In the December issue, we started this discussion of HVAC design for typically humid climates similar to the American South. We broadened the definition of “hot-humid” to include parts of the country we might not have considered otherwise, and we touched on the building envelope and acceptable indoor conditions. Finally, we discussed ways of keeping moist air out of the building via pressurization and controlled intake.

This month, we will discuss what we do to dehumidify that outdoor air so that we don’t create problems when we distribute it through the building, and we will look at that distribution and the importance of pushing and pulling that air into and out of the right places. And we will wrap things up with the importance of planning, from schematic design through commissioning.


In most southern climes, the biggest source of moisture within a building has very little to do with what or who is in that building. If the envelope is tight, and the building is properly pressurized, the biggest problem is a code-dictated dilemma: ventilation air.

Luckily, the solution is pretty simple. In the most basic terms, you should bring in the minimum amount of outdoor air possible (while meeting ASHRAE 62.1 and pressurization requirements) and dehumidify it directly and constantly.

In ASHRAE 62.1-2004 a number of revisions were made that, on average, cut the outdoor ventilation rate by about 15% to 20% when compared to the 2001 version. For designs in humid climates, that’s a good thing. Less air means less moisture to wring out. But some folks don’t like the cut in the rate, which in an office environment cuts you from the old 20 cfm/ person to an average of about 17 cfm/ person.

The LEED® rating system dislikes the new numbers enough that it awards an additional credit (Increased Ventilation - EQc2) for an increase in the rate of at least 30% over the 2004 calculated values. They went with 30% because the folks at the USGBC would actually prefer a number 50% higher than the 2004 rate (about 25 cfm/ person), but a 30% bump was seen as a “compromise between indoor air quality and energy efficiency.” 1

Now, this is one of those LEED catch-22s that drive me batty. If you are designing in a humid climate, you could actually be rewarded for increasing the potential for mold and moisture problems! I know, I know, an intelligent designer should know better. But when a job goes “point shopping,” this is the kind of “easy” point pickup that can ultimately backfire on you.

A more progressive and safer approach would be to skip the LEED point and bring in the code minimum. Then, on projects that merit the added control complexity, apply a demand control ventilation strategy to actually cut that quantity down further whenever possible. You might even get that lost point back under the Optimize Energy Performance Credit (EAc1).


In most commercial mechanical systems, a single unit with a single cooling coil is used to provide both temperature and humidity control. In most mixing systems, the return air combines with the outdoor air upstream of a cooling coil and the mixture is cooled to a fixed supply air temperature for distribution to the spaces. More often than not, even if humidity is measured, it often isn’t directly controlled. Instead, drybulb thermostats rule the day.

In a humid climate, a number of problems can arise with such systems. If there is no humidity control sequence of operation, the cooling sequence may cycle on and off. With no cooling, unconditioned air mixes with the return air to create a mixture that may be just fine from a space temperature standpoint, but with an unacceptably high dewpoint. By the time the system reacts to the increased drybulb, rh levels have spiked.

If a typical cooling/reheat control sequence is employed along with a a single coil, the outdoor air may be safely dried out, but you’re still using more energy than is necessary. Why? Because the return air that was just a bit over room temperature has now been unnecessarily over cooled and must be reheated as a constituent of the mixed air passing over the cooling coil.

Instead, I suggest that you decouple the latent from the sensible cooling and do a better job on both. By decoupling, we aren’t committing to two separate AHUs, but we are committing to two cooling coils and addressing the outdoor air separately from the return air.

There have been numerous articles written about DOAS and the idea of decoupling sensible and latent cooling all the way down to the distribution into the rooms. Dr. Stanley Mumma is the most eloquent advocate for such an approach and has a website dedicated to the science of the DOAS at If you have the stomach for it, I encourage you to apply a DOAS. That’s not a backhanded compliment to Dr. Mumma and his ideas. It’s merely an acknowledgement that you will likely have to overcome a lot of skepticism regarding first costs and control complexity. Accommodating two complete systems in one building will almost always draw challenges from the peanut gallery.

Sometimes it’s easier to just fly under the radar. A relatively straightforward approach is to stick with one unit and two coils: a return air coil and a dedicated outdoor air coil. Together, the two cooling coils maintain the target condition in the zone. See Figure 1 for an example of a dual-path system I applied on a project in the tropics. Trane’s Engineers Newsletter2 provides an excellent discussion, but here are the particulars.

The humidistat directly controls the latent capacity of the outdoor air coil to maintain the desired rh, and in turn, provides sufficiently dry air to remove the latent load, from both outdoor air and the zone.

And the thermostat directly controls the sensible capacity of the return air coil, providing the balance of cooling needed to ensure that the supply air temperature satisfies the sensible load. As an added bonus, if no sensible load exists, the unconditioned return air can temper the preconditioned outdoor air, thus providing free reheat.

figure 1. An example of a dual-path system applied on a project in a tropical region.


So we have minimized the quantity of outdoor air, directly dehumidified it, and mixed it with the return air. Now it’s time to distribute our dry air into the building and back to our AHU.

While maintaining a building that is net positive to the outdoors sounds pretty clear-cut, things get a little fuzzy when you realize that within that net positive building you will be trying to negatively maintain some spaces in relationship to other spaces. Theoretically, this may be a simple concept, but air follows the path of least resistance, and if you aren’t careful you may create negative pressure microzones that could undermine an otherwise acceptable building pressurization strategy.

Why taunt the mold gods by placing negative pressure spaces where they could do the most damage? In particular, rooms like toilets, storage rooms, or any other room that might be constantly exhausted, should not be located on the building perimeter. When certain areas must be kept under negative pressure, they should be placed near the center of the building so that air traveling to the depressurized space is conditioned and the HVAC system has a better chance of maintaining the overall positive building pressurization.

With VAV systems, normally occupied spaces that may experience low loads when the rest of the building is at high load present some unique challenges. So before we talk about plenum return vs. ducted return, let’s touch on return air in VAV systems.


On a typical VAV system in a commercial or institutional environment, the supply and return systems do not control together. In more sophisticated installations, constant volume or VAV terminals may be employed on the return side, but more often than not, this isn’t the case.

In a classic VAV system, the supply varies with the load in the space. But the return varies as a proportion of the total flow of the return system, regardless of the supply to the space. So, it is a very real possibility that the supply to a room may be at its minimum setpoint (say 40%), while the overall return system (and in turn the return from the space) is at 75% of design.

To put numbers on this, say the supply box maximum setpoint is 1,000 cfm and the return grille in the room supplied by the box is 900 cfm. The state described above would supply 400 cfm at minimum with 675 cfm of return; the room pressure relationship is completely reversed from design. This condition could easily exist in an office or conference room that is vacant while the rest of the building is occupied.

The designer must recognize that he cannot apply diversity to a return system like he can the supply. The return total is always equal to the fan supply minus the outside air, the constant exhaust, and the air that escapes the building through exfiltration. That means the balance at the return grille must be in proportion to the maximum return available. You can only balance the system proportionally. To balance the return, set the supply system to maximum, with diversity set so that the supply boxes total the fan scheduled airflow. Set the outside air and exhaust to design, and then proportionally balance the return system at its maximum with diversity, similar to the supply system.

Once you have balanced the system for design maximum, put the AHUs into automatic mode and keep an eye out for trouble spots. Opportunities for excessive negativity at low load will be more prevalent at points in the return air system closer to the central return fan.

If the pressure relationship is critical, never put the space on a proportional VAV return system.

If the space pressure is less critical like a corner office or some other benign function, when and if the space goes negative because of proportional return, you can provide a safer makeup air path. Instead of the path of least resistance being through the exterior walls or windows, provide an air transfer assembly, in turn pneumatically linking the trouble space to the spaces around it.


One thing I love about facility systems engineering is the fact that there is so much individuality. You put five engineers in a room and you will likely get five different solutions to a problem. Each solution may be different, but they hopefully are all correct. The problems arise when two solutions conflict and are, in fact, mutually exclusive. Now you have to choose.

How you return air in any building, not just buildings in humid environments, is one of those areas where conflicting theories and experiences clash.

There is a school of thought that contends plenum returns are a simpler and more forgiving system. They are certainly less expensive and easier to balance with fewer dampers to fiddle with. And advocates say the negative pressure associated with plenums is so slight, that it is not a factor in moisture migration through the envelope.

But I have reached a contrary conclusion. I would suggest that plenum returns, while acceptable in simple buildings or in more forgiving climates, should be avoided in all cases when designing in the American South. ASHRAE’s Humidity Design Guide makes the same point no fewer than three times, and further instructs the designer to ensure that the return ductwork is sealed tight.3

Though plenum returns may be easier to balance from a physical labor perspective, they are impossible to balance from an operating standpoint. Remember what I said about potential negativity problems on proportional returns increasing as you near the central return fan? If you have no ductwork, then you have no means of correcting such problems closer to the AHU or ensuring that sufficient negative pressure exists remote from the AHU.

And if you agree that negative pressure on the building perimeter is a sticky wicket, then creating a negative pressure situation throughout the building, where the plenum interfaces with the exterior walls is certainly unadvisable.

If you must use a plenum return, I would suggest following the U.S. Government Services Administration's criteria of a partially ducted return.4 The horizontal distance from the farthest point in the plenum to a return duct intake should not exceed 50 ft. And no more than 2,000 cfm should be collected at any one return grille. But be prepared to discuss the envelope with the architect to ensure the walls are properly detailed to avoid moisture migration.


Finally, when it comes to HVAC design in humid climates, you must be proactive from the beginning. Understand the architect’s design and don’t take anything for granted when it comes to the building’s exterior. From the beginning, plan your systems with moisture control as the watchword.

Understand your sequence of operation, and be sure it is tailored for the project. Locate building pressure sensors on the drawings, and don’t assume the control contractor will locate them properly based on a narrative alone. As you develop the specifications, be sure that commissioning, startup, and final test and balance are sufficiently covered.

Once the project is awarded, don’t lose track of those important submittals you asked for, like the pre-TAB report, control floor plans indicating all device locations, startup checklists, and functional test procedures. Too often, these items get put off until late in the job, when there isn’t enough time to properly react to deficiencies.

Don’t assume local contractors, designers, and tradesman understand moisture control. The reasons problems exist in places like Miami and Houston are because the buildings were designed and/or built wrong … and quite often by folks in Miami or Houston. Just because a hard knock lesson may have been meted out, doesn’t always mean the lesson was learned.

Get to the site as often as possible, so that you aren’t a stranger when commisioning and startup begin. Once startup begins, get involved. Don’t wait for a phone call if problems arise. If you do, chances are the problem you are asked to address is actually a couple problems downstream of the original culprit. And be prepared to walk the systems through low load scenarios as well as peak load scenarios to ensure that dehumidification is constant.

In the end, you should have a building that is dry and healthy. Then, when folks occupy and operate your building and its systems, the term southern comfort will take on a whole new meaning.