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!

B Rate of Heat Flow

  1. Jun 30, 2018 #1
    I recently read in a Khan Academy article that the rate of energy exchange through heat across a material of thickness ##d##, surface area ##A##, and thermal conductivity ##k## can be approximated by $$\dot{Q} = \frac{kA\Delta T}{d}$$ where ##\dot{Q}## is the heat rate and ##\Delta T## the temperature difference between the two sides of the material. ac67696eb6a0451324b5a0855ce6888d34155198.jpg

    However, I am a bit confused by why the thermal conductivity of the media on either side of the material in question does not affect that heat rate across the "middle media," ##\dot{Q}##. Referring to the included image, what if the thermal conductivity of the media composing the body of ##T_1## was close to zero? Would the heat rate across the middle media not be affected? According to the equation for ##\dot{Q}## above, it is not.

    Is ##\dot{Q}## solely a function of temperature difference, independent of the "sandwiching" media involved?

    Khan Academy article I'm talking about:
    https://www.khanacademy.org/science...-heat-transfer/a/what-is-thermal-conductivity

    UPDATE: I think I understand what's going on. I believe that the heat rate across any media is, in fact, independent of the surrounding media's thermal conductivity with temperature difference between the surrounding media being the only external parameter of any importance. Please correct me if I am wrong.
     
    Last edited: Jun 30, 2018
  2. jcsd
  3. Jun 30, 2018 #2

    jbriggs444

    User Avatar
    Science Advisor

    Thermal conductivity is a property of the body of a substance, not its surface.
     
  4. Jun 30, 2018 #3
    Thanks for that edit! I updated my post quite a bit to reflect that.
     
  5. Jun 30, 2018 #4

    jbriggs444

    User Avatar
    Science Advisor

    Yes, this is correct.

    If the surrounding media is a good insulator and the media in the middle is a good conductor, a heat flow equilibrium of sorts will eventually be obtained with a high temperature difference across the insulating material and a low temperature difference across the conducting material. The result will be (once this equilibrium is obtained) an identical heat transfer rate throughout.
     
  6. Jun 30, 2018 #5
    Thank so much! That helps a lot. I'm new to this forum... is there a way to mark a question as resolved?
     
  7. Jun 30, 2018 #6

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    Of course it is relevant. That diagram is idealised. Heat loss or gain from the sides of the material has to be reduced as much as possible or compensated for. If you Google "Searle's Bar" and "Lee's Disc" you will see two methods for finding the thermal conductivity of, respectively, good conductors and bad conductors. The sides need to be insulated as well as possible and the apparatus can be operated in an ambient temperature that's about half way between hot and cold sinks.
    Only occasionally, when the Mods consider a thread has totally run its course. As far as this thread is concerned, I should say the night is young.
     
  8. Jun 30, 2018 #7
    Okay, yes. I was thinking that something must be idealized. Why must heat exchange from the side material be minimized though? Doesn't that just lead to a change in ##\Delta T## and thus a time dependent ##\dot{Q}##? In other words, in that case we may rewrite the equation to say something like $$\dot{Q} = \frac{Ak\Delta T(t)}{d}$$
    Do you just mean that there must be some sort of adiabatic boundary around the system (sandwich and middle media)? Wouldn't removing an adiabatic boundary again just lead to a time dependent ##\dot{Q}##?
     
  9. Jun 30, 2018 #8
    If you suddenly change one of the end temperatures while holding the other end constant at the same temperature that both ends started with, the temperature within the conductor will depend on time, and ##\dot{Q}## will be a function of both time and position. However, at long times, when the conduction reaches steady state, the temperature profile in the conductor will become linear and the rate of heat flow will become constant.
     
  10. Jun 30, 2018 #9

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    If you don't eliminate / minimise it then the simple formula no longer applies - that is if you want to actually measure the conductivity. There are many ruses to eliminate the effects of heat loss in thermal experimental work.
     
  11. Jun 30, 2018 #10
    Is the steady-state condition for the conductor a fixed ##\Delta T##? In other words, must the surrounding media at temperatures ##T_1## and ##T_2## be approximate heat sinks?
     
  12. Jun 30, 2018 #11
    Yes.
    In the scenario depicted in your figure, yes. But, more generally, whatever the temperatures are at the two very ends of a conductor, at steady state, the rate of heat flow is given by your equation and the temperature varies linearly through the conductor from T1 at one end to T2 at the other end.
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted