Flow Turbulence - Pipe Fittings

[Galuck]

I am looking for information on how to calculate the amount of turbulence created by the use of 45 and 90 Deg Elbow. Also interested in the same for straight pipe vs corrugated metal hose.

Thanks,

 

[BigInch]

Turbulence isn't measured directly in pipe flow, never mind calculated, but it is related to Reynolds number, as that is used for the calculation of head lost by flow through the fitting.

"Make everything as simple as possible, but not simpler." - Albert Einstein (1879-1955)
***************

http://virtualpipeline.spaces.live.com/

 

[Prestz]

Hello. You can get information in White Frank, or Hansen Fluids Mechanical. There is good information and methods for energy loss calculations.
good luck

 

[zdas04]

Like BigInch says, the only measure of turbulence that I know of is Reynolds Number, but that tends to be a go-no-go (i.e., if RN is greater than 7,000 the flow is turbulent, at 1,000,000 is is still turbulent, but you really can't say it is "more" turbulent). There are some flow-visulazations floating around that show zero increase in vorticity or in random flow motion when going through a LR 90 or 45 fitting.

I don't think that what you're looking for exists.

David

 

[btrueblood]

Galuck,

What specifically are you trying to do?

Both prior posters are hampered somewhat by your use of the language, it's like you asked "how to measure blue?". You can measure Vrms, the rms value of flow velocity, you can measure swirl, you can measure mixing vs. length...all of which depend to some degree on how turbulent the flow is. You can also measure the pressure or head loss due to bends, pipe walls, etc.

 

[khardy]

what is the fluid?

 

[Galuck]

sorry for the language issue. I am not only trying to calculate the pressure drop in 45 and 90 Deg elbows but also the turbulence created by the change of direction when those elbows are used. The media is water.

 

[BigInch]

Why calculate turbulence? Isn't it the lost head that you're interested in? Besides I don't think turbulence is a scaler quantity, so its easier to see it than it is to calculate it. The best method (I think) is to calculate Reynolds number. If its below 2000, its laminar; no turbulence. If its between 2000, beginning turbulence to fully turbulent at appx 8000. Anything more than 8000 is just "more turbulent". You could draw streamlines around the Ells and calculate the velocities along each streamline then calculate the Re along each streamline and you might gather that turbulent flow would be slightly more promenant in the outer streamlines, those streamlines with the greater radaii. Shears created by different velocities between streamlines as the fluid moved through the ell, would have to be disipated by turbulence by the time the fluid reaches the end of the ell, and all fluid particles once again return to the original velocity profile, symetrical blunt-nose bullet pattern, immediately downstream.




"Make everything as simple as possible, but not simpler." - Albert Einstein (1879-1955)
***************

http://virtualpipeline.spaces.live.com/

 

[rcooper]

I think this is a language issue. Flow is turbulent or laminar, or somewhere between the two. Flow stops being laminar when the kinematic forces are sufficient to make the laminar flow unstable.

Flow can be uniform or non-uniform. With uniform flow, the velocity (speed and direction) is the same throughout the fluid. This is a perfect case, and rarely achieved. The flow can be axially uniform, but radially non-uniform. In turbulent flow, the flow is non-uniform on a microscopic level, but can be uniform on a macroscopic level.

Flow can be steady or unsteady. Its velocity varies with time. This can arise due to changes in flow rates, vorticy shedding, changes in fluid properties.

With water, you almost certainly have turbulent flow going in and out of your bend.

The bend will affect how uniform and how steady the flow is. I think this is what you are asking, how to work out how steady and uniform the flow is after a bend.

 

[btrueblood]

Pressure drop across fittings or lengths of pipe is fairly well understood; any good fluid mechanics text will have the information you need, or get ahold of a copy of Crane Co.'s Technical Report #410.

There is gross eddy motion caused by flow turning through pipe bends, this motion can be visualized as two independent axially rotating eddies downstream of the bend (the fluid at the inside of the bend wants to drop thru the pipe centerline towards the outer wall, the fluid at the outer wall has to move to get out of the way, and so flows along the outer wall towards the inside of the bend...). There is not a lot of good information on how these eddies combine with the standard wall friction to induce additional head loss in piping systems with multiple bends and/or straight lengths, nor is there a lot of information on the persistance of the eddies. For the most part, the eddies are ignored, and people analyzing the systems add a bit of fudge factor into the known friction factor values for bends to compensate. There is some data available on how well the eddies induce mixing, if you look hard enough and deep enough for it. It takes persistance, and access to a good engineering library. Start with ASME technical papers and reports, and move outwards from there.

 

[Zapster]

Galuck,

You state that you are trying to calculate turbulence. Just what units do you propose to use to characterize chaotic property changes? To my knowledge, I don’t believe there is an IP or SI unit for turbulence. If there is a language barrier problem here, you could state what base units you would like to express turbulence in. This would certainly help folks understand what you are tying to accomplish.