Corner Radii Square Pipe Fluid Flow Heat Transfer

[Mandrill22]

I'm trying to prove that increasing the corner radii of a square tank will improve heat transfer because of improved convection, based on ease of flow from hot to cold. Does anyone know what equations I will need to prove this?

 

[hydtools]

Maybe this may help. As I recall, we kicked this question to the curb in another thread.

Ted

 

[racookpe1978]

Is your "square tank" axis "sideways" or "vertical"? (What are the dimensions of the tank?)
What is the temperature difference between internal (a fluid?) and the external (air ?)
Is the tank insulated?

What difference does it make? (Literally!) How important is the loss of heat transfer capability compared to the (increased or decreased!) cost of fabrication?

Example: Bending the two intersecting sidewalls might be much cheaper than welding two flat pieces of steel plate. Or maybe not - It depends on what tooling is available to bend the walls, and how big the plates are, and how "square" the corners might become if bend and if welded with an internal fillet and external fillet. "Rolling" a completely smooth corner might be cheaper than either - or maybe not, if you have to buy a 15 foot long roller, and only have a 3 foot roller now.

 

[SandCounter]

Consider that heat transfer is heavily influenced by surface area. By going from square to round you lose 21.5% of the surface area at the corners. You will need to determine if the increased heat transfer from convection makes up for this loss in conduction.

I used to count sand. Now I don't count at all.

 

[IRstuff]

Please refrain from duplicate posts; you asked essentially the same question a month ago.

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

 

[Mandrill22]

racookpe1978, it is vertical. It is not insulated. It is about 100 degrees C hotter in the tank than ambient.

 

[btrueblood]

Strange that you want to increase heat transfer from a hot tank, but whatever. Tanks store fluids. To change the properties (temperature, pressure, etc.) other devices should be used. Changing temperature is best done with a heat exchanger, not a tank. You can modify the tank to make it more like a heat exchanger (add penetrations and cross- or counter-flow, add fins and other means of increasing surface area, etc.), or pipe a closed recycle loop through a heat exchanger, or ...

But rounding the corners would seem to lower, not increase, the heat-exchanger-iness of the tank, like SandCounter said.

 

[racookpe1978]

If it is a vertical tank, then you can either round the four intersections between the (assumed flat!) bottom plate and the 4x vertical walls, or round the circular intersection between a flat bottom and the single round wall.

Heat transfer is by convection from the fluid to the inner surface of the wall, conduction through the walls and bottom plate, then by convection from the walls and bottom plate to the(assumed still!) air in the room. (Radiation losses assumed near-negligible from a tank containing 100 C fluid = boiling water at atmospheric pressure?)
If outdoors, wall convection will be usually turbulent in wind. If indoors, you can make it turbulent by putting a fan in the room .

Either way, there will be very, very little difference between the smooth transition of bottom to sidewalls compared to a square transition (fillet welded). IR photographs of the fluid film surface on vertical surfaces show mixed flow begins quickly, even at low differential temperature.

(For a horizontal tank, the rounded edges becomes either a horizontal pipe (maximum curve) or near-square box configuration.)