Thermosiphoning?

Eric Johnson

I think IFC stands for "Integral Flow Check." Taco and Grundfos circulators with IFC have those letters in the model number.

A gravity flo-check is one of those red cast iron fittings you'll find on boiler piping. They have an arrow cast into the body indicating the direction of flow. I think they're considered superior to the swinging-flap style. There's probably a cheaper way around your problem than buying a new, bronze, IFC circulator, although a cast-iron flow-check is probably not going to do it in fresh water.

Charles_8

Valve to prevent thermosiphoning?

I think I've found the next problem with my homebrewed external-storage hot water system (original tankless coil in steam boiler, circulator pump, 40 gal. water heater tank). There may be some thermosiphoning going on, since although the thermostats and pump are properly set, the first gallon of water feels 20 or 30 degrees hotter than that.

What I think is happening, and please tell me if this makes sense, is: During a heating cycle while the pump is running, water is pumped from the hot water outlet downwards through the coil and back into the bottom of the tank. However, the boiler is at 160-180F after the thermostats turn off the pump when the tank reaches 120F. Now the coil is hotter than the tank, so hot water rises from the coil (reverse flow, backwards through the pump body) and back to the top of the tank. The cooler water at the bottom of the tank sinks, being more dense, and pushes water backwards through the coil, completing the thermosiphon loop.

I could put a swinging-flap check valve in the vertical line from the pump to the coil. That would prevent any reverse flow of hot water but could silt up or corrode over time.

Would a vertical loop somewhere in the line work to prevent thermosiphoning instead of a valve with a moving part? I don't see how this would stop the siphoning, though, since the flow is driven by the heat differential.

thanks for any help

-Charles

radioconnection_2

unwanted gravity circulation

Unless I am misreading your problem, this is usually controlled by using a circulator (or flange) with an IFC and a gravity flowcheck on the other side.

Charles_8

What are those?

I have a Taco 006B circulator, which appears to have no such reverse flow prevention devices.

I'm not familiar with the terminology - what does IFC stand for, and is a "gravity flowcheck" the same as a swinging-flap check valve like this:


http://www.pexsupply.com/Categories.asp?cID=3&brandid=#

?
thanks

Charles

hr

a circulator with integral check

would handle any potential thermosiphoning. They have about a .5 psi "pop" So flow in either direction should be handled.

You may still get some "stacking" in the tank DHW side. As the warmer water rises. A good, listed, thermostatic mix valve may be a good addition to prevent scalding on the first draw. It would also allow you to run warmer temperatures for extended drawdown, and help prevent bacteria growth potential if the tank is kept to low in temperature.

I like the Honeywell brand mixes. Watts, Conbraco and others have listed valves.

hot rod

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

Charles_8

Wish I'd thought of that...

... before buying my circulator

There shouldn't be much problem, if any, with "forward" thermosiphoning (hot water leaving the tank), since the heat is on for eight or nine months a year up here, and the coil should always be hotter than the tank. Even in summer, firing the boiler only for hot water, I doubt it'll cool far enough below 120 (in between water-heat cycles) for much thermosiphoning to occur. So a single check valve should be ok, unless there's a similar one that works in both directions?

I already purchased a Watts thermostatic mixing valve, which I was going to use to replace the nonfunctioning one attached directly to the coil output, before I decided to install the storage tank setup. Guess I can hook that up too... more pipes for the basement "nest of snakes"...

-Charles

Unknown

Can you tell us

exactly how the piping is connected from the coil to the tank ? Or better yet , can you post a picture of the system ? We've installed many a setup like you have ( we call them aquaboosters ) , and the only time there was an issue of too hot water right at the start was a broken dip tube on the cold inlet . We never use IFCs or checks with our aquaboosters . I'll see if I can find a pic of a steamer and aquabooster .

After that initial blast of hot water , does it come out at the set temp of the aquastat ? Are you using a surface mount aquastat ? Does your tank have 2 of them , and did you use the lower one to control the bronze circ ?

Charles_8

Picture

OK, great, you've done this before Here is the best picture I could do (the basement isn't that big)

http://i12.tinypic.com/4tgu3pl.jpg

The copper lines have been insulated since I took this picture. Water comes out of the tank, runs along the ceiling, to a tee (run feeds the house, sidearm down to the circulator input). Circulator outlet descends to the coil. The coil output further drops to floor level and goes into the drain port of the heater (there is a tee and valve on this line for draining).

Hot water appears to come out at the proper temp after one or two gallons are drawn, although I have not done actual temp measurements yet.

The boiler aquastat is set to 180 with 10 diff, and it is wired in series with the T-T contacts on the SR501 relay (tucked up on the joists with the expansion tank), so that the boiler only fires if the tank is calling for heat AND the boiler is below 170. (Or if the wall stat is calling for heat, of course).

The water heater (tank) has the standard two-thermostat setup for lower/upper heating elements, both currently set to 120. In order to avoid modifying the heater's wiring (should I want someday to hook it up to 240v) I simply connected its line leads (red/black) to the single-zone input terminals on the SR501. (Although it may look strange, the voltage drop from the series resistance of a 4500 watt element is negligible in an 0.3 amp thermostat-type circuit and I have verified that the circulator does turn on and off properly).

The manual for the heater says that the lower thermostat satisfies first, then switches to the upper one until both are satisfied (at which point the circuit is completely open and the circulator turns off). Should I turn the upper one all the way low, so that the bottom stat setting dominates?

thanks

Charles

Tilmon

Just a few thoughts: If you are using this type of setup you should raise the differential on the boiler (25 F is good) to raise the boilers efficiency.
If you can get an integral flow check for your circulator it would probably work and cost about 5$, they are very easy to install, just cut the water and remove the pump. Push the valve in the end and re-install. These valves are usually made of plastic and stainless so they should last (I use them a lot and work very well(grundfos))