Thinking about convection currents.

 

If you had a closed-loop of pipe full of liquid, in the shape of a square, stood it uupright, and then heated the left vertical side whilst cooling the right vertical side, would the liquid flow in a clockwise direction? Rising in the left side and sinking in the right etc?

 

If not, could it be made to do so by maaking some kind of adjustments to the set up?

The radiating heat would cause the liquid to radiate out radially, and not simply up, although there will be a pressure difference that will cause this as well. I think you will see the fluid that is rising streaming along the walls, and the same with the radiated fluid, it will also stream along the wall but on the opposite side. I believe you will see a dispersion that coheres into a stream that will subsequently engulf a cooler liquid 'sheeted bubble' which will quench the hot spot and recycle the process. There will probably be two cycles one from above and one from below. The one from below will be smaller. Overall the cycle I would imagine would be quite turbulent.

 

I have no solution for your problem short of using a u tube that allows for the hot water to stream up and flow back around. I am obviously just making unsupported generalizations, the outcome is going to have a lot of dependencies that no one can easily account for in a brief topic of conversation. The proper answer to your question is probably no one really knows from the description given. I'll stop talking now, and just post my reply regardless!

Yes, as long as the liquid expanded when heated and contracted when cooled. Early cars used this system where the water around the engine rose and the water fell as it was cooled by the radiator.

 

 

"Engine cooling

1937 diagram of engine cooling entirely by thermosyphon circulation

Early cars and motor vehicles used thermosyphon circulation to move cooling water between their cylinder block and radiator."

 

 

http://en.wikipedia.org/wiki/Thermosiphon

 

 

Yes, as long as the liquid expanded when heated and contracted when cooled. Early cars used this system where the water around the engine rose and the water fell as it was cooled by the radiator.

 

Yes, a U tube! :/

Yes, a U tube! :/

The OP said a closed loop? By this I assume a complete circuit of fluid, rather like the car cooling example.

 

 

 

ahhh I thought he had a square tube and wanted a closed loop of current. :/

 

**Like not a loop but just a single chamber in the shape of a square!

 

** somebody neg rep me, make me feel better . ..

..........................** somebody neg rep me, make me feel better . ..

No! (Reminds me of the old joke about a masochist who met a sadist in a pub)

Edited April 10, 201213 yr by Joatmon

A simple tube will have the same effect, with heat being absorbed on the bottom, vaporising the liquid which travels up the tube. It is then cooled at the top by fins or fans and the liquid cascades back down to the bottom.

In a distillation column 'reflux' is estabilished when equilibrium is reached between the liquid being vapourised and the vapour being liquified.

 

 

ah yes sorry, it is a closed loop of pipes, so basically just a tube, in the shape of a square.

 

sorry for the confusion

 

i simply said square because it was easier to describe the heating of one side and the cooling of the other.

 

yes sorry.

 

and MigL, i was thinking along the lines of it all staying in liquid form, not vaporising, but judging from what Joatmon said i'm guessing it would still work?

 

thank you all for your answers, very helpful!