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

Problem at calculation of convection

  1. Dec 6, 2009 #1

    I am working on a case in which i'm trying to calculate the heat load of a house (actually mine).
    I want to be sure that the calculations were done correctly during the design phase.
    Disregarding the methods at various web sites (most of which are commercial), i'm
    doing it in a scientific way.

    I'm kind of stuck at calculating "h", the coefficient of thermal convection. (Here i'm
    calculating h1 and h2. h1 is for the convection between interior of the house and inner surface
    i.e. the wall; h2 is for the convection between outer surface i.e. the wall and the exterior
    environment. I can't measure the temperature values at the inner and the outer surface which
    is the main problem.
    I can not use the correlations that uses Rayleigh and Grashoff numbers for the reason that
    i do not have the inner and the outer surface temperatures.
    I could not find any case studies. Am i following the wrong way here?

    Any help is greatly appreciated.
  2. jcsd
  3. Dec 6, 2009 #2
    Yeah. Your not going to be able to accurately calculate the heat transfer coefficient of your house using only theoretical models. The heat flux through the walls (and windows) not only depends on the conditions outside (solar load, wind speed, etc) but also on the amount of heat transport due to infiltration and exfiltration. In other words, your system is to complex to just simply apply a correlation.

    Houses with automated thermal systems and home energy auditors will have thermocouples in every room of the house along with the outside. By heating the house to a set temperature and measuring the rate at which it cools off you can estimate a bulk heat transfer coefficient and equivalent area for your house. These values can then be used to determine heat flux out of your home under most conditions.
  4. Dec 6, 2009 #3
    Thank you Topher925,

    I am O.K. with your comment about the complexity of the situation. Calculating the load is far away from a simple correlation. Simplification is just going to affect the result in a misleading way.

    But what are your comments on calculation of "h" when the surface temperatures are not
    available? Is it possible to do so? I searched some books and academic papers but i didn't find anything related.
  5. Dec 6, 2009 #4
    I don't think you will be able to find h without the surface temperatures. If you have no information about temperatures, flow velocities, or thermal properties about the wall, then I don't believe you will be able to find h.
  6. Dec 7, 2009 #5


    User Avatar
    Science Advisor
    Gold Member

    You will need the teperature of the surface and the bulk teperature of the air to calculate the free convection coefficient. Think about it this way- a wall at 100 degrees will have a higher convective coeffcient than a wall at 50 degrees.

    If you were calculating a forced convection coeffcient, you might be able to make some educated guesses about the fluid properties of the air and get to a reasonable solution. For free (natural) convection, temperature is the driver that causes convection (and therefore you need to know it to know how much air is flowing).
  7. Dec 8, 2009 #6
    Thank you. As long as my case is free convection, i will at least try to assign some approximate values for the temperatures mentioned, to get a reasonable result.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook