“The system passed.” It is a reassuring sentence, and on many projects it closes a chapter. But a passing result and a properly performing building are not always the same thing.
What a passing test actually proves
A test result is a statement about a defined portion of the system, measured by a defined method, under defined conditions, against a defined criterion. A duct section that passes a leakage test at its specified test pressure has shown exactly that: this section, at this pressure, met this limit.
It has not necessarily shown that the whole system, under real operating conditions, delivers the airflow, pressure and distribution the building needs.
Where the gap opens
Partial coverage. Specifications often require testing of a sample, or only of certain pressure classes. Untested sections may leak differently from the tested ones, and the connections between sections are easy to overlook.
Timing. Many tests are performed at a particular construction stage, before ceilings close, insulation is applied or other trades work around the ductwork. Damage, disturbed joints and modifications that happen afterwards are not captured.
Test conditions versus operating conditions. A leakage test is run at a controlled pressure. A working system sees varying pressures, fan speeds, damper positions and thermal movement. Behaviour at test pressure is a good indicator, not a guarantee.
Isolated components. An air handler may be certified, a duct section may be tested and a balancing report may be signed, yet the interfaces between them (flexible connections, plenums, access doors, terminal connections) may be where the real losses are.
Acceptance criteria that do not match the objective. If the criterion is “leakage below a given class” but the real objective is “design airflow at every terminal,” passing the first does not prove the second.
From component results to system proof
Demonstrating real performance means connecting the individual results to the outcome that matters. A practical approach:
- Define the performance objective in measurable terms: airflow, pressure, leakage, temperature, energy.
- Confirm what each test covers and, just as importantly, what it does not cover.
- Measure at the system level, comparing fan airflow, branch airflow and terminal airflow.
- Investigate discrepancies between component results and system behaviour rather than accepting either number in isolation.
- Re-test after any corrective work to confirm the system, not just the repaired part, now performs.
This is the logic behind a Measure → Diagnose → Rectify → Verify process: the final step is only meaningful if the earlier steps were designed to prove the outcome, not simply to tick a box.
A closer look at what each result tells you
It helps to separate what common test results do and do not demonstrate:
Test or report
What it shows
What it may not show
Duct leakage test on a section
That section met its leakage limit at the test pressure
Leakage at untested sections, joints to equipment, or damage after testing
TAB report
Airflow at measured points when the report was written
Whether airflow has drifted, or whether leakage explains a shortfall
Equipment factory test
The unit’s capacity under rated conditions
How it behaves with the installed ductwork and real pressures
Commissioning checklist
That listed items were checked and signed
Whether the whole system meets the performance objective
None of these is wrong. Each answers a narrower question than “is the building performing?”
A short example
Imagine a commercial floor where the supply ductwork passed its leakage test during construction. Months later, occupants complain that perimeter offices are warm. Investigation shows that after testing, ceiling installers disconnected a flexible duct to move a light fitting and reconnected it loosely. The original test result was honest and accurate. It simply could not predict a later change. The only way to catch this kind of problem is to measure the finished system, not just the system at one point in time.
Questions to ask before accepting a pass
• What exactly was tested, at what pressure, and on which date?
• What proportion of the system does that test represent?
• Has anything been installed, moved or modified since?
• Does the acceptance criterion reflect the building’s real objective, or only a component requirement?
• Is there a system-level measurement that connects the result to delivered airflow and comfort?
How this fits with commissioning
Commissioning is the natural place to connect isolated results to whole-system performance. A thorough commissioning process reviews test records, witnesses key measurements, checks interfaces between trades and confirms that the systems operate together as intended. Where commissioning is limited or compressed, the risk is that isolated passes are accepted as proof of performance without anyone asking whether they add up.
What owners and consultants can do
Owners and consultants can reduce this risk at the specification stage by writing acceptance criteria around outcomes as well as components. For example, in addition to a duct leakage class, a specification can require measured airflow at the fan and at representative terminals, with stated tolerances, and a repeat measurement after any corrective work. This costs little compared with the cost of discovering performance problems after occupation.
Putting it into practice
The practical lesson is not to distrust test results, but to read them precisely. Treat each pass as a confirmed fact about a defined scope, then ask what additional evidence is needed to show that the whole system delivers what the building requires. Where that evidence is missing, a short, targeted round of system-level measurement is usually inexpensive compared with the cost of a comfort or energy problem that surfaces after handover.
Conclusion
A passed test is valuable evidence, but it is evidence about something specific. Owners, consultants and contractors who ask “what exactly did this test prove, and what did it leave out?” are far more likely to hand over a building that works, not just a building that passed.
