Myths and truths about boiler return temperature protection.

As the number of radiant floor heating installations continues to increase, so does awareness of what can happen when a conventional boiler is operated at low water temperatures. The problems fall into two categories:

    1.Rapid fireside corrosion of the boiler's heat exchanger due to sustained flue gas condensation.

    2.Thermal shock resulting from relatively cool water entering a boiler that's already at a high temperature.

Surefire methods for preventing these conditions are not yet well understood nor applied by some practitioners new to radiant system design. This month we'll look at both myths and truths regarding boiler return temperature protection.

The Way It Used to Be

Prior to the rapid growth of radiant floor heating installations, most hydronic systems had relatively low thermal mass and operated at fairly high water temperatures. Many were equipped with tankless coil water heaters that required minimum water temperatures in the range of 150 degrees F or higher at all times. Significant boiler oversizing was also common. Such operating conditions made flue gas condensation a short-lived problem at the beginning of a cold-start firing cycle. Any condensation that formed inside the boiler quickly dried up before doing any harm.

Occasionally a bypass pump was installed between the supply and return connections of the boiler. Its function was to create a mixing point upstream of the boiler's inlet to temper modest quantities of cool water before it could enter the boiler's heat exchanger and set up thermal stresses that could crack cast-iron like a sledgehammer.

In essence, a bypass pump borrowed heat from the hot side of the boiler's heat exchanger to boost temperatures on the cool side. The net effect was a reduction in the temperature gradient across the heat exchanger, which in turn reduced thermal stresses. The bypass pump's ability to provide this buffering relied on the thermal mass of the boiler being somewhat comparable to that of the distribution system it was coupled to.

So what's changed? Well, for one thing the number of high mass/low temperature floor heating systems paired up with conventional boilers. Most people who install these new systems now realize they require some means of protecting the boiler against sustained flue gas condensation. Unfortunately, too many of them still think a boiler bypass pump (alone) is the solution. This is a myth - one that has undoubtedly led to serious corrosion in many boilers paired with low temperature distribution systems.

Aquastat vs. Nature

Let's look at boiler temperature protection from the standpoint of thermodynamics. The famous first law states that energy (in this case heat) can be moved from one place to another, but in the process can't be "misplaced." Think of it as nature's way of accounting for Btus the way a bank auditor keeps track of dollars. A good example of this principle is how nature always seeks a balance between the rate heat is being added to the water in a hydronic system and the rate heat is released from that water. Nature always "knows" the temperature where these energy flow rates will balance and steers the water toward that temperature.

Say we have a boiler rated at 100,000 Btu/hr., and it's connected to a radiant floor slab as shown in Figure 1. The floor slab is large enough that it can dissipate 100,000 Btu/hr. when the average water temperature in it is 100 degrees F. In other words, the floor can release all the heat the boiler can throw at it without need of its average water temperature going above 100 degrees F. The boiler of course has to run nonstop to keep delivering 100,000 Btu/hr. to the floor.

Its aquastat setting is essentially irrelevant. It could be set at 140, 180 or even 240 degrees F. As long as it's a few degrees above the water temperature the floor operates at, all the aquastat can do is keep the boiler firing nonstop. The floor slab, not the aquastat, is clearly controlling the water temperature in this situation. And by the way, water entering a gas- or oil-fired boiler at 100 degrees F will cause copious and sustained flue gas condensation. That's right - even with the boiler running continuously.

Hungry For Heat

he scenario above assumes the floor slab was operating at a steady temperature and removed heat from the water at the same rate the boiler added it. Although we often design hydronic systems around such presumed steady-state conditions, they seldom exist for any length of time. Instead our systems are often trying to raise the temperature of a cool (or even cold) floor slab. It might be during system start-up in fall, or while bringing a weekend house up to normal temperature following a deep setback. Perhaps it involves the warm-up of a snow-melting slab. In all these situations, the chilled slab doesn't hesitate to show its voracious appetite for Btus.

When a floor slab is cold it can "strip" heat out of a passing water stream much faster than when it's up to normal operating temperatures. It's entirely possible for the cold slab to remove heat from the water three or four times faster than the boiler can replace it.

To understand what happens, think of a bucket half full of water. At the top of the bucket is a hose adding water to the bucket at, say, 5 gallons per minute. At the bottom of the bucket is a 1/4-inch diameter hole through which water leaks out. What happens to the water level in the bucket under these conditions?

It's pretty likely your intuition tells you a 1/4-inch diameter hole can't drain water out of the bucket as fast as the hose can add it, and hence the water level goes up. Now think of the water level in the bucket as representing the water temperature in a hydronic system. If the heat emitters (be they baseboards, panel radiators, floor slabs or the like) can't release heat from the water as fast as the boiler adds it, the water temperature goes up, just like the water level in the bucket.

To see the flip side of this reasoning, imagine you could instantly enlarge that hole in the bucket to 2 inches in diameter. If the flow coming in from the hose stays the same, what happens to the water level in the bucket? Again, intuition strongly suggests the water level drops. Why? Because the outlet flow rate is now much greater than the inlet flow rate. The same thing happens to water temperature in a hydronic system when the boiler can't put heat into the water as fast as the heat emitters can remove it.

The presence of a mixing device or bypass pump between the boiler and heat emitters is irrelevant to the accounting of energy required by thermodynamics. You could put a 5-horsepower pump between the supply and return of the boiler under these conditions, and all it would do is vigorously stir the water as its temperature drops.

Some of you have probably watched the boiler temperature gauge plunge with a combined sense of amazement and helplessness during the start-up of a cold slab heating system. Maybe you reacted by cranking up the boiler aquastat with the attitude that you're in control. Perhaps you didn't notice the boiler was already running nonstop, putting out every last Btu/hr. it's capable of. You glanced over at that humming bypass pump and realized it's about as effective as an airplane engine without a propeller. Mother Nature is taking your system down.

So What Does Work?

A bypass pump is no match for a hydronic system with lots of cold thermal mass. The only sure-fire way of raising a cold thermal mass up to normal operating temperatures without creating sustained flue gas condensation is to install a device that both monitors return water temperature and responds to low temperature conditions by restricting the flow rate of hot water entering the distribution system.

That control could be a modulating two-, three- or four-way valve, or an injection pump operated by a variable speed control. Typical piping configurations for these options are shown in Figure 2.

Contrary to what many first think, this method of boiler protection does not slow down the process of warming the slab. Why? Because the boiler is running full tilt while it's being protected. Every available Btu is being sent out the slab. What boiler protection does do is establish the maximum rate of heat output and prevent the slab from gulping down those Btus any faster. This allows the boiler return temperature to remain stable at, or just above, the set minimum temperature as the slab warms up.

Are Bypass Pumps Worthless?

On the contrary, a boiler bypass pump helps buffer a boiler against temporary influxes of cold water. The mixing point it creates takes the edge off that cold water, helping prevent thermal shock.

A common configuration is to install the pump as part of a boiler loop that supplies hot water to a pair of closely-spaced tees, which in turn connect to the mixing device as shown in Figure 2. This allows bypass of hot boiler water to boost return temperature when the load is less than the boiler output.

Just remember that a bypass pump alone can't see or react to what's happening to boiler return temperature. Therefore it can't limit the rate heat is removed from the boiler by the distribution system. It's not a substitute for a temperature-reactive control when connecting a high mass/low temperature load to a conventional boiler. Pumps are obviously indispensable in modern hydronic systems - just don't expect them to rewrite the laws of thermodynamics.