Thread: convection & Inverted siphons.............

I'm currently making a convection water cooled thing.............the problem is i need to make an air trap so the air and water in the system separate.

I have come up with something that looks like an inverted siphon - this is good for water but I am worried about the convection side of things.

My question:

Would the inverted siphon be able to work as well with convection as it would with normal hydraulics?


I've come up with a few models but the whole thing seems rather paradoxical!

I can't picture it. What is an inverted siphon?

Originally Posted by Harold14370

I can't picture it. What is an inverted siphon?

an upside down hose ?

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

'Inverted siphon.
An inverted siphon is not a siphon but a term applied to pipes that must dip below an obstruction to form a "U" shaped flow path. Inverted siphons are commonly called traps for their function in making expensive articles like rings and electronic components easily retrievable.[citation needed] Liquid flowing in one end simply forces liquid up and out the other end, but solids like sand will accumulate. This is especially important in sewage systems or culverts which must be routed under rivers or other deep obstructions where the better term is "depressed sewer". Large inverted siphons are used to convey water being carried in canals or flumes across valleys, for irrigation or gold mining.'

The romans used them to cross valleys. Simply put - the exit is lower then the entrance - so over goes the water.

I'm using it as an air trap - the entrance (being higher) should trap the air (which I have an outlet for). Even though buoyancy plays a large role in convection currents, i'm thinking in terms of advection it should work - even if the siphon in question might have to get a little hotter to match the flow earlier on in the system.

I am also guessing that currents where the hot part is at the top and cooler part at the bottom are not as fast or turbulent as the other way round - has anyone any further insight on this?

Much obliged.

If I understand you correctly, you have some kind of heat source at the bottom of the system and a cooler or radiator of some kind at the top. (It won't work with the hot part at the top.) In between, in the hot leg or the cold leg, you have a trap, which is partly full of air, partly with fluid.

That could work, but any kind of loops in the pipe like that would make it harder to get the siphon started. Also, you would have to make sure that when you vent the trap, you are not letting air in. That would break the siphon.

Originally Posted by Harold14370

If I understand you correctly, you have some kind of heat source at the bottom of the system and a cooler or radiator of some kind at the top. (It won't work with the hot part at the top.) In between, in the hot leg or the cold leg, you have a trap, which is partly full of air, partly with fluid.

That could work, but any kind of loops in the pipe like that would make it harder to get the siphon started. Also, you would have to make sure that when you vent the trap, you are not letting air in. That would break the siphon.

correct but a little different.

it is a loop that is filled with water. The air trap is to bleed the system of air that diffuses out of the water when it is heated. The point is so no air gets in the radiator.

Its not like a siphon where i am worried about 'breaking' the siphon as when the water at the bottom of the system heats convection currents will take it to the top rather than it working like a siphon. The 'inverted siphon' part is the air trap i have made to bleed the air out and still allow the same rate of flow and it is horizontal. Since it is horizontal (as inverted siphons generally are) I am worried about what this will do to the convection currents...............

Originally Posted by fatman57

Its not like a siphon where i am worried about 'breaking' the siphon as when the water at the bottom of the system heats convection currents will take it to the top rather than it working like a siphon. The 'inverted siphon' part is the air trap i have made to bleed the air out and still allow the same rate of flow and it is horizontal. Since it is horizontal (as inverted siphons generally are) I am worried about what this will do to the convection currents...............

The thing is, you have created a loop in the piping where there is a section that goes up from the heat source to the trap, then back down, then up again. So when you first start it up, you have to heat up the hot leg to get the natural circulation going. If there is a trap in the hot leg, the hot water has to get past that by going down, which will require some suction on the downstream side of the trap. This is provided by the cooler water descending the cold leg piping which draws a vacuum on the hot leg. So, you will have to be concerned about breaking the vacuum.

Originally Posted by Harold14370

The thing is, you have created a loop in the piping where there is a section that goes up from the heat source to the trap, then back down, then up again. So when you first start it up, you have to heat up the hot leg to get the natural circulation going. If there is a trap in the hot leg, the hot water has to get past that by going down, which will require some suction on the downstream side of the trap. This is provided by the cooler water descending the cold leg piping which draws a vacuum on the hot leg. So, you will have to be concerned about breaking the vacuum.

thanks - i do see your point! much appreciated.

the air trap alone should help this by removing air so hopefully the tensile strength of the water will hold! My guess is it will as it also has a very short space to travel. Also the geometry is such that i don't think air will want to go sideways.

One thing i am worried about is that the water will expand rather than draw vacuum - this will stop/slow the flow. As long as the radiator which is after the trap (plumbed parallel rather than serial) can remove heat quick enough this could create enough of a vacuum, at least i hope!

What's the "tensile " strength of a fluid/
Or do you mean vsicosity?

Originally Posted by fizzlooney

What's the "tensile " strength of a fluid/
Or do you mean vsicosity?

you mean Viscosity?

lol....................Yes........................ ....... :-D

Couldn't you just have an open vent at the top of the radiator? Any air would leave naturally, and this would also allow for thermal expansion and contraction as the system heats up and cools down. Air drawn in during cooling periods would be forced out again when it warms up. I'm assuming your system operates at atmospheric pressure.

Originally Posted by Bunbury

Couldn't you just have an open vent at the top of the radiator? Any air would leave naturally, and this would also allow for thermal expansion and contraction as the system heats up and cools down. Air drawn in during cooling periods would be forced out again when it warms up. I'm assuming your system operates at atmospheric pressure.

i could but i don't want the water to evaporate.

It will work like a car cooling system where the heat makes it pressurised. This will raise the boiling point if slightly and the slightly raised pressure will make it a slightly more effective cooling system then atm would be able to offer if not more thermally efficient.

The big problem i had was when i had to change everything to bleed the air out when its working - major headache! Any air in the radiator will make it less effective.