I consider myself lucky to live in a rural dairy-farming valley in upstate New York. About 30 feet from my office is a barbed wire fence that usually keeps the cows in the adjacent pasture out of our lawn and garden.
In October of any given year, those cows can decide the grass really is greener on the other side of the fence. On a crisp fall morning, we may wake to the sound of a dozen or so Holsteins grazing on our lawn and feasting on the garden. Those girls simply found a hole in the fence and took advantage of it.
It's interesting how lots of things in nature have that inherent ability to find a hole in the proverbial fence. One example is how hot water in a boiler can spot a cool piping loop above it. Nature quickly establishes a gentle but persistent "gravity flow" that allows some of those boiler Btus to sneak away from the system in a way the designer never intended.
It's simply nature's way of trying to reduce any imbalance in the amount of heat contained by water in different parts of the system. We might not even realize such pathways exist in the system, but rest assured nature will find them and exploit them to the best of her ability.
The development of circulators during the early 1940s was a turning point for the hydronics industry. No longer were designers bound by the constraints of early gravity-flow systems. Hot water now could be sent in any direction with relative ease.
Unfortunately, it's easy for those of us who missed out on these early hydronic systems to forget that the use of circulators doesn't nullifu the law of gravity.
Mending The Holes
Here's a list of several details you can use to minimize or eliminate gravity flow. The ways are illustrated in Figure 1.
Never substitute a swing check valve for a flow check valve in any situation where gravity flow would move in the forward direction of the valve. The flapper disc in a swing check is not heavy enough to prevent gravity flow. Been there, tried it - it doesn't work.
Never use a thermal trap as a substitute for either a flow check or spring-loaded check valve on the supply riser of secondary circuit. The trapping effect cannot stop gravity flow up the supply riser after the secondary circuit shuts off. The lighter (hot) water in the supply riser, in combination with the heavier (cooler) water in the return riser, produces a slow but persistent gravity flow that only stops when the primary circuit turns off and all the piping cools down.
Why should both sides be protected? Because nature has a way of setting up a two-directional flow in a single pipe. Hot water tends to slowly flow up the center of a vertical pipe while cooler water slides down along the pipe wall.
Don't believe it? Go feel the return pipes where they rejoin the return header on a multizone system without gravity flow protection. You'll find the return piping on inactive zones will be warm, even hot, several feet back from the return header. A swing check at the return end of each circuit puts a stop to this.
Call me a control freak, but when I see unnecessary heat migration in a system, I know that system isn't performing as well as it could be. Why leave any "holes in the fence" through which some of those Btus can sneak away to unexpected or undesirable locations? Paying attention to situations that allow heat migration makes good design sense and negates a source of potential callbacks. And who doesn't want that?