1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Using the Root Sum Square method to determine the uncertainty of heat transfer coeff.

  1. May 5, 2012 #1
    1. The problem statement, all variables and given/known data

    This is a lab report that I'm working on. I'm almost done, except for the last part which consists of finding the error in the result for our heat transfer coefficient h using Newton's law of cooling. I've tried this in different ways but I can't seem to get it right.

    We have 51 data points for temperature and 51 data points for their respective times.

    2. Relevant equations

    ln( (T-T(room))/(T(initial) - T(room)) = -hAt/(C*p*V)

    Where T is the variable of temperature of the metal sphere whose temperature is being measured, T(room) is room temperature, T(initial) is initial temperature of sphere, h is the heat transfer coefficient, A is the surface area, t is the time, C is the specific heat, p is the density and V is the volume.

    The root sum square method says, that if we have h(T,t), then the uncertainty of h:

    Uh = +/- [ (T/h * ∂h/∂T * U_T)^2 + (t/h * ∂h/∂t * U_t)^2 ] ^(1/2)

    U_T is the uncertainty of the measurement device, which is 0.5 degrees C. U_t I assume will be 1 second, since the timer was read to the nearest second.

    3. The attempt at a solution

    First, solving for h(T,t):

    h = - [CpVln((T-T(room))/(T(initial)- T(room)) ]/(At)

    Partial derivatives:
    ∂h/∂T = -CpV/(A*t*(T-T(room)))

    ∂h/∂t = [CpVln((T-T(room))/(T(initial)- T(room)) ]/(At^2)

    Multiplying by T/h and t/h for the respective components above:

    ∂h/∂T *T/h = 1/(T-T(room)) * (1/ln((T-T(room))/(T(initial)- T(room))

    ∂h/∂t * t/h = -1

    But then when placing these two in equation for U_h, I'm not sure which T to use for the portion of ∂h/∂T *T/h*U_t.





    Another way that I tried solving this was by letting f(T) = ln(T-T(room)/(Tin - T(room)).

    But then the function will be come h= h(f(T), t), so the uncertainty would be:

    U_h = +-[ (∂h/∂f(T) *f(T)/h*U_f(T))^2 + (1)^2] ^(1/2).

    ∂h/∂f(T) * f(T)/h would reduce down to -1, so I would get:

    U_h = +-[ (-U_f(T))^2 +1 ]^(1/2).

    I wasn't sure if this second method was right, and if it is, what would I use for the uncertainty of the f(T) function U_f(T)?

    I hope this post makes sense. I've been working on this particular part of the lab for so long. Any help would be greatly appreciated.
     
  2. jcsd
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Can you offer guidance or do you also need help?
Draft saved Draft deleted



Similar Discussions: Using the Root Sum Square method to determine the uncertainty of heat transfer coeff.
Loading...