Minimum Pipeline Flow 4

[magusinp]

I am designing a device that uses the energy from the flow velocity in a pipeline to function. So I need the forum members to tell me what you use as the minimum practical pipeline flow velocity. This device may be used in mainline and plant environments so I would like the full range. So far my device works down to 10 fps but it is more efficient as the velocity increases.

 

[DrillerNic]

Your question is a bit like how long is a piece of string, but most long distance pipelines operate in turbulent flow, so the minimum flowrate to attain turbulent flow would be a reasonable minimum. This rate can be looked up on the Moody Diagram for different pipe sizes and fluid properties.

 

[JStephen]

Seems like I recall that water piping is usually sized for 5 ft/sec or less. If the pipe is intentionally oversized for any reason, velocity will be less. And at times of less-than-peak usage, velocity will be less.

I would think that in most cases, the minimum flow velocity would be zero. The minimum velocity is usually not the design criteria, and any given pipeline will be shut down at some time or another.

It might help to indicate your application more clearly. Liquid? Gas? Water? Petroleum? Steam? 1/2" pipe? 120" pipe?

 

[zdas04]

For gas lines with liquid, GasUnie recently published some data that showed liquid drop-out occured at around 3 m/s (11 ft/s). Above that the gas tended to drag the liquid along with it. Below 3 m/s the liquid collected in the line.

For gas lines, I'm now using 3 m/s as the "minimum" design conditions in gas lines.

David

 

[magusinp]

Thank you for your replies. JStephen, I'm developing a chemical injector so it might be used in any application we can show that it works in. DrillerNIc, there is no problem with turbulent flow, that has plenty of excess kinetic energy. It's at the lowest end of the flow regime that we reach our limit. Zdas, Another thread in this forum mentioned that a Dutch laboratory study found 13 f/s is the limit for liquid drop out. So your 11 f/s is good real data.

I am still welcoming any more replies!

 

[vanstoja]

For the design of a pipeline chemical injector device that depends on energy of the flow velocity, I think you should be concerned about determing the minimum flowrate to completely fill a pipe of any diameter and figuring out what the fluid velocities are in partially filled pipes at lower than filled-flowrates. Seems like you should also be concerned about piperun orientation (horizontal, vertical, canted) to ascertain the correct radial positioning of the device for flowrates less than filled flowrate. The only input I can provide to these problems is to pass on an unproven (to me) equation for minimum filled flowrate found in Costa,,D.P., 1963, Design News Reprint RS 528, Vol. 18, No. 10, 5/15/63. The equation is attributed to Greve,F.W., 1929, Bulletin of Purdue University, Vol.12, No.35.
The equation is Q=7.3*D^2.56*K^1.84 where Q is in GPM, D is inside diameter of standard iron pipe of 2 to 6 inches diameter and K is fraction of ID under fluid from 0.2 to 0.6. I ran some numbers for several pipe diameters in and out of the ranges specified and got results for a near 12 inch ID pipe that seemed plausible from my experience with reactor coolant pumps and piping. For 2, 6 and 12 inch IDs the equation results for K=1.0 are Q=43.05, 716.79 and 4226.95 GPM with respective full-flow velocities of 4.40, 8.13 and 11.99 fps. I can't offer anything reliable at this time on how to get in-fluid (not area averaged) velocities for partial fill fractions but it looks like a possibly simple circular segment geometry problem. Gas-water flow technology may provide an answer to the part-filled pipe orientation problem.

 

[BenThayer]

the rules of thumb i use are:

liquid flow: 10 ft/sec max. unless specific short uses and 7 ft/sec for reasonable economic sizing (pipe costs vs. pumping costs).

vapor flow: 100 ft/sec max.

 

[magusinp]

vanstoja: Thank you for bringing up an issue I had not thought of - partially filled liquid lines. I will take your information and digest it and get back with you.

BenThayer: Thank you for the maximum flow Rules of Thumb but I am actually concerned with minimum velocities. If you can forward me that information it would be most helpful.

 

[greg87]

I interpret your question as: you would like to know what is the lowest flow rate that would be typically found in typical industrial pipes, for insight into what flows your “device” may encounter. If my interpretation is correct, then:

A "usual" range would be 5 to 8 fps, with velocities significantly higher not uncommon. I would say that at the low end it would not be uncommon to see some lines operating at 3 fps. Obviously one could encounter velocities less than that, but not likely.