thermosiphon question

[yazguli]

1 -Is there something like "pressure gradient" or "driving force gradient" in a thermosiphon system?If so,how is it drawn?
2 - Suppose a simple thermosiphon system of a heat source at bottom ,a sink at top and two risers.If there is another sink in the middle of source and sink how the system will work? Which sink will receive more heat?
The question arises from this practical question ;in a multifloor building when the pump is off,which floor will get more heat?

 

[btrueblood]

Thermosiphons work due to a difference in density, i.e. hot water is lighter than cold.

No easy answer to #2, it would depend on the heat transfer occurring at each floor, and the piping layout. One can easily envision a scenario where the nearer floor sucks all the available heat, starving the flow to the upper floors, or the opposite, where the density gradient is high enough to cause the hot fluid to zip right past the first floor tee.

 

[racookpe1978]

Or an intermediate or startup/shutdown condition where flow may actually be trying to go the wrong way.

 

[chicopee]

Assuming parallel risers, the riser with less heat loss and less frictional flow should provide more heat.

 

[btrueblood]

But, with parallel risers, there is at least the ability to control the flow to each floor via suitable control valves. The OP described "two risers" which I ASSumed was one feed and one return line. Thus, I was envisioning a system with two floors fed by a single riser, which I think is inadvisable.

 

[chicopee]

In boiler parlance, with water tube boilers, one pipe would be a riser and the return pipe would be called a downcomer. So again in many of these posts, terminology is everything.

 

[yazguli]

Thank you every body
By "riser" I mean both supply and return pipes.A single supply pipe and a single return one with branches at each floor.
As racookpe1978 has mentioned I have experienced cases where flow tends to stop or tries to go the wrong way. Also as btrueblood has mentioned there were cases where the first floor has starved.
And what about gradient? Can we consider the heat source like a pump and draw the gradient graph with the least head just before the source and the most head just after it?

 

[LittleInch]

You can't really think of thermo hydraulic effects in the same way as a pump or else all the flow would go through the first floor as the least resistance. As it is a density driven system the driver is gravity or desire of the lighter product to get to the highest point. To make a middle floor work you really need a vertical fork or Y and then go horizontal from one branch after a short vertical leg and vertical up to the top floor from the other

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[btrueblood]

I like Chicopee's terminology, let's try using it.

An efficient, self-starting thermosiphon has to have a large vertical region that can be made to be the hottest place in the system. If the expansion of the water can cause some of the hot fluid to "back up" into the downcomer pipe, then the siphon won't start.

Once started, it becomes a balancing act to try and get the flow equalized - back to my first reply. Thermostatic valves on the radiators (heat sinks) might help (but these can be slow to react). Even better might be some active control systems that sense the temperature of the fluid along the length of the riser, and at the inlet and outlet of the radiators, and open/close motorized valves to vary the flow in/out of each radiator. Check valves in the feed line to each radiator might also help...but the biggest problem happens when a slug of hot fluid gets forced into the downcomer, something only a finely tuned controller can hope to prevent, I'd think.

 

[racookpe1978]

BTrueBlood:

Your summary was what I was trying to get at in my earlier reply: Routine operation may work, but only after all of the "hot parts" are hot and the "cold parts" are cold and the fluid (the working chemical) are moving in the right direction.

 

[btrueblood]

Right, not to take anything away from your reply racooke, just trying to restate it and give (some of?) the parameters that cause upset to the desired flow. A lot of northern European hotels use what appear to be un-pumped, thermosiphon type systems, and control the radiators with little wax-actuated thermal control valves. You end up either shivering or sweating a lot of the time, because the valves are so slow to respond (not that cheap US hotels don't have crappy thermostats either). My point being that the slow response of those types of cheap, simple, unpowered valves might tend to cause lack of control of the types of problems we are discussing (reversed flow, poor heat distribution, poor startup response).

 

[yazguli]

Thanks .Good lessons for me.