Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Calculating Film Coefficients

  1. Jul 23, 2012 #1

    vhf

    User Avatar

    I'm trying to estimate the heat transfer coefficient from the surface of a hot object moving through open air. After much searching, all I can find is coefficients for sundry fluids confined within the tubes of heat exchangers.

    I'm trying to calculate for a body at between 300-460K moving at up to 35m/s through air at about 300K.
     
  2. jcsd
  3. Jul 23, 2012 #2

    Simon Bridge

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member
    2016 Award

    Welcome to PF;
    I'd want to model that as Newtonian cooling since the air is moving over the body fast enough to carry the heat away. Objects in air lose heat mostly by convection. You seem to have been looking at heat transfer by conduction.
     
  4. Jul 26, 2012 #3

    vhf

    User Avatar

    I'm well aware that the problem is one of convection not conduction as are the references on heat exchangers I've been reading.

    Out of 60 or so scenarios that are listed, the one that seems to be of closest relevance is turbulent flow of a cold gas over a "warm wall". In this case the Prandtl and Grashof numbers can be disregarded and the coefficient is a function of the Reynolds number:-

    P= 0.055k/l R^0.75 (Imperial Units) and R= vL/n where

    k= thermal conductivity
    l= length of pipe
    n= kinematic viscosity
    v= velocity
    L: "for lengths greater than 2 feet use L=2"


    This begs a whole host of questions:-

    What exactly does a "warm wall" consist of?
    Should I read "length of wall" for length of pipe?
    Why are we dividing by l when the transmitted power is already per unit of surface area anyway?
    What is the relevant dimension for irregular objects?
    Is l the same as L, and if not is "lengths greater than 2 feet" referring to l or L?
    Why is R apparently being calculated from a longitudinal rather than a lateral dimension?
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook