The U.S. Environmental Protection Agency (EPA) encourages use of technologies that drive the highest efficiency and the lowest emissions. Additionally, owners and builders must sustain and document performance over the lifetime of the application.

In that context, optimal HVAC performance is critically important and it is possible while also reducing operational cost, raising efficiency, and minimizing environmental impact.

One method to accomplish this is by using what are called low-flow, low-temperature, high-efficiency systems (LLH). These systems drive supply temperatures down and temperature differentials up. In turn, this drives flow rates down and reduces the size of the fans, ductwork, pumps, and piping. LLH can be used in many applications, but it lends itself particularly well to variable air volume (VAV) systems.


Rockwell Collins, a manufacturer of aviation electronics, had a 410-ton chiller installed in 1979 that was starting to require maintenance. The base load of the building is 600 tons and the owner wanted to find a way to add one 600-ton chiller to handle the load and do it efficiently without adding a lot of infrastructure modifications.

The original chiller was designed for 58-45°F on the evaporator with 3-gpm/ton of 85° condenser water. The new 600-ton chiller was built to do 60-40° on the evaporator with 2-gpm/ton or 85-100° condenser water. The lower water temperature enabled the existing air units to handle the added load with colder air and the added capacity came without requiring any changes to air units, fans, ductwork, chilled water piping or condenser water piping.

The new chiller design added capacity and efficiency to this existing building by changing the chiller and tower, reducing the flows, and lowering the chilled water temperature without touching the piping or airside. The new chiller is more efficient at 40° leaving water temperature (lwt) at 2 gpm/ton (0.63 kW/ton actual or ARI-adjusted kW/ton, which would translate to better than 0.50 kW/ton on an ARI-adjusted basis) than the old chiller was at 45° lwt at 3 gpm/ton (0.66 kW/ton).

While this is a specific example, the question of whether this saves energy and is the optimum design needs to be addressed using hour-by-hour lifecycle cost analysis tools. These tools are now available to help determine operating cost savings potential by calculating the optimum supply air temperature and the right leaving water temperature.

And while the Rockwell Collins site was but one part of the system, efficiencies can originate in every part of the HVAC system.

Here we will look at the various locations including the cooling towers, cooling coils, pumps, and chillers.


Any tonnage range is applicable from 50 tons to 50,000 tons plus, but for this discussion an 800-ton, two-chiller application was selected. For a conventional system, this 800-ton load would require 2,400 gpm at a 10° delta tee or 3 gpm/ton. A typical cooling tower for this application would require about 40 brake horsepower (BHP).

An LLH system would require 1,600 gpm at a 15° delta tee or 2 gpm/ton. A typical tower for this application can be selected that uses a 30 BHP motor. This tower is smaller, has a smaller footprint, and weighs less. The higher delta tee causes the chiller to work at a higher head and hence consume more energy. Running a lifecycle cost program can help strike the optimum balance between the tower and the chiller.


Consider a conventional system in which the typical design delta Tee is 10°. That would mean if the temperature entered at 44° it would leave at 54°. To produce 504 MBtu (41 tons) it would require 101 gpm of water. (MBtu is equivalent to 1,000 Btus per hour.) For a system using a 16° design, the same coil, now with 41° entering water temperature, could produce the same 504 MBH with 63 gpm or 37.5 percent less water.

This would be true if the system were designed for new construction. However, it is also true if an existing coil has a two-way valve, the chiller water pump has a VFD, and the chiller water temperature is turned down from 44 to 41°.

With a two-way valve in place, the leaving water temperature does not go down, it goes up. Typically it goes up 0.5 to 1° for every 1° reduction in entering water temperature depending on the type and circuitry of the coil. The MBtu delivered is the same, so the leaving air temperature did not change.

This means that the same MBtu can be delivered at substantially reduced water flows or if the gpm were held the same and the entering water temperature reduced, additional capacity could be obtained. For the right applications, these alternatives can provide opportunities to reduce pumping BHP or handle cooling capacity problems.


Typically, the reason to reduce the water temperature is not the coil, it is producing the same MBtu; the reason is to reduce chilled-water pumping energy consumption. In our example, a conventional 800-ton design would require 1,920 gpm (2.4 gpm/ton). If one assumes a head requirement of 110 feet with pump and motor efficiencies of 80-95 percent respectively, the kW pulled by the motor would be 52 kW.

On this same system, if a VFD were applied to the pump and the chiller could produce the low temperature, when the supply water temperature setpoint was reduced from 44 to 41°, the required gpm would have been reduced from 1,920 to 1,200 gpm (1.5 gpm/ton). The total feet of head in this example would have been reduced to 49 feet, by essentially following the pump law of head which states that friction loss is reduced by the square of the gpm reduction.

In these applications the pump and motors may need to be resized to get back to the original motor and pump efficiencies. If they were, the kW requirement would have been reduced to 16 kW; or with a 36 kW savings it represents nearly a 70 percent reduction in pump energy consumption.

The reduction in pump energy illustrates the pump law, which states that power varies by the cube of flow reduction. This is especially true if a critical zone reset strategy is used that allows the pressure in the piping to be driven only by the most pressure-demanding valve.

If the pipe sizes remain the same, the pump energy can be cut by more than 66 percent. On a new construction job, the pipe size may be reduced with the goal of optimizing either first cost or operating cost or both.


When you turn down the leaving water temperature, the kW consumption on the chiller(s) will go up. In our example, to deliver the 800 tons we used two 400-ton screws. Those two screw chillers, producing 44° water, would have consumed 464 kW. When the leaving water temperature was reduced to 41° the consumption jumped to 490 kW.

Energy consumption indeed increased by 26 kW; but the meter is not hooked to just the chiller, it is hooked to the entire building. And while the chiller consumed 26 more kW the chilled water pumps alone saved, on the design day, more than 36 kW for a net 10 kW savings.

From a Coefficient of Performance perspective, chiller efficiency has improved more than 70 percent in the last 30 years; while pump and tower energy efficiency has not improved significantly. Building managers would do well to take full advantage of this evolution by using highly-efficient chillers.

Most jobs have at least two chillers for redundancy; the vast majority of these chillers are piped in parallel. Instead of piping in parallel, the full 26 kW or more can be recovered if the chillers are piped in series. This takes advantage of cascade cooling and costs less. This means the first chiller produces higher leaving water temperature and works with a higher suction temperature.

Even when combined with the downstream machine, which is producing colder water than a conventional system, most selections will show a saving in first cost (because the upstream chiller can produce more tons) from 2-4 percent while showing energy savings of 4-10 percent. And even with the extra waterside pressure drop of placing the two evaporators in series, the overall system efficiency will typically be significantly improved.

An important reason to turn to series chillers, in addition to reducing both first and operating cost, is to facilitate the use of variable primary flow systems. In the case of a common application with two chillers - with series flow chillers - when the second chiller comes on there is no change in flow through the chiller that is running.

Contrast this with a parallel flow application on a common pump set. When the second chiller is activated, valves on the second chiller are opened and the flow through the first chiller is reduced. But it’s a race to make sure the second chiller can be brought on-line and both chillers stabilized before they trip on safeties or a chiller barrel freezes.


Typically almost half, or more, of the energy consumption of an HVAC system comes not from the refrigeration side, but from the airside of the system. That’s why these designs frequently are extended to the airside using supply air temperatures in the 45-48° range for chillers and 50-52° for packaged equipment.

The goal is to optimize the energy consumption of both the airside and the refrigeration side of the system. In our experience, if you have a very efficient centrifugal chiller the right balance supply air temperature (SAT) is around 45°.

If somewhat less efficient screw chillers are used, 48° SAT frequently offer the best balance. If scroll compressors are used, especially in air-cooled applications like Rooftop/VAV, the right balance may be in the range of 52°.

These technologies and best practices are delivering for both business and the environment every day. Building owners and facilities managers are reducing operational cost, raising energy efficiency, and minimizing the impact of building emissions and energy use on the environment. They are able to sustain and document these improvements, and meet this challenge cost-effectively, thanks to these state-of-the-art, environment-friendly HVAC systems. Optimum building performance is here.

Sidebar: Downdrafts, Condensation

Two considerations in low air temperature applications are cold downdrafts and condensation. There are millions of square feet of space in humid climates such as Florida, Texas, and Georgia that are successfully using low air temperature applications.

For cold downdrafts the key lessons are:

• Use parallel fan-powered VAV boxes on the perimeter.

• Use cooling-only DDC/VAV boxes on the interior.

However, importantly, use linear slot diffusers with aspirating characteristics on both the interior and perimeter VAV boxes.

For condensation the key lessons are:

• Cold surfaces must be kept inside the humidity-controlled envelope. Air condition the equipment room and use nonducted returns where codes allow.

• Night setback and morning pull-down controlled off interior dew point sensor. Don’t pull the system down all at once; do it following a planned process controlled with input from an interior dew point sensor.

• Positive building pressure is critical for no condensation and IAQ. Make sure your building automation system has the ability, ideally on a floor-by-floor basis, to maintain a slight positive building pressure.

• Design the P-traps and pitch the condensate drains correctly. These systems will produce a good deal of condensation and, especially for draw-through fan applications, it is important to design the P-traps correctly and to pitch the condensate drains properly on all systems.

• Install a vapor barrier with reasonable construction. This is a basic that is important to check.

Publication date:05/12/2008