Heat transfer coefficient change as deltaT change?

[Jensdsds]

Hello, i'm a student in electrical engineering and i'm doing a project using COMSOL multiphysiscs tool to anticipate the temperature of a cable being subjected to a specific current.
The test is done in real life and then simulated in COMSOL, to see if its possible to anticipate.
Currently we have some problems understanding the nature of the Heat transfer coefficient, in the Heat fLux module.

My simulation is tuned in so that it fits the cable dimensions etc. and the last parameter tuned was the Heat transfer coefficent[HTC].
This was set to 12 W/m^2*k. The first test was a copper cable which reached 32degC at 50Amps.

For my second test, i tested a aluminium cable which got a lot hotter than my copper cable due the the diffrent size.
At the higher temperatures the simulation became less accurate.
At temperatures around 32degC, the simulation became more accurate.

So from my standpoint is seems that as the deltaT between cable surface and ambient air becomes larger, the Heat transfer coefficient also becomes larger.
Is this correct to assume?

In this picture, the HTC is dependent on the deltaT and the heat flux.
The heat flux is almost constant, except for higher resistance at higher temperatures.
But if the deltaT rise, the HTC should become smaller? when looking at the formula.

So to sum up my question, is the HTC a constant value, or does it go up or down with a deltaT rise?

Thank you in advance, and pardon my bad spelling errors and thermodynamic newness.

Mvh. Jens

 

[IRstuff]

You should be posting in the student forum. But your modeling shows the issue that you are using a convective heat transfer coefficient; has your instructor not covered radiative heat transfer?

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[davefitz]

h is normally assigned as the convective heat transfer coefficient, so it is associated with the flowing fluid properties. If the fluid is an ideal gas or a perfect gas then h will vary according to the shear that occurs in the boundary layer near the surface of the solid object being heated or cooled, and so most correlations will vary h as a function of flow properties such as Re , modified with the prandtl number Pr. The fluid properties used are those in the middle of the boundary layer, so as DT increases the properties in the boundary layer slightly increase as well. If the fluid is a liquid close to either the saturation temperature or if it is a supercritical fluid near the critical temperature then the fluid properties vary dramatically in the boundary layer and conventional correlations for h are not valid.

In the 1970's there was an alternate method of computing q and h proposed by E Audiatori in his textbook " the new heat transfer", but it never caught on.

"...when logic, and proportion, have fallen, sloppy dead..." Grace Slick

 

[georgeverghese]

The natural convection htc does vary with surface temp. See any college or Uni heat transfer text. The natural convection htc is a function of the Grashof no and the Prandtl no, both computed at the film temp, where Tf = (Ts + Ta) *0.5.

 

[pierreick]

Hi,
You may be interested with this ppt .
Good luck
Pierre

 

[IRstuff]

Here's what's in one of Mathcad's older handbooks

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[Jensdsds]

Hello,

Thank you all for the answers, it was very informative and helpful.

Kind regards - Jens