15 ft/s speed limitation in liquids 4

[zarrus]

Hi all,

I have been trying to find some explanation for the commonly used figure of 15 ft/s speed for liquid streams but I can not find it. In my experience, I have found a lot of cases when we use a figure that anybody knows from where is coming and for me this looks like one of these.

Could somebody shed some light over this?

Thanks,

 

[bimr]

Fluid velocity is determined by the economics of pumping, not some arbitrary factor. A fluid velocity higher than 15 ft/s is possible. However, you can be sure that high fluid velocities and long pipe runs will not be economical.

Reasonable pipe velocities depend on the application. There is no correct velocity for all applications. Here is a general guideline.

Reasonable Velocities for the Flow of Water through Pipe (from Cranes Technical Paper 410):

Boiler Feed.............8 to 15 ft/sec
Pump Suction ............4 to 7 ft/sec
General Service.........4 to 10 ft/sec
City.......................to 7 ft/sec
Transmission Pipelines...3 to 5 ft/sec

Go to a basic hydraulics book. Try Cranes Technical Paper 410 as a reference for the above velocities.

http://www.eng-tips.com/viewthread.cfm?qid=111206

http://www.eng-tips.com/viewthread.cfm?qid=348061

http://www.amazon.com/Fluids-Through-Valves-Fittin...

 

[zarrus]

Thank you for your quick response Bimr,

I am happy to hear that answer. According to it, money is the key factor but there are not technical issues with higher velocities.

Let's talk about the case I am studying. I have an industrial fluid in the top of a tower 200 meters height. Therefore, the fluid has a huge static pressure head that I want to use to move this water to a certain point. After performing some calculations I have more energy that I need so I would like to reduce the pipe size as much as possible in order to save money.

According to your statement, I could increase the speed above 15 ft/s without any technical risk.

Am I right?

 

[moltenmetal]

For water, free of solids and not cavitating, erosion isn't a huge issue.

For corrosive services where erosion-corrosion is a concern, or for services containing solids which may be abrasive, you do need to concern yourself with erosion. But your problems then will be in control valve trims etc. first and in the bends/elbows of the pipe second...

 

[zarrus]

So, Moltenmetal, if we have a clean, single phase liquid you also agree we can increase the liquid velocity, don't you?

 

[EmmanuelTop]

What is this? A quiz?

Dejan IVANOVIC
Process Engineer, MSChE

 

[rconner]

Agree there is really no one answer to this question - however, you may be interested in some of the historical etc. background explained in the FAQ "What is the maximum recommended flow velocity..." at

http://dipra.org/faqs/#.

This passage includes a determination based on "kinetic energy of a particle" that happens to come up with a number very close to your subject flow velocity.

 

[cvg]

think considering cost factors alone is an oversimplification. there are technical issues. high velocities increase the risk of cavitation, erosion, water hammer, etc. you may need to evaluate all of these if you wish to operate at higher velocities.

 

[BigInch]

It is a common waterhammer guide limit. It can start to become a serious problem with liquid systems operated above 15fps.

 

[LittleInch]

Although it is often taken completely out of context, api rp 14e, section 2.3,a gives a maximum recommended single phase liquid velocity of 15 ft per second.

The fact that this section refers specifically to lines transporting liquid from one pressure vessel to another by pressure differential and is designed to prevent vapour breakout has unfortunately become lost and in experienced engineers looking for a "code" or written limit have leapt on this figure as a do not exceed velocity. This is total rubbish and while I fully agree higher velocities above 3 to 4 m/sec present difficulties in terms of excess press drop, increased surge etc, there is no limit.

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

 

[bimr]

zarrus, with the information that you provided, the answer is yes.

As the others have noted, the piping would have to be designed for the fluid, the entrained solids, and the expected fluid velocity. It may be less expected to use a lower velocity and less rigorous piping design.

 

[DubMac]

On the low end of the spectrum, I was always taught to watch out for anything below 5 fps, fearing solids dropping out.

 

[georgeverghese]

The end of this line at grade - is it leading into a liquid filled vessel with this incoming pipe fully submerged - if there is an air break at the bottom , the riser may never fully liquid prime.

If there are high point pockets in this downcomer, these will need to be vented out.

Is there a vortex breaker in the exit pipe at the tower top ?

 

[wangduyu]

Think about valves, it will be also difficult to close valve in such high speed.

 

[LittleInch]

Other than a tower block, I can't think what sort of industrial thing is 200m high, but what you describe is like a hydro power dam. If you're flow is reasonably constant, why not generate electricity if you don't needs the pressure at the base for flow? But otherwise I believe the answer is yes. Just make sure you can control the speed of valve closure or otherwise slowly bring the flow to halt and avoid small radius elbows and tees at the high velocities.

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

 

[georgeverghese]

If air can vortex itself into this exit line at 200m elevation, then this line may be vibrating with air and water phases fighting each other as they rush down this pipe. There are limits to this kind of two phase gavity velocity - it is quantified in the Francis formula for overflow or gravity drain lines. If there is some kind of low pressure holding tank at the top, the resulting air draw throught the free vent on this tank could implode the tank if free vent sizing is not adequate. No details provided for these at either end of this line. This air draw can also happen if there is low level in this tank, even if there is a vortex breaker.

 

[EmmanuelTop]

Not to forget the electrostatic charge that may (and will build up) if dissipation gradient is lower than the charge buildup. There are no universal numbers with regards to velocity limitations, because these are dependent on fluid properties, velocity, piping material, grounding etc. API 2003 (legally binding document in the US) provides general guidance:

https://law.resource.org/pub/us/cfr/ibr/002/api.2003.1998.pdf

There are some materials published by NFPA and others:

http://www.nfpa.org/Assets/files/AboutTheCodes/77/77_STA-AAA_ROCAgenda2_10-12.pdf

http://people.clarkson.edu/~wwilcox/Design/flamliq.pdf

http://www.wolfsonelectrostatics.com/01_hazards/pdfs/guidanceforplantengineers-staticelectricity.pdf

I have rarely seen any liquid lines designed above 3-4 m/sec, for many reasons.

Dejan IVANOVIC
Process Engineer, MSChE

 

[EmmanuelTop]

I found this interesting article in my archive (see attached). It recommends the following:

a) Liquids with conductivities >100 pS/m, no flow velocity restrictions
b) Liquids with conductivities <100 pS/m and no immiscible components, flow velocity should be less than 7 meters per second
c) Liquids with conductivities <100 pS/m and containing immiscible components, flow velocity should be less than 1 meter per second

Locate filters and valves as far as possible from the entrance to the receiving vessel. The filters and valves must be placed upstream of the discharge point to provide at least 30 seconds of “relaxation time” prior to the liquid discharge. The relaxation time1 would depend on the liquid conductivity, flow velocity, and type of filter or valve; hence, additional relaxation may be warranted."

Dejan IVANOVIC
Process Engineer, MSChE

 

[davefitz]

There is the criteria of economic pipe size based on capital cost of the piping vs pumping KWE usage over the time period for which the rate of return is requested. This had historically been about 10 ft/sec, but will vary based on cost of electricity, cost of borrowed money ,and cost of piping.

There is also the criteria for erosion/corrosion of the piping. Carbon steel pipe with most water flows will erode at local velocities of 18 fps, and the average velocity of 10 fps does not adequately illustrate that the local velocity at a pipe entrance ( vena contracta) will be 40% higher than the average velocity due to turbulence, and a similar effect occurs at bends and elbows. Higher grade piping can allow higher velocities before erosion becomes an issue.

Some metals erode at very low velocities ( copper and copper alloys), while others require a high velocity to prevent MIC microbiologically induced corrosion; a copper alloy condenser may not have velocities above 7 fpps, yet a stainless tube condenser should not have velocities less than 9 fps.

"Whom the gods would destroy, they first make mad "