Joukowsky Head - Water hammer advise - formula use and surge mittigati

[jamesbanda]

Dear all,

I've Joukowsky Head method to estimate surge flow for a liquid line. This calculates 130% overpressure. Our internal standard only allows for 100% for lined and plastic pipes. We also plan to increase line capacity
1. Is this normal pratice.
2. What the experience with surge protections (on corrisive service ?) do NRV's work ? if so how effective etc.
3. Ive assumed 1500 m/s for the propagration speed eg water. but it is hydrochloric acid - do you know how to calculate this value ?

 

[BigInch]

If you're building to code, you won't be allowed 130% for transient pressures. The maximum allowed values are in the range of 10% transient over MAOP.

Increasing capacity will increase the velocity and worsen the transient pressures.

Usually speeds of sound are measured and bulk moduii calculated from that value, but it can be done backwards if you know the bulk modulus of HCL.

Some pressure balanced NVR's will work (especially for pump start and stops in small systems), but their helpfulness may be limited, if transient pressures high and sustained. In those cases a relief valve, surge relief valve, or surge tank, may be your best alternatives.

"I am sure it can be done. I've seen it on the internet." BigInch's favorite client.

"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.live.com

 

[stanier]

Suggest you get hold of

Fluid Transients in Pipeline Systems by ARD Thorley

Pressure tarsnients in Water Engineering Ellis.

In answer to you questions the Joukowsky equation does not provide the worst case in transient events if column separation occurs. Pressures have been experienced at 6-7 times the operating pressure.

NRVs in themselves do not solve waterhammer events however they need to be considered to avoid the problems. Refer

http://www.noreva.de

Thereferences above will help you derive the celerity with HCL.

 

[MortenA]

I could also suggest that you consider a transient flow study (computer simulation) to clarify if you have a problem. Joukowsky is the theoretic max. You make have some margins. And also you could then use the correct pipe and fluid parameters, and test mitigating measures.

Best regards

Morten

 

[BigInch]

Morton, I of course agree that it's a good idea to do the transient study for a number of reasons, but especially because the theoretical maximum for pressures from the Joukowsky equation, as Stanier mentioned, are those for velocity head only and do not consider pressures generated from recombination of separated fluid columns.

"I am sure it can be done. I've seen it on the internet." BigInch's favorite client.

"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.live.com

 

[PaoloPemi]

I agree with BigInch and MoprtenA, Joukowsky (or Allievi or Frizell or...) method gives a realistic solution for a limited number of cases, see some comments here

http://www.kypipe.com/new_stuff/JAW200701_AWWAaward.pdf

(I have found the paper with Google), you should correctly interpret the results or apply a more accurate methodology (keeping in account the complexity of your system)

 

[MortenA]

There will forever be one thing that will puzzle me - why is it that people of so many different nationalities think my name is Morton (when its Morten)...

Best regards

Morten

 

[BigInch]

Why not just consider it as the result of different accents.

My apologies Morten.



"I am sure it can be done. I've seen it on the internet." BigInch's favorite client.

"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.live.com