Heat transfer coefficient of water and air with tank problem

[Unibond81]

Hi all :)

I'm trying to figure out how much heat I can save from covering a tank which is open (uncovered). The tank is rectangular made of steel and is filled with water, heated to 50C, and is not insulated with any material. The outside temp and humidity is assumed to be 25C and 60% respectively with a very slight breeze.

I've managed to figure out the evaporation rate of the water using q=(25-19v)*A*(xs-x)/3600 where v=velocity, A is the surface area of the exposed water, xs =humidity ratio in saturated air at the same temperature as the water surface, x=humidity ratio in the air. and finally Q=2270*q where 2270 is the evaporation heat of water.

What I cannot figure out are the heat transfer coefficients for water and air to calculate U in the equation Q=U*A*deltaT. I know 1/U=1/h(water)+deltax/k(steel)+1/h(air).

In the class room you would usually be given these values but what does one do for real situations? Do I need to calculate h form the Nusselt number (only option I can think of)? and if so what equation of Nu do I use for the water in the tank and the air as well as the reference lengths to calculate the Grashof number?

Thanks in advance for any help

P.S. For completions sake I've calculated the radiation even though this can be neglected

 

[edgepflow]

For the outside tank walls, h(air), apply a natural circulation convection heat transfer coefficient for a vertical plate. The Churchill and Chu correlation is a good choice. Any good heat transfer textbook or handbook will have the details.

For the interior of the tank, h(water), you could apply an enclosure natural circulation heat transfer coefficient. See "Convection Heat Transfer" by Bejan or similar reference.

 

[Unibond81]

Thx a lot edgepflow, thought of using the churchill and chu correlation for air but wasn't sure, guess should be more confident in myself. As for water hadn't read of enclosure natural heat transfer so a BIG thanks for that.