Can I measure Water Hammer? 1

[DrDarrell]

Is Water Hammer the resultant "Noise" from the shock of acceleration of water? If so, Water Hammer is nothing more than vibration.

Is there a point when Water Hammer is pressent and not detectable by the human ear? If so, would there be any damaging affect because of the seemingly small amout of Water Hammer?

Is there a method of measuring Water Hammer?

Darrell

 

[vpl]

Is Water Hammer the resultant "Noise" from the shock of acceleration of water? If so, Water Hammer is nothing more than vibration

No, it's not the resultant "noise." It is also much more than vibration. A water hammer is a "layman's term" for a pressure transient, often severe, usually caused by fluid rapidly changing direction in a non-full pipe. Just accelerating the fluid doesn't do it, you also have to rapidly change the direction of the fluid. Having a partially steam filled line with a rapidly increasing pressure also adds to problems.

Is there a point when Water Hammer is pressent and not detectable by the human ear? If so, would there be any damaging affect because of the seemingly small amout of Water Hammer?

Yes a water hammer can be present and not detected by the human ear IF THERE IS NOBODY THERE TO HEAR IT. It depends on how close people are to the location of the water hammer and what intervening material (distance, amount of concrete, etc.) as to whether someone will actually "hear" it. And yes there could be damage, even if no one hears it. You can often "see" the after-effects of a water hammer by the line of dust on the floor that was shaken off the pipe. However, the major concern is with the pipe supports that might be damaged.

Is there a method of measuring Water Hammer?

Generally, you want to design against it, because of the very large pressure transients which can occur. However, if you think you have it going on, then some high-speed pressure gauges and accelerometers would allow you to measure it.

Why do you think it is water hammer and not cavitation?

Patricia Lougheed

Please see FAQ731-376 for tips on how to make the best use of the Eng-Tips Forums.

 

[BigInch]

Its about as much like vibration as an earthquake is. Water hammer is often not heard at all. What is typically "heard" is the effect of a pipe moving into a support or slamming up against a wall, anchor point or other supporting structure, if not actally coming lose from its mountings, or breaking up a flange joint completely, if it was severe enough. If the support structure was not there, the pipe might even have moved relatively silently, and probably moved a lot more. The event triggering the high pressure transient in one place could have originated 20 miles or more away from the point of observation. They are quite common on long pipelines with high fluid velocities that experience rapid +/- accelerations.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[sailoday28]

BigInch (Petroleum) Velocity in long liquid lines does not have to be high to cause water hammer problems. Consider what the unbalanced forces are due to changing the velocity by say 5-6fps.

regards

 

[MortenA]

5-6 fps is also a fairly high velocity in a pipeline i would say. I know you can go double that velocity but it not a requirement.

Best regards

Morten

 

[sailoday28]

MortenA (Petroleum)5-6 fps may be high, depending upon what system the flow is in.
At any rate, Try 1-2 fps change in velocity in a long pipe and I believe you will see the resulting unbalanced forces are relatively high.
Regards

 

[MortenA]

Sailoday28: I agree.

BTW going back to the original question: I believe that you can get both electronic pressure gauges that are fast enough to catch the peak pressure and mechanical gauges with "stickyness" so that the gauge wont drop back but stay at the max pressure.

Best regards

Morten

 

[sailoday28]

MortenA (Petroleum)While fast response devices are available and can be used, why not design to minimize the potential impact of "fast/slow transients"?
What seems to be a slow closing valve is relative. In a long pipe, such a valve can be considered fast closing.
For example, neglecting friction and flow velocity, and say 2 miles of pipe with a sound speed of 3000fps- under 3 seconds closing time is fast.
Regards

 

[budt]

When I suspect pressure spikes in a hydraulic systems I put a pressure gauge in the suspect line with a check valve in front of it, free flow to the gauge. When a pressure spike occurs the gauge will see it and the check valve will hold it. It usually takes a couple of hours to finally trap the spike pressure or at least near its maximum.


Bud Trinkel CFPE
HYDRA-PNEU CONSULTING, INC.
fluidpower1 @ hotmail.com

http://www.fluidpower1.us

 

[sailoday28]

budt (Industrial)I follow your reasoning, but the gauge reading is now dependent upon the response time of the check valve.

regards

 

[DrDarrell]

I would like to thank all of you for the informative conversation.

I think it is time to give a little background.

I am building Valve/Actuator packages and I would like to design into the actuator a closing profile which slows the closing time toward the 100% close position. I have a fellow associate who feels that it isn't necessary.

The velocity in an 8” pipe with the valve fully open is approximately 20 ft/sec^, and the length of pipe varies from 50 ft to 500 ft.

From your comments, it appears that I should press forward with a closing time profile.

Of coarse I would like to eliminate the complexity of the closing time profile, so If I were to build a prototype and test it for water hammer, and found that there isn’t a need to change the speed of the valve’s closing, everyone would be pleased. But, I would need a definitive test method to determine the amount of water hammer.

 

[budt]

sailoday28 wrote:

"budt (Industrial)I follow your reasoning, but the gauge reading is now dependent upon the response time of the check valve."

I know its not perfect but gets close enough with available inexpensive equipment. It is also a way to get info without the machine operator or others knowing you are checking.

Try it, you might like it.


Bud Trinkel CFPE
HYDRA-PNEU CONSULTING, INC.
fluidpower1 @ hotmail.com

http://www.fluidpower1.us

 

[BigInch]

For pipelines 5-6 fps is normal, typically 3 minimum and 10 maximum.

A 3 second total closing time on a pipeline is speed of light, 1 sec/inch being more reasonable, unless its an ESD valve. Great care is needed in selecting valves and actuators when velocities are above 5 fps, as many valves effectively open in the first few seconds, or close in the few seconds, even though their total opening time and closing times may be only 1 sec/in diameter due to poor Cv vs actuator position curves. One must be carefully in slowing down the mainline velocity well before a valve gets close to its shutoff point.

If there is a snap acting gas operated ESD valve in your pipelines, you must do a transient analysis.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[BigInch]

This type-ahead (of brain) is killing me. Make that 1min/1 ft diameter (5 sec/in-D)

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[sailoday28]

BigInch (Petroleum)"A 3 second total closing time on a pipeline is speed of light,

Depends on time for inital reflection and responding reflection to return to point of origin.
For a short lengths, 3 seconds may be a slow closure.


Regards

 

[BigInch]

"Pipeline" in my book is well over 10,000 feet long.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[MortenA]

I agree biginch - to me this sounds more like "piping".

The Jucowsky equation give a conservative estimate:

h(wh)=a*dV/gc

where

a=sqrt(1/((den/gc*(1/K+D/(bE))

Keep it strictly metric and it all works out (head in metre liquid column)

gc:gravitational acceleration
V:velocity
K:bulk modulus of elasticity of THE FLUID
E:bulk modulus of elasticity of THE PIPE

See e.g. Perry for reference

Best regards

Morten

 

[MortenA]

oh yes: The (origina) question was really just about measuring water hammer pressure - not preventing or predicting it...

Best regards

Morten

 

[BigInch]

In the above, a 3 second transit time would probably not be acted upon at all if the valves had a closure time of 25 seconds, so the max transient pressure for that event would (hopefully) have to had been below the max design pressure, which is why a transient anlaysis should have been done beforehand to verify that indeed it would have been.

Caution, the Joukowsky equation is for a straight single level pipe filled with one product and closed on one end, so it does not tell the whole truth when applied to an operating pipeline that typically will cross one or two mountain ranges and river valleys, maybe with slack flows, wide variations in product density (due to +2800 meter differential elevations), or perhaps 10 products in the line at the same time, have open connections into other systems, pumps starting, stopping, check valves chattering, and control valves changing %Open at several (or many) points along its length and have multiple reflected (algebracially additive) transient waves moving back and forth between stations.





BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[MortenA]

Biginch i always considered Joukowsky to be conservative. Is that not the case. Can columseparation and following cavitation cause even higher surges?

Best regards

Morten