I was calculating conduction heat transfer through a glass window and I got confused conceptually. Lets assume that we have a glass window of area 1m2. Thermal conductivity (k) of glass is around 1 W/(m.k). I probed into ASHRAE's handbook to find the overall heat transfer coefficient for a 3.2mm single glazed glass window and found the U value to be around 7W/(m2.k).

We know that U = k/L where L is thickness of the body.

Here is the confusing part, if we assume the window to be 100% glass, we can calculate its thermal conductivity by multiplying the U value with thickness of the glass. By multiplying 7W/(m2.k) with 3.2mm i.e. 0.0032m we get a value of 0.0224W/(m.k).

Where am i mistaking? One one hand we have a thermal conductivity og 1 W/(m.k) and on the other hand 0.0024W/(m.k). Why such a huge difference?


The overall heat transfer coefficient U includes the air boundary layer resistances on the inside and outside of the glass, which are in series with the window glass and dominate the resistance.

| cite | improve this answer | |
  • $\begingroup$ Even if we do the maths and include those values of air film, i am sure that it cannot compensate for that HUGE difference between 1 W/(m.k) and 0.0024 W/(m.K). $\endgroup$ – Engineer SAM Oct 12 '17 at 17:07
  • $\begingroup$ Oh yeah? According to the OP, the value of 0.0024 W/(m.K) is equivalent to an overall heat transfer coefficient of 7 W/m^2K. In Transport Phenomena by Bird, Stewart, and Lightfoot, they cite typical values of 3-20 W/m^2.K for one-sided natural convection heat transfer with air. Two-sided overall heat transfer coefficient would be half of this, neglecting the glass resistance. It all seems totally consistent. $\endgroup$ – Chet Miller Oct 12 '17 at 18:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.