Reynolds Number (Turbulent Flow) 3

[dharr25301]

Hi,

I am working out Reynolds number to get a turbulent flow for flushing out our system.
I have found various website’s which state different answers when turbulence is achieve turbulent flow (Re> 4000, 3000 and 2320).

Just wondering which is correct? I initially thought it was more than 4000.

Many Thanks,

D.

 

[Latexman]

That has been a nebulous, subjective number since Reynolds day. I think he said 2100, IIRC. I use 10,000 just to be sure in my design work.

Good luck,
Latexman

 

[Latexman]

Oops! Reynold's said 4000.

Good luck,
Latexman

 

[dharr25301]

Thanks for the advise.

 

[zdas04]

I did some actual experiments a couple of years ago on a project that HAD to have turbulent flow. For Reynolds Numbers around 4,000 the flow occasionally had "random 3 dimensional movement superimposed on the bulk flow", but not reliably and not much different than 3,000. As I increased the flow rate (and therefore the Reynolds Number), I found the amount of 3-D flow (and coincident scouring of the pipe walls which was my measure) increased to a maximum around 9,500 and then didn't seem to change much as I increased it further (my apparatus could only go to about 15,000 so I can't swear that there isn't another max somewhere else. Like Latexman I use 10,000 as my threshold for turbulence.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[ione]

There are application (i.e. microchannel or helical flow) where the flow could still be laminar up to Re = 12,000 – 15,000 and even above these values, upon certain conditions are applied (there are a bunch of studies on this topic and you can check this link

http://arxiv.org/ftp/arxiv/papers/1007/1007.0810.pdf

for your reference).
Said what above I’m with both Latexman and zdas04 in considering Re = 10,000 as a valid threshold for turbulence.

 

[BigInch]

Is your system composed of pipe?
If there are flat plates involved, turbulent Rn can be over a million.

"People will work for you with blood and sweat and tears if they work for what they believe in......" - Simon Sinek

 

[dharr25301]

This system is pipe (stainless steel), 1"od with wt of 0.065, 1/4"OD with a wt of .048.

 

[zdas04]

Flow in instrument tubing is a whole different matter. The boundary layer thickness is such a large portion of the total flow area that you get turbulence at very low Reynolds Numbers. In 0.15 inch ID pipe you rarely see Laminar flow. The 1-inch is a bit more representative, but still a problem. Once you get above 1-1/2 inch the stuff I was saying above makes more sense.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[SNORGY]

Maybe turbulent flow can be defined when:

Solve(64/Re=Solve(1/(f^0.5)=-2/ln(10)*ln(k/(3.7*d)+2.51/(Re*f^0.5)),f),Re)

 

[SNORGY]

To qualify...since that was obviously wrong...sort of...

At some point, the friction factors computed one way versus the other will diverge, with Re on the abscissa, presumably somewhere to the right of where they are closely equal.

 

[zdas04]

The problem is that all the arithmetic gets really wonky when less than 80% of the flow is outside the area adjacent to the no-flow boundary. I usually take no-flow boundary in turbulent flow in pipes to be on the order of 0.01 inches. I usually take the layer as 10 times the boundary so if you reduce the ID of 1/4 inch tubing by twice the boundary layer you get a negative number (0.15 in - 0.01*10*2 = -0.05 inches. Hard to define a flow profile that can put 80% of the flow into a non-space.

For instrument tubing you need to go to the equations for capillary flow and they are pretty messy. I did some work a few years ago (1998 actually) to try to determine how much gas was vented to atmosphere with pneumatic valves operated on natural gas. Part of that work was to try to asses flow rates in small tubing. I evaluated this by timing the shut time on fail-closed pneumatic devices. With no exhaust tubing I was able to calculate close time to within miliseconds. When we added 3 ft of exhaust tubing I was able to develop an empirical relationship that matched closing time pretty well--it was nowhere close at 3.5 ft of tubing or 2.5 ft of tubing. I finally gave up and wrote my paper assuming local exhaust, a little bit sleazy but the paper was accepted by SPE and no one ever asked "what happens to these numbers if there is exhaust tubing" so I never had to admit I didn't have a clue.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.