Air Sweeping Velocities in Pipeline Design

[kakarot98]

Hello All,

I am trying to find a good reference on required air sweeping velocities in pipelines. I work in the mining industry where air valves are often considered a nuisance and commonly removed after the first sign of trouble. That being said, I am trying design the pipeline to ensure that any trapped air can be forced out of the line via the upstream movement of water. I cannot seem to find a good reference with a fairly simple set of equations to deal with this. Separately, but perhaps also slightly related, there seem to be multiple variations of equations out there that project the maximum possible surge pressures in a given line and they result in slightly different answers. I have attached the two methods I am currently using and as a note, I believe the Square Root symbol in the first method is meant to encompass the entire bottom half of the equation, otherwise it results in an inordinately low surge pressure. Does anyone know a definitive equation to calculate maximum potential surge pressure in a line, given diameter, flow, etc.? Or perhaps can verify the veracity of one of the two attached equations over the other?

Any help would be appreciated.

Thank you.

 

[LittleInch]

I've always gone for 1 m/sec myself.

But see

https://www.eng-tips.com/viewthread.cfm?qid=470064

and scroll down about 2/3 of the way and I attached a paper which might be useful.

Surge is a transient event and a simple equation is usually very conservative.

The kowalsky? equation is the most commonly used to get a worst case.

But surge is a strange thing and can be affected by many parameters, especially valve closure speed, flow velocity, type of pipe, length of pipe etc.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[kakarot98]

Hey LittleInch... I was actually hoping you would reply; I saw one of your responses elsewhere, but didn't know how to message you directly (if thats even possible). I will take a look at the paper, thank you. That being said I am aware of many of the factors that surge is dependent on. We have VFDs on all the pumps, and there will be explicit instructions on valve closure times or they will be spec'd to utilize some sort of slow close mechanism. The one scenario I am worried about is a power failure as the mines are typically very remote and outages are not uncommon; in this case the VFD woudn't do anything. I didn't really want to spec a surge anticipation valve, although that is exactly what I would do on some sort of municipal supply system. Again, thank you for your help, and I look forward to future discussions with you. Regards.

 

[1503-44]

Jakowsky.. yes it is the sqrt of all the terms denominator

Surge pressures, valve closing times,

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

Reality used to affect the way we thought. Now we somehow believe that what we think affects reality.

 

[LittleInch]

No problem - There's no DM on this site so it's all public for view, which is normally a good thing.

You don't give much (i.e. none) detail so it's difficult to say much more, but one thing to note is that if you're doing down mine shafts, that's a different issue to running pipework or pipelines inside a tunnel or something similar.

If the system is pumped then it's generally OK - it's the gravity fed systems which can suffer from air locks and similar and tend to have lower velocities.

On surge the equations should give you a worst case for instantaneous closure. Anything else you really need a transient network analysis. I've normally found anything less than 3m/sec is usually OK, once you get 5+ then you've got some serious issues.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[1503-44]

You can estimate surge vs valve close times. Its in the link I posted.

Reality used to affect the way we thought. Now we somehow believe that what we think affects reality.

 

[kakarot98]

Super. Thanks guys. I appreciate your help. Till next time!

 

[gerhardl]

Hi all,
I have with interest read the postings and references, and agree with the main conclusions. I am however a bit 'hanging in the air ' about your real practical problems. (Taken down by first problems?)

1. Yes. Normal practical water filling of pipelines is (in my experience) 0,5 m/s up to 1 m/s. Might be for higher under filling, but reduced down to 0,5-1 when nearing full.

2. Yes, water and /or air pockets can remain and cause trouble or water hammer depending on layout of pipeline.

3. Yes, proper drainage of rest air or water pockets are essential, and will require air outlet valves or steamtraps for water outlet.

4. Suitable valves or regulators must be used for the slow water inlet, fear of caitation if standard BF valves are used for regulation if only crack opened.

5. For air outlet dimensioning of airvalves you can roughly calculate air escape at 40-60 m/s ar smallest orifice (suggested value).

Good luck!