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

 

[MortenA]

dcasto without having looked into your system i would say that biginch has a point: Surge/water hammer is a phenomena with a much higher trancient resolution than 3-10 seconds. The peak pressure may only last for less than a second. Im sure that i dont have to explain Shannon sampling theory (and being a process engineer im surely not the right person) but i do understand that much: You have to have double the samling frequency compared to the system that you cant to measure (excuse my poor english - cant put it any better).

Best regards

Morten

 

[BigInch]

It might be your recorded peaks were chopped-off versions of the true wave that may have actually missed the high point. Depends on the shape of the wave, speed ...

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[dcasto]

We may not have caught the peak pressure as they went by the transmitters, but there is no denying that we saw the wave. I have set up an ocilliscope on a very sensitive pressure transmitter to read the pulsations from a PD pump trying to prove if the pulsation dampener was working. We could see the peak pressures that the transitters could not record.

This may be off the thread, but there is a sonic wave sent through the system whenever a valve is opened or closed. The pipeline I refered to above had some test run on this theory by NASA (the line went through Friendswood) and they used highspeed pressure recorders to sense when a valve opened. Then by listening along the pipeline at different stations and using time differential they could pinpoint the valve opening location. The application was leak detection. the opening of the valve was a leak.

It worked. The hold back then was cost of the system. especially the time factor, all the locations had to have exact time to "triangulate". Today I'd bet they could use GPS to get the time very accurate.

 

[BigInch]

Exxon? Nobody's denying you can't see them, hear them, or measure them. Just a question of how well sometimes. Other times, there's no question at all. They can make a rather loud bang as they pass through a meter or marine terminal loading station.

Actually there have been a lot of ultrasonic leak detection packages on the market for quite awhile now. Its claimed they work rather well.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[dcasto]

In 1993 for OxyChem, now Equistar,the lines are ethylene. We tried various leak detection models and tyipcal ultrasonic and they didn't work. We were installing a LIC pressure point analysis type system with a heavy front end processing for SPC of the measurement instruments when I left the company in 1995. The package even compensated for the hieght of fluid difference between the instruments and the line. The density ranged from 23 to 8 lbs/ft^3.

The NASA stuff looked promising, but I never knew what happened to it.

 

[BigInch]

From what I hear, there are a lot of fantastic ultrasonic claims that just don't hold water after installation.

The only thing I know that works is a hot real-time transient model comparison to second-by-second data for large leak alarms and detecting small leaks by long term differential volume summation, and following the big blue flies to a gas leak.

LIC is one of the good ones.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[stanier]

My experience has been that you need fast response high accuracy pressure transducers connected to an oscilliscope or a data logger that is capable of logging all the pressure peaks otherwise you will miss the transient. Generally SCADA systems just dont cut it for accuracy or response times.

I demonstrated this to an international oil company when they had transients as shown by the pipes jumping all over the place yet their SCADA didnt pick up the event.

Go to

http://www.ventomat.com

and read the articles there on how they developed their air release valve. Ventomat didnt understand the pressures being experienced unitl they used the right gear.

Geoffrey D Stone FIMechE C.Eng;FIEAust CP Eng

http://www.waterhammer.bigblog.com.au

 

[stanier]

In response to MortenA,

Joukowsky does not give you the worst conditions. For a start it doesnt predict column separation and the reforming.

Prof ARD Thorley inhis book describes this phenomenon. Also trey Walters in the manual for AFT's Impulse explains why Joukowsky cannot be relied upon.

This is one of the common mistakes in the engineering world. Ignorance is bliss for some as they do a quick calculation and advise their clients that the worst pressure is X as a result of a pressure transient.

When you consider that the piping codes all say that the design pressure shall include surge pressure, this ignorance is unforgiveable. how many engineers avoid surge analysis because it is too hard?

See my papers on Risks & Surgee at

http://www.pipingdesign.com

or

http://www.olaer.com.au/products/water-surge-vessels.htm.

Geoffrey D Stone FIMechE C.Eng;FIEAust CP Eng

http://www.waterhammer.bigblog.com.au

 

[BigInch]

Vent-O-Mat looks like a pretty nice source for air valve info, but must be the slowest site on the web. I lost patience.

Stanier, do you have a link to Trey's AFT paper? I couldn't find it.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[sailoday28]

stanier (Mechanical)I think that Joukowsky will give conservative results when used with the assumptions it was derived for. That is low velocity realtive to sound speed and highly incompressible.
The elasticity of piping should serve to reduce sound speed and with it peak pressures.
Yes column seperation is important and as you state should be considered. As I uderstand it, typical water-hammer anlysis assumes the water-void based on isothermal analysis. That too may not be conservative, since it is possible during an expansion to have flashing and pressure drop in the two phase region as the flashing temperature is reached.

Regards

 

[BigInch]

A manual analysis may assume isothermal conditions. A good transient program will not. Column separation and vaporization reduces temp in the region, whereas recombination of the columns with associated vapor condensation and pocket collapse releases heat and raises the temp.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[stanier]

BigInch

It wasnt a paper by Trey it was from the Impulse manual. I will get it to you.

Geoff

Geoffrey D Stone FIMechE C.Eng;FIEAust CP Eng

http://www.waterhammer.bigblog.com.au

 

[sailoday28]

BigInch (Petroleum)I doubt that the water hammer programs out there can handle adiabatic flow. Whether by hand or computer, isothermal solutions and those using constant wave speed only require the intersection of two characteristics (dx/dt = c and dxdt=-c) to obtain solutions at future times.
With adiabatic and or heat transfer, the method of characteristic equations also require the additonal complication of solution on a particle path ( of dxdt=u, where u is the local velcity).

I'd be interested to know the available programs out there which handle the more complicated MOC solutions.

As a correction to my last post, The elasticity of piping should serve to reduce WAVE speed and with it peak pressures.

Regards

 

[BigInch]

You can model the effects with the better programs. They allow column separation to be accurately tuned to measured field responses, if not entirely rigorously calculated. The temperature effects act to slow down vaporization on column separation, since the temperature reduces resulting in a corresponding reduction in the vapor pressure and slower vaporization, with the reverse being true upon column collapse when condensation liberates heat, thereby increasing vapor pressure and slowing the vapor condensation. Vapor pressure changes are calculated resulting in corresponding velocity changes in the segments affected. The timing can be manually adjusted to match field observation.


BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[sailoday28]

BigInch "The temperature effects act to slow down vaporization on column separation, since the temperature reduces resulting in a corresponding reduction in the vapor pressure and slower vaporization.."

As an expanding fluid drops in pressure and as you state there is a reduction in vapor pressure. However there is an increase of quality for two phase homogeneous fluids.
What is your basis for slower vaporization?

Also, sound speed will radically change at the point of zero quality and in the two phase region as quality increases.
I'd be interested to know of water hammer programs which handle the varying sound speed for the two phase region.

Regards

 

[BigInch]

Let's just say it would be stopped completely if the vapor pressure on liquid cooling dips below pipeline pressure. I believe it slows between vapor presure and bubble point, but I don't know exactly how rigorous a solution it becomes there and also regarding the quality. I can't see the calculations or access that detailed data and the manual is not explicative. The sonic velocity is adjusted. Suffice it to say that the solution is adequate for most pipeline work.

Send me an e-mail and I can show you what the manual has on it.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[sailoday28]

BigInch (Petroleum)
For sending the manual
sailoday_28@yahoo.com
Regards