pairs can be computed from:

Fm =Maximum unbalanced Force = dp * Area

Design load in pipe link (between elbow to elbow pair in axial direction) = Fd= (DLF)*(L/Lc)*Fm

Page 7 of 8

Dynamic/Equivalent Static Analysis of Dynamic Events (Notes Prepared by S Koley)

Dynamic Analysis (Force Spectrum Method): On the pump or valve discharge side the maximum magnitude of the pressure wave is the difference between the fluid vapor pressure and the line pressure and on the supply side a positive pressure wave moves away from the pump at the speed of sound in the fluid. The magnitude of the pressure wave is equal to the sum of the suction side pressure and dP. On the discharge side a negative pressure wave moves away from the pump at the speed of sound in the fluid. The maximum magnitude of this negative pressure wave is the difference between the pump discharge pressure and the fluid vapor pressure. Once the pump shuts down, the pressure at the discharge begins to drop. The momentum of the fluid in the downstream piping draws the discharge pressure down. If the fluid reaches its vapor pressure the fluid adjacent to the pump flashes. As the negative pressure wave moves away from the pump these vapor bubbles collapse instantly. This local vapor implosion can cause extremely high pressure pulses. In addition, there may be a fluid backflow created due to the rapid drop in pressure. In this case the backflow slap at the idle pump can be accentuated by the collapse of created vapor bubbles, resulting in an extremely large downstream water hammer loading. Water hammer loadings will cycle to some extent. The pressure wave passes through the system once at full strength. Reflections of the wave may then cause secondary pressure transients. Without a transient fluid simulation or field data the usual procedure is to assume one or two significant passes of the pressure wave. The time history waveform for both types of water hammer pulses is shown as follows:

Pod - Discharge pressure Ps - Source (tank or static) pressure Pos - Suction pressure (while running) dp - Pressure fluctuation due to the instantaneous stoppage of flow through the pump Pv - Liquid vapor pressure at flow temperature

In the dynamic analysis, the piping system is idealized as a mathematical model consisting of lumped masses connected by weightless elastic members. These lumped masses are carefully located to adequately represent the dynamic characteristics of the piping system. For computer programs utilizing the modal-superposition method, enough modes (or appropriate cutoff frequency) should be specified in the computer input such that the dynamic response of the piping system can be truly represented. The timehistory unbalanced forces are applied to all pipe segments. Where critical piping is concerned or where the maximum loads on snubbers and restraints is to be computed, the independent effect of a single pass of the pressure wave should be analyzed for each elbow-elbow pair in the model. A separate force spectrum load set is defined for the elbow with the highest pressure as the wave passes between the elbow-elbow pair. The direction of the applied force is away from the elbow-elbow pair. An individual dynamic load case is run for each separate force set, combinations of different force sets are usually not run. To define a force spectrum some parameters are calculated as below: Critical length: Rise time (Time to reach maximum force) Dwell time Falling tine (Time to fall maximum force) Total time duration Maximum load in pipe link Lc = C * tc Where tc= Effective valve closure time tr = tc*(L/Lc), for L<=Lc Where L= length between Elbow to Elbow pair. (If L >Lc, then further water hammer effect need not to be incorporated) td = L/C tf = tr t = tr + td + tf F = (L/Lc)*Fm

Page 8 of 8