August 2009 
Innovations in Comfort, Efficiency, and Safety Solutions. 
Techniques and technologies: the basics for beginners


Measuring airflow in a duct is a simple concept on paper. In reality it’s historically an oftentimes troublesome practice to implement properly and with any degree of accuracy. This article will discuss the theory behind the methods used in HVAC, and will describe the methods themselves, from the simple to the sophisticated. You can then choose for yourself the most appropriate method to implement for any given application.
But first, why do it to begin with? Well, there’s good old ASHRAE Standard 62.1 (the ventilation standard), which details the requirements for acceptable Indoor Air Quality (IAQ), and outlines strategies for implementing ventilation control. And since you can’t control what you don’t measure, airflow measuring stations are a crucial piece of the puzzle for implementation of this standard. Then there’s the whole LEED thing. With credits to be earned in the categories of Energy & Atmosphere and Indoor Environmental Quality, that either directly or indirectly call for metering and airflow monitoring, precision airflow measuring techniques are no longer a luxury reserved for the higherend systems, they’re a necessity at many levels!
To calculate velocity (FPM) and thusly airflow (CFM) in a duct, we need to be able to first measure what’s called the velocity pressure, or Pv. A pitot tube is a device that can be inserted into a duct, that has some holes in the side and at the end of it that sense pressure in the duct. The holes on the side pick up static pressure (Ps), and the hole at the end of it, pointed directly at the oncoming airflow, picks up total pressure (Pt). The equation relating total pressure to velocity and static pressure is Pt = Pv + Ps.
So how do extract a value for Pv out of this, so that we can calculate airflow? Well, if you rearrange the equation to solve for Pv, you see that velocity pressure equals the difference between total pressure and static pressure, or Pt – Ps. If you hook the pitot tube up to a differential pressure sensor with some poly tubing, connecting the total pressure port to the high side of the pressure sensor, and the static pressure port to the low side of the pressure sensor, the outcome is that the sensor does the math for you! And the resulting signal output from the sensor, that which can be read via an analog input at your Building Automation System (BAS), is proportional to the velocity pressure in the duct.
To get FPM and ultimately CFM from this information, the BAS needs to perform a couple of calculations. First is to derive FPM from Pv. It’s a simple calculation involving a constant, and a value for the “air density”. If the air density is assumed constant, then the calculation is a sole function of Pv. Now to get CFM from FPM, you need to know the cross sectional area of the duct. This value gets multiplied by the FPM, and viola, you have your CFM reading!
The pitot tube method of measuring airflow in a duct is fundamental yet rudimentary. The single point method of measuring total pressure leaves a bit to be desired. So manufacturers of airflow measuring products have taken their products to the next level. If you’ve seen a VAV box and its flow ring or flow cross, then you kind of have an idea of where I’m going with this. Generally the more sensing points that traverse the path of airflow, the more accurately the “average” pressure can be measured. So the simple pitot tube method has given way to the multipoint monitoring, selfaveraging pressure probe.
For large ducts with a good long section of straight duct, a multiple probe approach is among the best solutions. Each probe will have evenly distributed total and static pressure sensing ports. The larger the duct, the more probes required. This can vary upon application, and from manufacturer to manufacturer, but to get you in the ballpark you can expect to use one probe when the short duct dimension is less than 12”, two probes for duct heights up to 28”, three probes for duct heights up to 54”, and so on. The probes are inserted into the short duct dimension, typically the vertical dimension, equally spaced from each other and oriented in the airstream such that the total pressure ports are directly facing the oncoming airflow. Doing so ensures as well that the static pressure ports are properly aligned for maximum accuracy. For round ductwork, typically just two probes are needed, both of length equaling the duct diameter, and mounted in the duct perpendicular to each other. As far as minimum placement requirements, there needs to be some straight, unobstructed duct both upstream and downstream of the probes, up to 5x upstream and 2x downstream, where x is the diameter for round duct, and the “equivalent diameter” for rectangular duct, which when calculated turns out to be somewhere in between the two dimensions of the rectangular duct.
For applications in which there is not enough straight section of duct to satisfy the minimum placement requirements, there are products that are marketed as “airflow straighteners”, which are short sections of duct with honeycomb or eggcrate style straightening foils that can help obtain a more accurate measurement when placed in the duct upstream of the measuring probes. These don’t take up much space, as typically the whole assembly is no more than 8”12” wide, and they’re good for use in turbulent conditions, as might be immediately downstream of an elbow. Just be careful because they do introduce some static pressure drop into the system, so make sure that a calculation is performed that ensures the fan system can handle the additional static.
For highly critical applications, or for those of which the abovementioned straighteners are not an option (for whatever reason), the next available option is fan inlet probes. These are not unlike the duct probes, in that there are multiple averaging ports for total and static pressure. However as the name expresses, these probes get mounted at the inlet of the fan. This approach to airflow measurement is simpler to apply in that straight duct runs are not needed. A pair of probes is required for a single inlet fan, each being mounted on either side of the fan shaft, in the inlet “bell”, at its minimum diameter. This ensures that the probes measure the highest velocity of air, and obtain the most accurate measurement. For dual inlet fans, an additional pair of probes is required. In terms of accuracy, this method is arguably the best discussed thus far, however there are circumstances that could limit or even prohibit its application. Just be aware of the issues and the application guidelines, and you should be just fine.
The last method of airflow measurement that we’ll discuss is based on the concept of “thermal dispersion”. This method does not rely on pressure measurements to determine airflow. Rather, it does so using the laws of thermodynamics and the power of modern electronics.
Thermal dispersion probes are multipoint as are pressure probes, yet at each sensing point there are two thermistors, or temperature sensors, one that measures the ambient temperature of the air that flows across it, and another that is “selfheated”, meaning that it’s powered up to give off some heat. The temperature of the heated thermistor is regulated, and is known at zero airflow, as simply measured when there is no air flowing across it. The temperature of the thermistor then varies as a function of the velocity of the air moving across it: the faster the air, the more the thermistor cools. The variation is then measured electronically, and the airflow is calculated. Of course as you can imagine, the temperature of the air moving across the heated thermistor plays a part in the calculation as well, and so the job of the other thermistor is to continuously measure that air temperature, which is figured into the calculation. There’s more to it than that, but at least you can grasp the basic concept here. The resulting calculation for airflow is converted to an analog signal that’s proportional to the average airflow.
The science behind it is solid, and the technologies afforded in this day and age make it a very accurate method of measuring airflow, in practice, particularly at low airflows. Accuracy will be affected by other factors, and as with the pressure probes method, the larger the duct, the more probes are required in order to achieve specified accuracies. Minimum placement requirements apply here as well, although perhaps not as stringent as with the pressure probes (don’t quote me on that!).
Tip of the Month: Perhaps the best tip I can offer, in order to ensure accurate, economical airflow measurement, is to be proactive in terms of thinking about the requirements for airflow measurement. In other words, before or at the time the duct system is laid out, the thought should be in the head of the designer, as to where to physically locate the requisite flow measuring probes, such that the location abides by all the rules with minimal disruption to the design and physical layout of the duct system. Once that ductwork is in, if airflow measurement is an afterthought, you may be forced to “pay the piper”, and pop for the premium methods of airflow measurement. Again, some simple foresight can save some real headaches down the road, as is true with most things, I suppose! 
[Home Page] [The Automator] [About] [Subscribe ] [Contact Us]