Those new to designing or troubleshooting hydronic systems often misunderstand how the speed of water flowing through a heat emitter affects heat output. More than once I’ve heard someone reason that if water flows through baseboard too fast, there isn’t enough time for the BTUs to “jump off.” Apparently BTUs are smarter than Tom Cruise in Mission Impossible. They know when the “water train” they’re riding is moving too fast to safely jump off, and therefore stay put and get swept out the other end of the baseboard. This reasoning implies that slowing down the flow rate through the baseboard would increase heat output. This, however, is not the case.

More Speed = More Heat: Heat moves from a fluid to a solid (such as baseboard) by convection. The rate of heat transfer depends strongly on the speed of the fluid along the surface. The faster the fluid moves, the faster heat transfers.

For example, think about a small hydronic fan-coil unit in comparison to a fin-tube baseboard of equivalent output rating. The fan forces air across the hot surfaces of the coil much faster than the air naturally flows across the baseboard’s element. The resulting higher rate of heat transfer per square inch of surface area allows a relatively small coil to release as much heat as several feet of baseboard.

Humans experience this effect as well. The concept of “wind chill” is familiar to almost everyone. Simply put, the faster cold air passes over our skin and clothing, the faster heat is “scrubbed off” by convection. The moving air isn’t any colder than the thermometer reading, nor can it cool an object below this temperature regardless of wind speed. It just feels colder because of the higher rate of heat transfer from our exposed surfaces.

This principle applies to both gases and liquids. In a hydronic system it’s at work whenever hot water flows through the inside of tubes, heat emitters, boilers ... whatever.

A basic precept of hydronic heating is that increasing the water temperature inside a heat emitter increases heat output. Suppose we pass 180 degree F water through a given length of baseboard and it exits at 160 degrees F. The temperature drop is 20 degrees F and average water temperature in the element is 170 degrees F.

Now let’s increase the flow rate so the water exits at 170 degrees F. This time the temperature drop is 10 degrees F. Some would argue that because the temperature drop is only half its previous value there will only be half as much heat output. Unfortunately they’re ignoring the fact that heat transfer depends on both the temperature drop of the fluid, and its flow rate, (see Hydronics Workshop, March 1997). What’s actually happened is the average water temperature in the baseboard has increased to 175 degrees F yielding greater heat output.

Tradeoffs: I’ve used my Hydronic Design Toolkit software to size the baseboards in the system shown in Figure 1 assuming several different flow rates. Notice how the required length of the baseboard decreases as the system flow rate increases. Not surprisingly, the head loss of the piping system also increases.

Nature is full of tradeoffs, and hydronic systems certainly have their share. For example, increasing the flow rate of water through the piping system in Figure 1 reduces the length of baseboard required. But at some point we’ll have to use a more powerful circulator to supply the extra head needed to maintain this flow rate. Not only will a larger circulator cost more to install, it will require more electrical power, and thus add to the operating cost of the system, especially when considered over the life of the system.

The objective in such a situation should be to use all the head a given circulator can produce to maintain flow as high as possible, while still avoiding flow noise or erosion of the piping components. The latter is generally accomplished by sizing piping so its flow velocity doesn’t exceed 4 feet/second. (Figure 2 lists the flow rate corresponding to a flow velocity of 4 feet/second for several sizes of copper tubing.) Also check manufacturers’ specs for any limits on flow rate through a specific device. This being accomplished, it makes no sense to throttle away any “extra” flow a circulator can produce just to match the flow rate the baseboard is rated at.

Diminishing Returns: The principle of increased heat output from increased flow rate also applies to floor heating circuits. Figure 3 plots water temperature at various distances along a 300-foot floor heating circuit operated at one of three flow rates. In all cases the circuit is supplied with water at 110 degrees F. (See Figure 3)

Notice the 20 percent increase in circuit heat output when the flow rate is increased from 0.5 to 1.0 gpm. However, bumping the flow up another 0.5 gpm produces a much smaller gain, only about 7 percent. It’s the law of diminishing returns. If the flow rate is increased farther, the rate of heat output continues to go up, but by a smaller amount each time. The situation quickly becomes limited by the rapidly increasing head loss and potential for flow noise taking precedence over the small gains in heat output.

The trends shown on this graph have two very practical implications. First, trying to regulate the heat output of a floor circuit — or any heat emitter for that matter — by adjusting its flow rate is a very “non-linear” process. Small changes in flow rate in the lower flow range will produce big changes in heat output. The same change in flow rate in the upper flow range will have minimal effect on heat output. Think about this the next time you’re adjusting balancing valves on a manifold.

Secondly, operating a floor heating circuit at a very low flow rate, (and thus high temperature drops) significantly reduces the heat output of the tail end of the circuit. This results from the “droop” in the curves of Figure 3 at low flow rates. A further disadvantage of this situation is a very noticeable change in floor surface temperature across a room. Although the total heat output may still be satisfactory, customers will question why the floor feels cooler in some areas. From a heat output standpoint it makes more sense to install two shorter circuits rather than one long circuit. The total heat output per foot of tubing will be higher. This of course has to be weighed against the extra cost of additional manifold connections, etc.

See For Yourself: Gather up technical specs for several types of hydronic equipment such as fan-coils, panel radiators, flat-plate heat exchangers, boilers, and even hydronic heat pumps. Look at how the heat output of these devices change when operated at different flow rates. In all cases you’ll see increased heat output as the flow through the device increases. If enough data is provided you’ll also see the law of diminishing returns in effect. In almost all cases the issues of head loss, flow noise, and even the potential for erosion of metal surfaces quickly becomes the limiting factor.

So in the end those BTUs have no fear of leaping from a fast-moving “water train.” In fact the faster the train careens through the system, the more BTUs line up to make the jump. The allure of a cooler metal surface is just too tempting to resist.