Surge Tanks - when to use?

[MidstreamEgr]

I'm not a pipeline person (more inside-the-fence terminal piping design), but I've noticed that some pipeline receipt points or pump stations have surge tanks, and some do not. Not familiar with the pipeline-side of the business, I was curious what drives the decision to use it some places and not others. Is it typically an end-of-the-line feature only (and not at midpoint boosters), only at the "low elevation" points along the route, or strictly dictated by transient analysis but with no real commonality on where they end up?

Thanks!

 

[stanier]

When the analyst who does the transient system design recommends that there is a need for a surge tank.

You cannot guess these things. Every system and system boundary is different. The surge mitigation device is used to absorb energy. The energy is a result of momentum being changed to kinetic energy. Hence if the pipeline velocity is higher then there is a greater need for such mitigation as the momentum is higher.

Generally a surge vessel, not tank, is used when the valve closing time is such that is causes a pressure wave greater than the design pressure of the system. The surge vessel absorbs the energy and keeps the system pressure lower than design.

If you really mean a surge tank it may be to take the flow from a surge anticipation valve that opens in advance of a high pressure epak occurring. Surge tanks are generally used on water pipelines not petrochem lines.

“The beautiful thing about learning is that no one can take it away from you.”
---B.B. King

http://waterhammer.hopout.com.au/

 

[LittleInch]

Your last point is basically the reality - you normally run a transient analysis and put surge alleviation or protection when you can't avoid it any other way. The issue at booster stations is normally one of pulling a vacuum if the pump stops suddenly, but you can get high pressure transients - as stanier says it is very system specific.

Commonly found on relatively short high velocity lines, but normally less so on longer distance pipelines.

The limiting pressure for most pipeline design is design pressure plus 10% for transient events.



My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[BigInch]

There's plenty of surge tanks on petroleum liquid pipelines.

I hate Windowz 8!!!!

 

[MJCronin]

Surge tanks are also necessary on liquid thermal heating systems ( commonly called hot oil systems)

The surge tank is typically placed on the suction side of the circulation pumps and is sized for 5-6 times the system oil expansion.

A short, vertical axis tank, with pressure indication and a fill hatch seems to work best.

More good information available from the fine and helpful people at PARATHERM:

http://www.paratherm.com/resources/articles-white-papers/

MJCronin
Sr. Process Engineer

 

[msquared48]

To my recollection, concrete lined surge tanks (50 foot diameter or so) were used on Diablo and Gorge High dams on the Skagit River in Washington, but not on Ross Dam. I recall that the penstock runs on the two former dams were longer than Ross Dam's, so the volume to stop may have been greater, although the head on Ross was one to two hundred feet higher than either of the others. Lots of factors to consider...

Mike McCann
MMC Engineering

 

[BigInch]

MJC, isn't that technically what is called an "expansion tank", because it is there to accommodate increased volume from the expansion of a heated fluid, and actually higher "surge" pressure never generated since there is free expansion into that chamber, and higher transient pressures at pump suctions normally are not a problem as 1) higher pressures there are usually good for NPSH and 2) pump suction is the usually the point where the lowest operating pressure in the whole system is found, so higher superimposed transient pressure waves won't normally overpressure the pipe there although I agree it is rather pointless to harp on such a distinction.

Penstocks surge chambers are common as the changes in flowrate often need to be made very quickly to keep generator output balanced with load. Higher available heads can generate similar power outputs with equal, or even lesser mass rates at equal or even still lower velocities, therefore with lower velocities prevalent in the system, surge tank perhaps not required on high head installations.

I hate Windowz 8!!!!