Water Hammer -

[Bilodeau]

Hi,

I have calculated the water hammer effect, and I have 500 psi (delta P), more my normal operation at the inlet valve (550 psig), my pressure can increase to 950 psig in my worst case. Also, I have calculated my closing time and I got 7 sec.

(with Joukowski's equations).


If my understanding is good, even if my closing time is longer than 7 sec, I will have the water hammer effect, but I reduce the pression reached.

MY QUESTIONS : How to calculate the pressure with different closing time, because in my case, I'm looking to put a closing time to respect my pipe spec.

Thank you,

Ken

I have already read :

http://www.eng-tips.com/viewthread.cfm?qid=168831

 

[Bilodeau]

500+550= 1 050 psig.

Drugs are bad.

 

[sailoday28]

Is the fluid cold water? Are you calculating the pressure rise from instant closure, (dp)=rho*c*change in velocity ?

Also what is rho*c, initial velocity and length of piping?

 

[Bilodeau]

Gasoline and diesel.

 

[Bilodeau]

10 in pipre

4 000 m = length

700 m³/hr = flow.

 

[BigInch]

Joukowski's equation with a rigid fluid column gives the answer for pressure rise that takes place during the time it takes for the fluid column to come to a halt. That is the time it takes for thw pressure waves to move from the valve to the ends of the pipeline. Those times are of course different if the valve is not at the center length of the pipeline. Assuming you have a valve at the downstream end of the pipeline and close it, the pressure will build to a maximum just as the pressure wave reaches the beginning of the pipeline. That is the time = Pipeline_Distance_toInlet / Sonic_Velocity. A pressure decrease occurs on the downstream side of the valve of equal magnitude, but Operating Pressure - Surge Pressure must be >= Vapor Pressure. But, that pressure decrease occurs during the time it takes for a pressure wave to travel from the valve to the outlet end of the pipeline.

Solve Joukowski's equation for velocity, given the maximum pressure increase you want to have. Consider that to be a maximum change in velocity that limits the surge pressure increase to your maximum desired value. Do not change the fluid's velocity in the pipeline by more than that amount during the time it takes for the pressure wave to arrive at the end of the pipeline.

For best response, take care of the valve selection by choosing a valve with a linear relationship of change in velocity (or flowrate) across the valve to the valve's closing time. That is usually accomplished by selecting a valve with a linear relationship between flowrate and %Open and using an actuator with a linear relationship between its position and travel time. Ball and gate valves do not have linear characteristics of change in flow rate vs Position and almost all of the change in flowrate will occur during the last 20% of the closing operation. Therefore if you are using a ball or gate or similar flow pattern valve, and you set a closing time for the actuator, use 20% of that actuator travel time for calculation and comparison with the answer you get from the Joukowski equation.

http://virtualpipeline.spaces.msn.com

 

[BigInch]

BTW, Your pipeline fluid velocity is pretty high, so beware of erosion, and 7 seconds seems slow for a 4000 m long pipeline with gasoline or diesel. I got a surge of 734 psi with diesel, but I didn't calculate V_sonic; estimated it at 2500 fps. Is your 1875 fps sonic velocity really that slow?

http://virtualpipeline.spaces.msn.com

 

[sailoday28]

For the best response,
1 You should use a water hammer program -one that uses the method of characteristics.
2-Obtain from the valve vendor(S)/valve operator, the valve operating characteristics-flow area vs. valve position or better yet valve Cv vs position. AND valve position vs time.

With the above AND the boundary conditions for the system you should be able to select the proper valve/operator to limit you pressure surge.

 

[Bilodeau]

Thank you for reply.

I'm sorry, I have forgotten to mention that is a ON/OFF valve...

 

[BigInch]

I knew that.

http://virtualpipeline.spaces.msn.com

 

[davefitz]

If the 7 sec period causes other issues, is it feasible to install a surge dampener, ie, a gas filled bladder attached to the pipig at a tee ?

 

[BigInch]

See

http://www.spxprocessequipment.com/sites/mandjvalve/literature/MJ-1687_surgerelief_mj.pdf

http://virtualpipeline.spaces.msn.com

 

[sailoday28]

Have you considered the elastic properties of the piping when coming up with an equivalent wave speed and therefore closure time?

 

[25362]

Gases, if present, may markedly reduce sonic velocities.
Also a visit to

http://www.flowguardusa.com/docs/Article_Surge.pdf

may be of some help.

 

[sailoday28]

25362 (Chemical)Your point is well taken. It would be interesting to find the two phase relationship as pressure increases.
Regards