Water Hammer and surge analysis (Guides, specification, procedures, ... )

[energeng]

Dears,

I am currently working on a project that requires the study of water hammer/surge analysis of several lines of about 4000 meters in length, a typical facility for feeding finished hydrocarbons to ships from storage tanks.
Our processes department is working on the hydraulic study of the water hammer/surge analysis using the AFT Impulse software, which can output the forces generated in the pipes. In this study, the calculation model is simplified and the entire route of the pipe is not defined in detail, simplifying some changes of direction, loops, etc., due to the fact that we do not have the definitive final layout of the pipes.
There are my questions:
1. How can I use the force values ​​obtained in AFT Impulse based on a simplified model and apply these forces in a stress calculation model where the model does have all the details?
2. Does anyone know if the large engineering companies have specifications describing how to proceed with the water hammer/surge analysis calculation from the hydraulic point of view and the application of forces in pipe stress studies?
3. Can anyone recommend a bibliography on this topic applied to refining facilities where the fluids are hydrocarbons and not water?

Thank you very much in advance.

 

[Alex Matveeff]

I can suggest this video
Forces from hydraulic analysis software should be transferred into pipe stress analysis software and applied at all bends and tees. At each moment of time you have the specific distribution of the forces in all bends and tees. The forces can be in same direction or in opposite directions, because waves traveling both ways. And if you perform static analysis, need to find the most dangerous moments of time. If you perform time history analysis, then you can analyze the whole system and apply time-force diagram at each bend and tee.

I'm the PASS/START-PROF Pipe Stress Analysis Software Developer

 

[MJCronin]

1. How can I use the force values ​​obtained in AFT Impulse based on a simplified model and apply these forces in a stress calculation model where the model does have all the details?

Incorporate the AFT forces generated into a CAESAR-II or equivalent piping stress anaysis program. Compare the transient stresses generated to Piping Code allowables. You will need an experienced stress analyst... You cannot learn this "on-the-fly:

2. Does anyone know if the large engineering companies have specifications describing how to proceed with the water hammer/surge analysis calculation from the hydraulic point of view and the application of forces in pipe stress studies?

Yes, the equations contained in ASME B31.1 and B31.3 contain these requirements


3. Can anyone recommend a bibliography on this topic applied to refining facilities where the fluids are hydrocarbons and not water?

You will have to do a bit of digging for this... lots of info available on the internet and eng-tips .... Why don't you contact the AFT people and ask them that question. They would have not problem answering questions from their licesenced software customers !


MJCronin
Sr. Process Engineer

 

[shvet]

3. Can anyone recommend a bibliography on this topic applied to refining facilities where the fluids are hydrocarbons and not water?

https://www.amazon.com/Fluid-Transients-Benjamin-Wylie/dp/0961014407

and para 5.9 BP's Tank Farms and (Un)loading Operations booklet for illustrations

Note that allowable overstress for pipe metal is referred to a slow rise of pressure&temperature (minutes or even hours in some cases) instead of liquid surge which is referred to milliseconds. This fact means that there is a chance that local overstress will be much higher than that calculated by conventional algorithms because those has been created for phenomenon of another nature. The widespread explanation of this is next:
imagine - you slowly stand up to a wooden crate or you jump to that - in second case it will be broken instead of your weight was equal in both cases. It is because stress had no time to spread equally to all elements of a crate.

 

[Snickster]

I have done such stress analysis by considering imbalance forces at pipe bends, tees, change in directions as equivalent static forces. I received similar surge analysis from process engineers that showed graphs of the increase in pressure along the pipeline versus time based on a generalized model. With that information I looked at the maximum increase in pressure in the lines and converted to a force by multiplying the pressure increase times the area of the pipe. I then applied the force at each elbow/change in direction. Since the force is only felt at one elbow at a time as the pressure wave travels down the pipe, I ran two or three programs to put the forces simulataneously on elbows not adjacent to each other so that the force on one elbow would not have an affect on the effects of the force applied simultaneously to ther the upstream or downstream piping. The option would be to run a separate calculation for each elbow with force only at that elbow but that would be very time consuming.

The way I ultimately handled the forces was to put an anchor or line stop in the center of each straight run to absorb the surge force while allowing that run of pipe to grow thermally. I did not hear of any issues with this design

If I recall this is the recommended method for static analysis in the Caesar piping program technical literature, which is the program which I used.

 

[GD2]

Just to make a point here - it will be a dynamic stress analysis.

GDD
Canada

 

[1503-44]

There will probably only be a few, if any, places where such loads may be concerning, if the maximum fluid velocity is kept below 10fps. 3.3 m/s. You should not have any higher velocity in the main pipeline. If you keep velocity low, you should not have any excessive loads in pipe, nor at supports. Any faster and you might potentially damage the ship's manifold. You really, really do not want to do that.

I would look at the potential forces (from the hydraulic transient pressure analysis.)
in 4 generic locations,

1 the immediate vicinity of your fast closing ESD valves, where max hydraulic transient pressure will be seen
2 at relief and surge valves
3 any lines to/from a surge/pressure relief tank
4 a generic 1 or 2 90°bend configuration, for which I would use this max for all support designs.

I would use an equivalent static design factor of 2 for dynamic loads applied to supports. I would never do a full "dynamic force design" for any preliminary configuration and probably, if ever, only at or near those specific locations named above.

Check the pipeline for Maximum operating plus Transient pressure, adding the transient pressure of #1 to the pipeline's maximum operating pressure at any point. Transient pressure rise must not exceed pipeline design pressure + 10% if higher, the options are
decrease ESD valve close speed,
or increase pipe diameter, or
add surge valve and tank.


Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[0707]

Maybe it helps

https://www.youtube.com/watch?v=xoLmVFAFjn4

https://www.armstronginternational.com/zh-hans/node/370588

luis

 

[davefitz]

A simple analysis can be found here: <

https://www.valvemagazine.com/articles/water-hammer>.

The hydrocarbon density and soundpseed is different than that of water, so those parameters can be easily modified in the analysis.
The true magnitude of the pressure pulse is also impacted by the flexibility of the piping material( E , youngs modulus), and the potential to fail the pipe is also limited by the energy required to fail the pipe, and not simply due to the forces without regard to energy .
The normal methods of reducing the magnitude of the force is to reduce the speed of closure of the valve ( mechanical measures are more reliable than control logic) and/or to add a bladder ( snubber) with compressible gas to dampen the pressure pulse.

"...when logic, and proportion, have fallen, sloppy dead..." Grace Slick