Airflow is one of the aspects of hvac that is critical — perhaps especially so in heat pump installations — but it is poorly understood, partially due to the cost of the required tools. This article is meant to help bridge that gap.

One of the prevalent problems that I see in the field is that many of the tools necessary to perform truly top-notch work are prohibitively expensive for the average technician or installer. This discourages field servicers from learning and implementing some of the more fundamental and critical aspects of field service.

First, let’s look at some basic principles of airflow in the hvac system.

Airflow is the medium by which we transfer heat energy from the conditioned space to the conditioning equipment and vice versa. The more air is moved to and from the conditioning equipment, the more heat can be moved.

Figure 1. A digital velometer and a homemade flow hood for measuring airflow.

The Basics of Airflow

Two aspects of airflow should be considered by people selling, applying, installing, and servicing air conditioning and heating equipment:

1. There must be adequate airflow to transfer enough heat energy to heat and cool the conditioned space.

2. There must be adequate airflow to allow enough heat energy to flow through the heat exchanger of the conditioning equipment.

Let’s begin with a discussion about adequate airflow to the heat exchanger of the conditioning equipment.

Heat pumps and air conditioners are particularly sensitive to abnormal airflow volume. An air conditioner, for example, is designed to operate per the published capacity and efficiency at 400 cfm per ton of cooling.

Inadequate airflow across its indoor coil diminishes its capacity and causes it to remove more moisture (latent heat) and less sensible heat from the air. Excessive airflow increases its cooling capacity but causes it to remove more sensible heat and less moisture from the space. In humid climates, more latent cooling is desired and in dry climates, more sensible cooling is desired. There is nothing good or bad about this, as long as the air conditioner is meeting the comfort needs in the conditioned space and will operate reliably.

Too much airflow is usually not the problem in our industry and when a system has too much airflow, it’s a relatively easy fix. Due to the cost-driven nature of our business, inadequate airflow is a more likely problem and it plagues many installations.

Inadequate Airflow

Inadequate airflow causes a loss in total capacity, a loss in sensible capacity, a loss in efficiency, and a reduction in reliability (due to refrigerant floodback to the compressor and distorted refrigerant system feedback to the servicer).

In 1996 a study commissioned by Arizona Public Service Co., the utility investigated total airflow in newly constructed houses in Phoenix. The study revealed that:

  • The airflow in 14% of the tested homes was at 90% of nominal airflow (360 cfm/ton).
  • 39% of the homes tested at 80% of nominal airflow (320 cfm/ton).
  • 14% tested at 70% airflow (280 cfm/ton).
  • 7% tested at 60% of nominal airflow (240 cfm/ton).
  • Of course, this sample does not necessarily represent the entire country, but I suspect that Phoenix is not the only place that suffers from this phenomenon. A test by Texas A&M University (Table 1) shows the ramifications of low airflow on capacity and efficiency. Notice that a 25% reduction in airflow (300 cfm/ton) causes a 7.5% reduction in cooling capacity and a 4.2% reduction in efficiency.

    This gives you some idea of what it costs the consumer to have a system with low airflow. Of course, the consumer also pays a high price in reduced reliability; 60% of the homes tested in Phoenix had a total airflow of less than 350 cfm per ton. Why does this matter? Because besides poor performance issues, 350 cfm/ton is the minimum airflow recommended by most manufacturers.

    Figure 2. A dominant return leak.

    Digital Velometer

    The less airflow a system has, the more vulnerable it is to failures of the compressor and compressor start-assist components due to refrigerant floodback. Low airflow also causes the refrigerant system to feed misleading information to the service technician. Low airflow lowers refrigerant pressures and superheat and raises system subcooling.

    Based on this information, inexperienced techs who rely mainly on pressures to evaluate a refrigeration system tend to overcharge a system, exacerbating refrigerant floodback to the compressor. Even more knowledgeable techs who use superheat and subcooling to evaluate a system will tend to undercharge with refrigerant, exacerbating performance problems that already exist as a result of low airflow.

    The net effect of low airflow is premature equipment breakdowns. Until recently, it has been very difficult for service technicians to incorporate airflow diagnostics into their daily routine because accurate airflow measurement tools are generally too expensive for the average technician to own.

    Recently, a tool has come out on the market that solves the expense as well as the accuracy problem. It’s called a digital velometer.

    Digital velometers measure the velocity of airflow, are available for as little as $100, and are more than accurate enough to serve the needs of field service and installation people. Figure 1 represents the one I use.

    Digital velometers measure the velocity of airflow in feet per minute (fpm). Although cubic feet per minute (cfm) is what is needed to evaluate airflow, there is an easy and inexpensive way to get these velometers to measure in cfm. All you have to do is build an airflow hood (Figure 1).

    Airflow Hood

    The one I’m currently using is made out of a cardboard box from a moving company. Of course, it can be made out of plastic or ductboard for more durability. (I’ve tried making them out of sheet metal, but they are just too heavy to use.)

    One end has to have a 12- by 12-in. square inside dimension; the other can be as large as you want. If you place this flow hood over a grille and measure the velocity of the air in fpm as it flows through the 12- by 12-in. opening, the fpm readout will be the same as the cfm flowing through the opening. Technically, the face area of the 12- by 12-in. opening should be sized at 144 sq in. plus the face area of the velometer you are using.

    Don’t get me wrong; I’m not saying this Rube Goldberg flow hood is as good as a store-bought version. What I’m saying is that it is accurate enough to diagnose 95% of the problems technicians typically encounter in residential and light commercial installations.

    1. When measuring the airflow through the flow hood, place the flow hood over a register with the 12- by 12-in. end away from the register. The air moving through the register will now be flowing through the flow hood.

    2. Leaving the velometer in the off position, place it in the airstream, allowing the propeller to come up to speed.

    3. Turn the velometer on (it should have previously been set to average fpm mode). The velometer will immediately begin to average the velocity of the airflow through the hood.

    4. Traverse the velometer across the opening of the hood to get an accurate reading of the average airflow through the entire face of the hood opening. As you can see in Figure 1, I use my velometer mounted on a stick so that my hand will not add restriction to the 12- by 12-in. area. This will give you the velocity, and more importantly, the cfm flowing through the grille.

    5. With the air conditioning system operating, measure the airflow out of each supply register; add all the readings together to get the total supply airflow.

    6. Now measure the airflow into each return register and add these together for the total return airflow.

  • If the total airflow of the supply system is close to the total airflow of the return system, the ductwork is probably intact and you probably have a good idea of what the total airflow is throughout the system. It should be 400 cfm per ton.
  • If the total airflow is short, more investigation is needed. We will talk about that later in this article.
  • If there is a significant difference between the return total and the supply total, there is duct leakage. If the return total is more than the supply total, the system has dominant supply leakage. If the return total is less than the supply total, the system has dominant return leakage.
  • What I mean by dominant leakage is that both the return and the supply sides could be leaking, but one of the sides has leaks that dominate over the other side. For instance, the return side could have 200 cfm worth of leaks and the supply could have 100 cfm worth of leaks. Inside the house this would show as a 100-cfm dominant return leak (Figure 2).

    If you find that a system has inadequate airflow, the next step is to determine why.

    Of course, the obvious thing to do is inspect the duct system for restrictions such as crushed ductwork, dirty filters, and dirty evaporator coils. Also, check for dirty blower wheels, slipping belts, slow-running motors, or misadjusted drive pulleys. Correct any of these problems and recheck the total airflow.

    If the airflow is still inadequate, the next step is to check the external static pressure (ESP) of the duct system. For this you need a manometer, which can read the pressure in the ductwork. By reading the pressure in the ductwork, you can determine if the ductwork is too restrictive for proper airflow and which part of the ductwork is causing the problem.

    Personally, I like to use a device called a Magnehelic gauge. These are made by Dwyer Instruments and can measure the pressure in the return side and the supply side simultaneously.

    There are two different types of pressure in the ductwork: velocity pressure and static pressure. Static pressure is the one that helps us determine if the ductwork is restrictive.

    To measure static pressure accurately, you need to use a device called a static pressure pitot tube. This is the pickup device that is inserted into the airstream. It is connected by a hose to the Magnehelic, which will give you the pressure at that point in the ductwork. Two pitot tubes are required.

    The return external static pressure is measured as the air enters the return opening of the equipment. The supply external static pressure is measured just outside the supply opening. Try to find the least-turbulent air to take the readings.

    By measuring the static pressure in the supply and the return, it is easy to determine if the ductwork is excessively restrictive. Generally speaking, the total resistance to airflow in residential ductwork (not including high-velocity systems) should be in the neighborhood of 0.5 in. of water column or lower (0.5-in. wc).

    Note, however, that this is a very generic number. Different air handlers perform differently, so don’t take this number too literally. Some machines won’t move 400 cfm/ton at 0.5-in. wc. If you have access to the application literature for the particular system you’re evaluating, this of course will be very helpful, but many times the paperwork is not available. The higher the ESP of the ductwork, the less air it will move. The lower the ESP, the more air it will move.

    So, if I’m trying to diagnose a ductwork system that has low total cfm, I will measure the ESP of the system while the fan is running. If it is 0.7-in. wc, I have a pretty good idea that the ductwork is too restrictive for proper airflow.

    The next step is to determine if the problem is in the return side, the supply side, or both. I will remove the pitot tube from the return and measure the supply only, making a mental note of the supply static pressure. Then I will reverse the procedure to determine the return static pressure.

    Generally, the return pressure in a properly operating duct system will be much less than the supply pressure. See Table 2 for a guideline of return-to-supply pressure ratios. By comparing the individual return and supply pressures, I can determine which side is causing most of the restriction. Let’s say it’s the supply side causing the problem. I can insert my pitot tube into the supply plenum and watch the results of any modifications I try.

    Flex Duct Specifics

    In the case of flex duct systems, sometimes just moving ductwork around can solve the problem, and that can be observed on the Magnehelic in real time. Common flex duct problems can include:

  • Turn radiuses too tight;
  • Too much flex in a run (installer did not cut off extra flex);
  • Flex not supported adequately, causing pinching at supports;
  • Flex run diameter too small; and
  • Flex runs not pulled straight and tight.
  • Everyone knows that flex duct is supposed to be pulled straight, but many installers fail to do so because it can kink the turns. The solution to this is to strap the flex to the structure entering and leaving any turns, then pull the flex tight between strapped turns.

    Another common problem is failure to support the flex properly. If there are not enough straps to support a suspended flex run, it will crush itself at the supports that are in the run. The rule of thumb is to support flex every 5 ft with a strap at least 1 ½ -in. wide. This support guideline works best if the turns are strapped and the flex is pulled tight.

    Limited Space

    Because builders are leaving less and less room for equipment, it is becoming increasingly difficult to attach the return duct to the air handler properly.

    This is a big no-no. Not only can this create excessive resistance to airflow, it can adversely impact air distribution across the face of the indoor coil and/or through a furnace heat exchanger. If the return has to approach an air handler from the side, the air needs to enter the air handler straight. If there is not enough room to accomplish this with the ductwork, a return plenum can solve this problem. The plenum allows the air to straighten out before it enters the air handler.

    Balancing Airflow

    Now that we’ve covered total airflow to the unit, let’s talk about moving enough air to the conditioned space. Let me preface this by saying that the proper way to determine how much air needs to be supplied to a space should be determined by a room-by-room heat load calculation.

    When you’re on a job trying to solve an airflow problem, heat load calculations are not always going to be possible. Here are some guidelines to get through many of those circumstances.

    The rule of thumb for airflow delivery to a conditioned space is 7.5 air changes per hour. This varies depending on the overall insulation quality of the structure and regional weather. If you use the 7.5 rule to get a general idea how a ductwork system is operating, this is how you would do it:

    1. Determine the internal volume of the suspect room. Let’s say the room is 10 by 12 by 10 ft, with an internal volume of 1,200 cu ft. That volume of air should be replaced 7.5 times per hour. Multiply the internal volume (1,200 cu ft) by 7.5. This will tell you the total air volume that must move through that room’s registers per hour (in this case, 9,000 cfh). Divide this number by 60 min for the room’s required cfm (in this case, 150 cfm).

    Of course, you have to correct for solar loads and such. By being able to easily check the airflow out of each register in the space, you can decide which registers do not have enough air and which ones have too much. Ideally you want to balance the registers by closing off the ones with too much air in order to push that air to registers that need more.

    You may have noticed that when you have tried to balance ductwork in the past, some systems will balance out easily while some won’t balance at all. One of the reasons for this is that the duct system could already be too restricted to deliver adequate air to condition the space, so moving air from one room to another leaves some rooms low on air. When you are attempting to balance register airflow and there is plenty of total air to work with, you are more likely to be successful.

    The bottom line is this: If you are attempting to get more air into a room and the external static pressure of the ductwork is low (plenty of total air), you can move air around by restricting registers that have too much air. If you are attempting to move more air into a room on a duct system that is too restrictive (inadequate total air), the solution is to relieve the backpressure in that branch and/or in the total system.

    An important thing to remember is that the rules of thumb I’ve given you are just that, rules of thumb. Know when it’s appropriate to use them and when it’s not.

    Leonard is president of Total Tech HVACR Training, Phoenix, AZ. His firm specializes in service, installation, and application training for service technicians. He can be reached at 602-943-2517.

    Sidebar: Pinpointing Leaks

    If you have a commercially made blower door, pressure pans, and manometers, you can spend a few hours to find the location of all the leaks from within the house. Although these tools are invaluable under many circumstances (and the resulting repairs can be quite lucrative if you market them appropriately), I find that for most typical jobs, a visual inspection of the ductwork will usually turn up most of the leaks.

    Typical return leak locations are:

  • Unsealed plenums under vertical air handlers or furnaces;
  • Return air routed through any infrastructure cavity;
  • Any poorly sealed connections joining the return trunk to the register box or the air handler plenum; and
  • Poorly sealed air handlers.
  • Many air handlers and furnaces leak like sieves, particularly some newer, flimsier ones. If the newer air handlers and furnaces are not hung properly, they can sag or twist, causing the door seals to leak. I always tape up air handlers to minimize leakage.

    Typical supply leak locations are:

  • Poorly sealed air handlers or coils;
  • Poorly sealed duct connections; and
  • The register box connection to the register.
  • If the supply register box is not sealed to the conditioned envelope barrier (floor, ceiling, or wall), the air exiting the box can strike the inside of the register and deflect backwards around the outside of the register box, into the infrastructure of the building. You might be surprised at how much air escapes this way.

    Publication date: 05/21/2001