Temperature and friction factors

[TeeAr]

Hi there,
Today I am posting a double.
I red this interesting post, thread378-227715, where I could find an exhaustive information concerning possible formulas for friction factors.
I'd have following questions: each of these formulas are affected indirectly by temperature.
Temperature influences density and viscosity, that are included in Reynolds number.In theory this would account for temperature change.
My question is: does any of this correlation loose precision when we are dealing in temperatures of water around 180F instead than when we are working with temperatures of 50F or vice versa?

Many thanks

 

[zdas04]

All of the "shortcut" equations were developed for slide rules. They are all trying to make the "transmission factor" (1/f[sub]f[/sub])^0.5 easier to solve (with the goal being to put it into a closed form equation). In MathCad, it is trivial to solve the Colebrook equation with a "Given/Find" structure. I guess a Reynolds Number, get friction factor from Colebrook, run the isothermal gas equation, calculate a new Reynolds Number, get a new friction factor, etc. I do it in a MathCad loop and the loop counter is never more than 10 for a 0.01% tolerance.

The real trick looked like it was going to be solving for a temperature-dependent viscosity. In the slide-rule days we always skipped that step and just used the same viscosity everywhere. I recently did a comparison of a dozen different empirical equations for gas viscosity as a function of density. I compared them to the values from NIST RefProp.exe and found that allowing dynamic viscosity to vary from 25% of NIST to 200% of NIST gave me a variance in flow rate of less than 5% for the new plastic pipes and 15% for commercial steel pipes. After a week of messing with this stuff I determined that we were closer to "right" with holding viscosity constant than the new guys are by working so hard to quantify changes.

David

 

[BigInch]

You can find tables for the viscosity of water at 0-100 C temperatures. Then you could run the numbers and answer that question for yourself. Use the Churchill equation to avoid iterations.

We will design everything from now on using only S.I. units ... except for the pipe diameter. Unk. British engineer