TECHNICAL FEATURE

This article was published in ASHRAE Journal, June 2014. Copyright 2014 ASHRAE. Posted at www.ashrae.org. This article may not be copied and/or distributed
electronically or in paper form without permission of ASHRAE. For more information about ASHRAE Journal, visit www.ashrae.org.

Technical vs. Process Commissioning

Final Exam: Functional

Performance Testing
BY DAVE MCFARLANE, MEMBER ASHRAE

So far in this series on technical commissioning, we have examined commis-

sioning functions and practices associated with the planning, design, and
construction phases of a new building project. We have outlined the owner’s
expectations in the Owner’s Project Requirements (OPR) document, and we
have reviewed and approved the Basis of Design (BOD) document, which
shows how the design meets the requirements of the OPR.
In addition, we have created a plan that outlines the Importance of the Issue Log
tasks everyone involved in the project must complete Up ’til now, at each step in the commissioning process
to ensure the project meets its OPR. We have also moni- either the commissioning authority (CxA) or the con-
tored the construction process to ensure that all equip- tractor has carefully documented issues of concern in
ment and systems were properly installed, and we have the project issue log (which we have not discussed previ-
verified that all equipment and systems can be properly ously due to space limitations).
started and run. But the issue log is more than just a chronicle of prob-
The next phase of technical commissioning—func- lems. It also should include the date the issue was first
tional performance testing (FPT)—is the “final exam” a identified (and by whom), a description of possible solu-
new building must pass before it can “graduate” and be tions, who is responsible to correct the issue, and the
turned over to a satisfied owner. date the issue was corrected (and by whom).
This is the seventh in a series of bimonthly articles that explain the technical commissioning process for The CxA must maintain the issue log, and a significant
new buildings. Some of these articles’ content is based on ASHRAE Guideline 0-2005, The Commissioning part of all commissioning-related meetings should be
Process (published 2005) and the National Environmental Balancing Bureau (NEBB) publication Proce-
dural Standards for Whole Building Systems Technical Commissioning for New Construction (revised April devoted to discussing possible resolutions, assignments,
2013). In addition, some of the information in this article series has been adapted from an unpublished costs, and completion dates for the items in the log. These
NEBB standard titled NEBB Standard Owner’s Project Requirements (OPR) Guideline (draft dated June 20,
2011); and from NEBB’s Procedural Standards for Building Enclosure Testing (March 2013).
discussions are vital, because the CxA should resolve

Dave McFarlane is principal project director at Atkins in Fort Myers, Fla.

14 A S H R A E J O U R N A L   ashrae.org  J U N E 2014
 TECHNICAL FEATURE 

issues as they come up, during any phase of the construc- confirmed that all equipment and systems have been
tion process, before moving to the next phase. properly installed and are fully functional. We’re now
For example, FPT cannot begin until all equipment ready for the project’s final exam: real-world, functional
and systems are verified to start up and function prop- performance testing of every building system. Our goal
erly during pre-functional testing. So when key startup- is to ensure that the building itself (not just the design)
related items are entered into the issue log but not cor- satisfies the OPR, so we must ensure that the occupants
rected, FPT will probably fail and need to be repeated find the internal environment to be efficient, healthy, and
after the issues are corrected. comfortable. We’ve given all relevant contractors their
So pay careful attention to the issue log, and make sure FPT forms, and we’ll now work with them during FPT,
all issues are resolved before FPT begins. observing and assisting as they perform every test.”
This “test-and-verify-everything” perspective is a big
Role of Functional Performance Testing difference between process commissioning and tech-
It’s important to note that ASHRAE Guideline 0-2005 nical commissioning. Process commissioning permits
requires functional performance testing. But the contractors to serve as their own SMEs, and allows them
Guideline does not describe the number or type of tests to to perform their own tests. However, technical com-
perform, nor does it define how stringent the tests must missioning requires the CxA to serve as the SME—and
be or even who should conduct them. This article is an as such the CXA either performs or supervises all func-
attempt to “flesh out” the Guideline in a practical manner, tional performance testing.
and from a technical commissioning perspective.
This article also assumes that every issue in the issue Test Example: A Variable Air Volume Box
log has been corrected, all pre-functional tests have been To illustrate how the FPT process works, this article uses
verified complete and successful, all building systems and the example of a single variable air volume (VAV) box
components are properly installed and functional, and all serving a single space. But note: Proper technical com-
valves and dampers open and close as required. missioning requires that the CxA measure and verify the
At this point, you may ask, “OK, if everything is well correct operation of all control loops—not just a sample.
designed, installed correctly, and fully functional—and It is true that the CxA is permitted to perform limited
all known issues have been corrected—what else is there sampling during pre-functional testing (refer to the pre-
to do?” vious article in this series). But no sampling is allowed
This is an important question, as my experience indi- during FPT. Instead, all systems and equipment must be
cates that a traditional process-commissioning CxA actively verified for real-world operation.
would give a very different answer than would a tech- You may ask, “What’s wrong with sampling? It can save
nical-commissioning CxA. I’ve found that a traditional time and potentially reduce overall costs.” The problem
process-commissioning CxA would likely answer: is sampling can work well for “homogenous” systems—
“You’re right. We’ve prepared the FPT forms, and the such as a batch of well-mixed concrete: Every sample
contractors have informed us that all of their testing was taken has a high probability of having the same chemical
successful. We’ve verified that all of the commission- composition as any other sample. So if one sample meets
ing forms have been completed, and we’ve randomly specifications, the rest of the batch should also meet the
sampled the building’s equipment and systems to verify specs.
correct installation and proper operation.” But mechanical systems are never homogenous. Why?
If our project employed traditional process commis- Every system has probably been touched by multiple con-
sioning, it would be complete; and we would be ready to tracting firms and multiple technicians. Our sample VAV
turn over the new building to its owner. box, for example, has had electrical, piping, sheet metal,
But when faced with the same question, a technical control, test-and-balance, and chemical-treatment con-
commissioning CxA will likely answer: tractors involved in its installation. Also, different employ-
“Our commissioning team has done a great job so far! ees of the same firm may build or install each of their com-
We’ve verified that the design satisfies the OPR. We’ve cor- ponents and systems differently. So it is likely that every
rected every issue in the issue log, and we’ve personally VAV box will respond differently to control inputs.

J U N E 2014  ashrae.org  A S H R A E J O U R N A L 15
TECHNICAL FEATURE 

During the “final exam,” each control loop FIGURE 1   VAV box control sequence is too tight.
of every system is checked to make sure its
80
control sequence actually does what it is
supposed to do. This is much more than 76

Temperature (°F)
simply testing to see if a valve opens in
72
response to a call for heat, or a reheat valve
closes in response to a call for cooling. In 68
fact, from a technical commissioning per-
spective, those tests would have already 10 20 30 40 50 60 70 80 90 100 110 120
been performed during the operational per- Time (Minutes)
formance testing (OPT) phase.
Instead, the FPT phase calls for hands-on FIGURE 2   VAV box control sequence is too loose.
verification that every controlled device 80
performs all of the functions it is required to
76
Temperature (°F)

perform, including:
•• The performance of all required steps it 72
is supposed to perform.
68
•• A controller/actuator response speed that
is appropriate for the application.
•• A stable response. 10 20 30 40 50 60 70 80 90 100 110 120
•• All events are controllable to the setpoint. Time (Minutes)
•• All adjustments take place automatically
FIGURE 3   Properly adjusted VAV box control sequence.
through the building automation system.
Figure 1 illustrates a control sequence that 80
is too tight. The figure shows a tempera-
Temperature (°F)

76
ture that fluctuates hourly between 76°F
and 68°F (24°C to 20°C). The problem is 72
that when the temperature drifts from its
68
setpoint, the controller causes the tempera-
ture to cycle unacceptably. If a thermostat
controls this VAV box, the air flow damper 10 20 30 40 50 60 70 80 90 100 110 120
and heating control valve responses are too Time (Minutes)
rapid, which causes the cycling—and I can
practically guarantee that the occupants will complain quickly enough—nor does the airflow increase at a rate
that it’s too hot or too cold. sufficient to maintain the setpoint.
But when called to resolve the problem, the building In this case, when occupants complain, the mainte-
maintenance staff may not be able to properly diagnose nance staff may simply see the temperature as too high
it because the temperature is fairly close to the setpoint or too low. So to resolve overcooling, they reduce air
much of the time. So the maintenance staff leaves, hav- flow; to resolve overheating, they throttle back valves.
ing done nothing—except to accuse the occupants of But neither of those solutions is the proper fix.
being oversensitive. Figure 3 reflects proper system calibration, adjustment,
Figure 2 illustrates a different challenge: A controller and tuning: When the setpoint is adjusted, the VAV airflow
that does not respond quickly enough. The problem is damper and the reheat control valve respond correctly, the
when the room temperature is below its setpoint, the temperature reaches the control setpoint within a reason-
reheat valve or air flow damper does not change quickly able time interval, and afterward the temperature is stable.
enough to meet the setpoint; and when the temperature In this case, most occupants would feel comfortable, as
is above its setpoint, the heating valve does not close OPRs often call for a temperature of 72°F (22°C), ±2°F (1°C).

16 A S H R A E J O U R N A L   ashrae.org  J U N E 2014
Advertisement formerly in this space.
Advertisement formerly in this space.
 TECHNICAL FEATURE 

So for the “final exam” for our sample VAV box involves controllers with pneumatic differential-pressure sensors,
verifying that: thermostats, and control valves.
•• Control sequences generate the appropriate tem- In one of the processes I oversaw, an explosion could
perature control responses; result if the chemical mixture got too hot. If the tem-
•• Control sequences adjust the temperature to the set- perature were too cold, the proper reaction would not
point; occur. In another process, the final result was a material
•• All sequences yield stable results; that could be either too brittle or too soft, depending on
•• All sequences occur automatically through the con- chemical flow rates.
trol system;
•• The box reacts properly to a
power failure; and
•• Room occupancy/CO2 sensors
are properly set to open the air inlets
in order to maintain proper CO2 lev-
els and room temperature.
Under the process commissioning
model, I have seen a contractor simply
set a thermostat to full heat and then
to full cool; and that was submitted as
the FPT. But this kind of test does not
measure response time or stability,
which is why such “tests” are unac- Advertisement formerly in this space.
ceptable for technical commissioning.
Space limitations do not permit me
to describe the FPT requirements
and processes for all of the systems
and components for an entire proj-
ect. Suffice it to say that the CxA
must perform functional perfor-
mance tests on every control loop in
the project, which encompasses all
of a building’s major systems and
equipment, including boilers, chill-
ers, air handlers, cooling towers,
pressurization systems, and so forth.

Importance of Control Loop

Adjustments
In the 1970s, I worked as a chemical
process engineer. Part of my job was
to ensure pumps started and valves
opened and closed as required. We
used tuned control loops to main-
tain chemical flows, pressures, and
temperatures within tight con-
straints. This was before the advent
of computerized controls, so we used

J U N E 2014  ashrae.org  A S H R A E J O U R N A L 19
TECHNICAL FEATURE 

My coworkers and I became adept at checking and tun- controller to small but definitive changes from the wall
ing control loops, which measured and controlled three thermostat setpoint and measuring the results. In our
parameters, together known as PID: VAV box example, the system is started and run at a set-
1. The proportional (P) parameter, which measures point of 72°F (22°C) until stable.
present error; Next, the setpoint is adjusted to 75°F (24°C), which
2. The integral (I) parameter, representing the accu- should cause the box to begin heating, and the system’s
mulation of past errors; and response is measured by readings taken every minute.
3. The derivative (D) parameter, predicting future errors. In a typical VAV box, the damper position should throttle
We were required to maintain our control loops within back, the airflow should be reduced to the heating (or
specific ranges; and our PID tuning resulted in quick minimum) setpoint, and the reheat valve should open.
adjustments. We controlled temperatures, pressures, The CxA then makes sure the required data points are
and flow rates to the appropriate setpoints and stabil- logged, ensures that the temperature has adjusted to
ity requirements—thus yielding materials that met the 75°F (24°C), and verifies that the system’s response time
quality standards. conforms to either the OPR or to acceptable responses as
Today, computerized controls are commonplace in described in the Commissioning Plan.
HVAC systems, but the same tuning of control loops is Finally, the setpoint is lowered to 69°F (20°C), which
still essential. That’s because technical commission- should cause the box to begin cooling. As before, the
ing is about minimizing energy consumption as well as system’s response is measured via readings taken every
improving occupant comfort. minute. In a typical VAV box, the damper position should
However, the tuning of control loops is normally not open to the maximum or cooling airflow, the airflow
performed during traditional process commissioning. should increase, and the reheat valve should close. As pre-
It is common for the manufacturer’s “standard” control viously, the CxA also makes sure the required data points
loops to be left in place until there’s a complaint. I would are logged, ensures that the temperature has adjusted to
be a rich man if I had a dollar for every time I heard a 69°F (20°C), and verifies that the response time conforms
control technician say, “We don’t need to adjust the PID to either OPR or Commissioning Plan standards.
loops; they’re preset at the factory.” If performance is unsatisfactory in any way, the CxA
collaborates with the contractor to investigate the prob-
PID Loop Testing Process lem, determine the cause, and adjust the control loops
It is important to ensure that a system’s control loops as needed.
are properly tuned. But because PID loop tuning is done Testing all control loops is another difference between
so infrequently, it may be necessary for the CxA to brush FPT in technical and process commissioning. A typical
up on how it’s done, in order to help the controls con- process commissioning approach calls for the CxA to
tractor properly test the equipment and systems they’ve sample 10% to 15% of a building’s control loops and by
installed. extrapolation conclude that all systems work properly. In
As previously noted, proper technical commission- contrast, under technical commissioning the CxA verifies
ing calls for the CxA to oversee the measurement and that 100% of a building’s control loops work properly.
verification of all control loops. This objective is accom-
plished by making minor system setpoint changes and Identifying and Resolving Problems
then data-logging the controlled event. For the VAV box As noted in the previous article in this series, project
example, the controlled event is the space temperature, contractors have the ultimate responsibility for their
so the CxA must ensure these five data points are logged: own quality control. Therefore, all contractors need to
•• Space temperature; perform their own tests before the CxA either conducts
•• Damper position; or witnesses FPT.
•• Reheat valve position; For our sample VAV box, multiple contractors would be
•• Discharge air temperature; and involved in a successful testing process. It’s worth noting
•• Airflow volume from the box. that each contractor should have verified their work during
For each loop, the tests involve subjecting the pre-functional testing. However, that is an insufficient level

20 A S H R A E J O U R N A L   ashrae.org  J U N E 2014
Advertisement formerly in this space.
TECHNICAL FEATURE 

of testing in the world of technical commissioning—if for during the time that passes since pre-functional testing,
no other reason than the amount of time that often elapses including:
(weeks or months) between pre-functional testing and FPT. •• A plugged strainer (responsibility: chemical treat-
While our sample VAV box may have passed pre- ment contractor);
functional testing, conditions can change in the weeks or •• A manual valve having been closed (mechanical pip-
months before FPT. For example, the box will fail FPT if ing contractor);
there is an insufficient flow of hot water to the reheat coil; •• Improper testing and balancing (testing, adjusting,
and there are many reasons why the flow might be reduced and balancing contractor);
•• The control valve failing to open
(controls contractor); or
•• The VAV box fan failing to start
properly (electrical contractor).
In a previous article, I noted how a
project’s contract documents should
permit the owner to financially
penalize a contractor for testing
beyond a specified limit; and retests
required during FPT would fall under
the financial-penalty stipulations.
But contractors can avoid those pen-
alties by ensuring their systems work
properly before the CxA shows up on
Advertisement formerly in this space. site to oversee the final tests.

Fully Tested = Fully Confident

Functional performance testing is the
“final exam” that verifies all of a build-
ing’s systems work in the real world.
After all project issues are resolved
and functional performance testing
is complete, the technical-commis-
sioning CxA can look the owner in
the eye and confidently state:
“Your building works as you’ve
intended. We’ve verified that your
project requirements have been
fully satisfied. Through careful and
detailed hands-on testing, we’ve made
certain that the building functions as
designed. All known issues have been
corrected, and we’ve personally veri-
fied that 100% of the building’s equip-
ment has passed every test.
“Congratulations! Your building has
been through a comprehensive tech-
nical commissioning process. And we
know this because we’ve overseen or
conducted all of the tests in person.”

22 A S H R A E J O U R N A L   ashrae.org  J U N E 2014