David Sellers

Damper tests provide valuable insights, Part 6; Minimum Outdoor Air Requirements and Building Leakage
June 25, 2007

After reading the previous post, some of you may be wondering how Abdi and Brent went about estimating the minimum outdoor air requirement for the Pacific Energy Center air handling unit they were working with. After thinking through the building and system, Abdi and Brent concluded that the minimum outdoor air requirement for the facility fell into three components; air required for ventilation purposes, air required to make up for the kitchen hood exhaust, and air required to ensure positive pressurization in the entry lobby.

Ventilation Air

One of the important realizations that Brent and Abdi had was that the occupant load in the PEC can vary significantly. On days with no classes, there may only be 20 or so people in the facility. But, if all of the class and meeting rooms are full, the population can easily expand to 100 or more people. As a result, they anticipated that the ventilation air required might vary over a range of values and estimated this range based on the minimum and maximum occupant count multiplied by 15 cfm per person. The 15 cfm per person number is a common rule of thumb that can be found in many places including Title 24 (the California Energy Code) and different editions of ASHRAE 62 (the ASHRAE Ventilation Standard). A survey of the toilet exhaust fans in the facility revealed that the air that was brought in for ventilation was generally taken out of the facility via the restrooms.

Kitchen Hood Make-up Air

Given the number of events that occur at the PEC, the kitchen located in the facility is used daily by the catering company that provides the food service for the various events. The kitchen hood represented a significant exhaust flow rate (nominally 1,000 cfm). Since there is no dedicated make-up air system for the kitchen, Brent and Abdi anticipated that the necessary make up air would need to be introduced via the AHU to ensure that it was properly conditioned. Otherwise, the operation of the hood could result in infiltration in the immediate vicinity of the kitchen.

Pressurization Air

The front door of the facility opens directly into a lobby area with educational displays and a reception desk. Brent and Abdi concluded that maintaining the lobby at a positive pressure would be desirable to minimize drafts and related discomfort for the receptionist during cool weather. Most of the time, this is accomplished at the PEC via the configuration of its economizer and relief system. Specifically, the AHU is provided with an economizer cycle, and barometric relief dampers and relief fans are provided to remove this extra air from the facility when the economizer is active. The dampers will not open until the building is positive and the relief fans are cycled based on lobby pressure. But, on a very cold day, with a low load, Brent and Abdi reasoned that the economizer might drive to minimum outdoor air and thus, the minimum outdoor air setting needed to reflect the pressurization requirements. They estimated the air required by using the “crack method”, a technique that relates infiltration to the length of crack around doors and windows. In formation about this approach and several other quick infiltration estimating techniques can be found in Part 8 – Infiltration Rules of Thumb of Arthur Bell’s book titled HVAC Equations, Data and Rules of Thumb, a really hand reference to have around if you are out in the field a lot or doing schematic design work. 

Total Ventilation Requirement

The table below presents the results of Abdi and Brent's estimate.

As a side note, a building pressurization test we performed at the PEC as a part of a different class revealed that the facility leaks quite a bit more than the “crack method”, or any method for that matter, would lead you to believe. The graph below depicts the results of a test we did at the facility. To perform it, we closed all of the doors, blocked the relief openings and shut down the relief fans. Then we used the AHU in a 100% outdoor air mode to incrementally pressurize the building; sort of a primitive blower door test.

I can explore the pressurization test and related resources on building leakage in a different string of posts if there is interest in it. I only mention it now because it relates to the pressurization estimate the class did using a classic technique and shows how easy it is to underestimate building leakage.

Based on their observations and estimates, Brent and Abdi developed a test that placed the AHU on minimum outdoor air and then simulated different supply flow conditions by driving VAV boxes into full cooling. For each condition, they planned to document the outdoor air flow, supply flow, and return flow. The goals of their test were to:

Identify if the desired minimum outdoor air flow rate was provided under all operating modes.

Identify if the operation of the kitchen hood impacted the amount of outdoor air that came in through the fixed minimum outdoor air damper position.

Determine if some sort of revised minimum outdoor air control strategy that could adjust for occupancy and kitchen hood operation merited consideration.

While Brent and Abdi did not get to run the full version of their test due to time constraints, they did obtain enough data to conclude that:

The current minimum outdoor air setting probably provided sufficient flow for the design condition, but over-ventilated during periods of low occupancy

Given the mild San Francisco climate, and the relatively low number of hours the system spent on minimum outdoor air, it would be difficult to justify a full blown, demand controlled ventilation strategy.  However, an approach that provided different fixed damper settings based on class room schedules may merit consideration.

As a result, they recommended additional testing and analysis using a procedure similar to the one they developed. Their recommendation became an action item for the class that followed, and their test became the foundation for the test Wayne and I ran. Come back in couple of days and we will take a look at the test procedure and begin to look at the results it generated.

Posted by David Sellers on June 25, 2007 | Comments (1)