Among the multitude of sessions at the ACI conference in Denver recently (a total of at least 180 sessions by my count), there was a very interesting half-day workshop on airflow testing hosted by Bruce Manclark and Paul Francisco. Their setup included simulated HVAC ducts of various shapes and sizes, and they used a Minneapolis Duct Blaster to provide air flow to test various measurement tools.
They demonstrated various methods of air flow measurement, including the use of flow hoods, large-vane and hot-wire anemometers, pito tubes, and, probably most interesting, the surprisingly accurate garbage bag method.
Before we get to the various measurement methods, a note on the “hosing” mentioned in the title of this post. In order to provide a specific amount of airflow to be measured, Bruce and Paul used the aforementioned Duct Blaster, connected to a manometer with a hose to provide air flow at a specific CFM to the ducts to be measured. While one attendee watched the flow rate on the manometer, several methods of measurement were demonstrated on rectangular and round supply registers.
Unfortunately, regardless of the method used, the amount of flow coming out of the registers never quite matched what was coming through the fan. After over an hour of testing and retesting, Bruce opined on the problem, suggesting that the Duct Blaster fan could be incorrectly calibrated. One of the attendees pointed out that whenever he has trouble with flow measurements he uses a different hose. Although the session leaders were skeptical, they agreed to give it a try.
In a matter of seconds the hose was changed, and like magic, the flow measurements started matching up and making sense. Although a little embarrassed by the episode, Bruce and Paul rallied and continued with their workshop.
Disregarding the hosing and resulting problems, one of the main takeaways from the session was that it is pretty hard to accurately measure airflow in duct systems and exhaust fans, something that reinforces my admittedly limited experience. Devices available to measure airflow include the previously mentioned anemometers, flow hoods, a pressure pan, a True-Flow air handler flow meter, and garbage bags.
Wide-vane anemometers can be moved around over grilles to obtain an average airflow or attached to a hood that fully covers a register, a method that appears to provide better accuracy. Hot-wire anemometers are used most often to measure pressure inside a duct through a hole drilled in the side, moved back and forth to obtain an average pressure throughout a duct.
Flow hoods appear to have varying degrees of accuracy, and can be difficult to hold in place. Pressure pans, used to measure exhaust flow only, are fairly simple, but it is critical to open the small door on them to the correct location for the airflow, and to match the manometer setting to obtain an accurate reading.
Simple and cheap = Accurate
Which leaves us with the garbage bag. By simply calculating the volume in a standard plastic garbage bag, and timing how long it takes to fill up or empty completely, you can fairly simply calculate the CFM of almost any supply or return register or exhaust fan. This method works best with two people – one to hold the bag and one to time it – but man, it sure is cheap and simple, and apparently, pretty accurate.
My takeaway from this session is that flow measurement is a difficult and somewhat inexact science. In my experience on building certifications, I have done a minor amount of flow testing, primarily of exhaust fans, and I have reviewed third-party test and balance reports provided to me by clients. In almost all cases, air flows are not close to the design and, since most residential ductwork is inaccessible, air flows are difficult if not impossible to adjust to meet designs.
I know that there are people out there meeting Energy Star 3.0 certification that requires HVAC systems and exhaust fans meet designed air flows. I would be very interested to find out if in fact they are really meeting the requirements, and how they are doing it.
I both look forward to and slightly fear the opportunity to test and verify these conditions myself on upcoming projects.