Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Estimate Friction Factor Accuracy

Download as pdf or txt
Download as pdf or txt
You are on page 1of 2

BIBLIOTECA

* Estímate
; fiictíon factor
- accurately
^Here is a very accurate equation
^for calculating the friction factor
^directly, without trial-and-error.
A

^ T. K. Serghides, Kerr-McGee Chmical Corp.

The Colebrook equation is a widely used method for


^redicting/, the friction factor for fluid flow:* C= -2.0log(^ e/Z) ^2.51^^
3.7 Re

I ^=-2.0 1og(ff +1^) (.) Eq. 2 is derived by applying Sieffenson's accelerated-


convergence technique to an iterative, numerical solu
^here € is the absolute pipe roughness (ft), D is the tion of Eq. 1. The constants A, B and C are approxima
nternal diameler of the pipe (ft), and Re is the Reynolds tions of Eq. 1 obtained by three iterations of the
«•«umber for fiow. direct-substitution method. Eq. 2 is lhe approximation
You cannoi solve the Coiebrook equation directly, as obtained by combining those constants according to
le friction factor appears on both sides. To solve it Steffenson's formula.
-^equires trial-and-error or a graph—which are difficult to It happens that the numerical solution of Eq. I by this
f^rogram. Or you can use one of severa! approximate technique converges very fast, so Eq. 2 is very accurate—
^ílutions ihat are explicil in /, and lherefore can be i.e., it approximates Colebrook's equation wiihin a small
^olved directly. fraction of a percent.
This article presents two new approximations, based Henrici [Z] explains this behavior in discussing the
numerical solution of the Colebrook equation. Both bounds of error on a numerical solution after a finiie
^pear to be more accurate than any of the other pub- number of iterations. A previous article [2] discusses the
íhed approximations. conditions for convergence in solving the Colebrook
equation by direct substitution, and explains Steflfen-
. he friction-factor equations son's method.
This explicit friction-factor equation is valid for iransi- A simpler version of Eq. 2 is nearly as accurate, and
-^^nal and turbulent fiow {Re > 2,100) ai any relative perhaps easier to use for hand calculaiions. Like Eq. 2,
loughness (e/Z)): this equation is obtained by applying Sieffenson's
method to a numerical solution of Eq. 1. It is valid for
Re > 2,100 and any value of e/Z):
{A - 4.781)^ \-
Vre: ^ = _2.0 log M) / 781 -
B -2A + 4.781
(3)

where: A= -2.0 log 4-^ )


Noif ihiii ihc/usfd hcii- is ihc Darcv íactor. »vlth whkh Irictional head loss (A.
•^lav bc calciilaicdas A= l{IJl)){i''/-K)- Hci e. l. is k-iiKth (li), i' is vclociiv(fi/s).
g is the gravitaiional constam (fi/s*).

CHEMICAL ENÜINEERINC MARCH 5. 1984 63

You might also like