I have an application which uses a large combined Pressure Reduction Valve (PRV) and desuperheater (12" steam inlet and 30" steam outlet). Are there significant forces resulting from the steam pressure and temperature reduction involved when using these valves? If so, are these forces to be modelled as applied externally to the system or are they absorbed internally in the piping system and thus, do not appear in the CAESAR model? I suspect that these forces are taken up internally but I would like get some input from other CAESAR users.

Also, are these large combined PRV/desuperheater valves supported independently or are they just welded into the piping system with supports being attached to the pipes (like any typical control valve application)? Should these valves be anchored on one side? What has been the experience of others out there?

Hi Bill,

The PRDS valves (AKA turbine bypass valves) that I am familiar with are in-line valves that are supported via the pipe. Of course it is important to know accurately what the total weight of the unit is so that hangers can be designed. Remember to include the weight of the cooling water piping that is transferred to the valve.

With these valves, the "bore thrust" forces due to internal pressure are self balancing. In a piece of pipe without a valve, the bore thrusts impinge upon components such as elbows and since the thrust is equal and opposite, the forces balance AS LONG AS THERE IS STRUCTURAL CONTINUITY (continuous pipe wall). If there is a structural discontinuity (e.g., an expansion joint interrupts the continuous pipe wall), there would be a pair of opposite forces to be addressed by design. With the PRDS valve in-line, there is a "high pressure (inlet) side and a "low pressure" outlet side but since both the HP and LP impinge upon the PRDS valve and some other component in the opposite directions the pressure remains balanced (discounting the possible structural discontinuity).

It is also important to know if the valve fails open or fails closed. What would happen if the flow of coolant were to be interrupted? Would the steam flow be shut down by the valve or would the downstream piping be subjected to higher than design pressure and temperature (expansion)? Just something else to think about.

DeZurik, Copes-Vulcan has an interesting Bulletin (Bulletin 1163 in PDF form) on the internet at:

http://www.dezurik.com/Literature_PDF/1163.pdf

What do the rest of you folks think?

Best regards, John.

A consideration for this type of installation is the change in fluid momentum due to opening or closing of the valve.

If the valve is in a service where opening and closing stroke times are long, then the forces are unlikely to be of concern.

If the valve has a rapid opening time (for example turbine bypass valve service), then the forces generated may be significant - you will need to look at the valve stroke time and the consequent change in mass flow from zero to full flow over this time.

Stroke times for hydraulic valves can be in the range of 0.5 - 1.0 s, whilst pneumatic valves can take from 2.0-5.0s.

Also, in order to keep nozzle loads to a minimum, turbine piping systems are often 'hung' and only guided where necessary (the local seismic design requirements often govern here), therefore can be sensitive to imposed forces such as these changes in momentum, or pressure relief reactions.

Jhon:
You posted:
"In a piece of pipe without a valve, the bore thrusts impinge upon components such as elbows and since the thrust is equal and opposite, the forces balance AS LONG AS THERE IS STRUCTURAL CONTINUITY (continuous pipe wall)"

1)Does this mean that it is not reqiured to use a thrust block at the elbow(high flow rate) ?
2)Is pressure thrust in a continous pipe is covered by(PD/4t)
3) if there is change in momentum (like a Wye with different flows in and out)does the same criteria applies?

Hello Akber and Ross,

Before I get to your questions, Akber, I want to underline Ross's excellent observations. You must be aware of the forces due to flow of fluid and design to accommodate these.

There are several forces involved here and perhaps it would be useful to consider them individually:

Internal pressure is equal in all directions. For the most part the forces impinge upon the inside of the pipe and they cause compressive forces on the inside diameter and tensile forces on the outside diameter (both of theses stresses are circumferential and they are considered by B31 Codes in the process of determining the pipe wall thickness (P * D / 2 * t)) - the continuous wall of the pipe allow these equal and opposite forces to counteract each other.

It should be noted that some types of expansion joints (e.g., bellows) also have an area that is greater than the internal bore area. A bellows has convolutions that include an internal "annular space". Normally all of the convolutions will have opposite and equal areas to counteract the longitudinal pressure - EXCEPT the two that are on opposite ends of the bellows. This feature of these types of expansion joints results in an additional longitudinal force that must be added to the "bore thrust" when the system is designed.

The longitudinal forces eventually impinge upon something that is aligned with (perpendicular to) the pipe inside bore - changes in direction come to mind. When these internal areas of impingement (e.g., inside pipe wall at bends) are connected together by a continuous pipe wall, the load path is continuous, allowing the opposite and equal forces to counteract each other - the stresses in the pipe wall caused by this are longitudinal stresses and they are addressed in the B31 Codes as part of the "additive longitudinal stresses due to weight (bending) and internal pressure (P * D / 4 * t)) " (i.e., "sustained longitudinal stresses").

Yet another force results from flowing fluids. The flow of fluids moving through the pipe must be "turned" by the pipe wall (change in the direction of flow) when a bend is encountered. This impingement of flowing fluid (energy) causes a force to be applied to the pipe that also must be counteracted. Obviously, the magnitude of this force is a function of the mass that is having its direction changed by the bend. If the force is great, it MAY require that an "anchor block" or similar restraint be used to transfer the fluid flow force from the pipe to surrounding soil or adjacent structure.

Further, some perturbations in flow will cause discontinuous or impact type applications of forces to the pipe ID at bends (and other changes in direction). As an example, if liquids are allowed to accumulate in a piping system that normally carries a gas or vapor, the "puddle" of liquid MAY be "picked-up" by the flowing medium and "thrown" downstream as a “slug” (this is called a “slug flow event”) until the "puddle" IMPACTS upon a pipe wall (bend). This is a very destructive event. AND, if that were not enough, the pipe designer must be aware of "steam hammer" and "water hammer" which occurs as a result of acoustic waves being propagated through the process medium in alternating compression and rarefication waves. These also result in impact loadings. “Hammer” events may be caused by product phase changes, “bubble collapse”, fast opening (or closing) valves or other such unpleasantness. These types of events are analyzed as a function of time and they have been well discussed in previous threads on this board.

Now, having said all of that lets look at you questions:

1)Does this mean that it is not required to use a thrust block at the elbow(high flow rate)?

NO.
Keep in mind what Ross is saying above. The flow of product through the pipe will cause a force to be applied that is a function of the mass of the flowing fluid and its velocity. This is a different force that that caused by internal pressure.

2)Is longitudinal pressure thrust in a continuous pipe covered by (PD/4t)

YES, in so far as the pipe stresses are concerned. It is part of the B31 Code "sustained stresses". The circumferential stresses caused by internal pressure are considered in the wall thickness calculation.

3) If there is change in momentum (like a Wye with different flows in and out)does the same criteria apply?

Yes, the same design issues are presented but remember that the "bore thrust" due to internal pressure is a function of the inside diameter and if the "Y" changes the diameter the designer must address this change in force - keeping the pressure-caused forces "balanced" is very important. This is VERY well explained and illustrated (with good diagrammatic figures) in the Standards of the Expansion Joint Manufacturers (EJMA).

The forces applied to changes in direction due to flow of product through the pipe must be countered by including thrust blocks, struts, restraints or by other design considerations.

I hope some of this will be of value to you. Good luck with your projects.

Best regards, John.

Thank you John for your helpful notes , now the question is what about diplacement resulting from internal pressure.Do they need to be considered in anchor forces.And if pipe is not anchored are the stresses resulting from these displacements covered by (PD/4t)

Regards

Hello Akber,

The loadings that occur at anchors, supports, hangers and any other piping system restraints MUST be calculated and they MUST be addressed by the system design. This loading is one of the "Design Conditions" that the Code requires the design to address.

Of course, if the design can control the displacement close to the point where the the pressure impinges upon the pipe wall, it may be possible to prevent a large moment at a more remote point like an anchor.

The PD / 4t term ONLY addresses the longitudinal tensile stresses in the pipe wall as a result of the longitudinal pressure (bore thrust). It does not address bending moments (and their stresses) caused by any pressure related displacement. The stresses caused by bending moments are addressed AT THE POINT WHERE THEY OCCUR by the second term in the "sustained stress" equation: iM / Z).

Regards, John.