Functional performance testing

An HVAC system can be correctly installed, started up and balanced and still fail to perform. The economizer may never open, the boiler may not stage, the backup pump may not start when the lead pump fails, or the system may behave correctly in normal operation and fall apart in a power failure. Functional performance testing (FPT) is the part of commissioning that finds these problems by putting systems through their full range of operating modes and verifying that they respond as designed.

FPT is where the sequences of operation meet reality. This guide explains what is tested, how test procedures are developed and run, and how to avoid the common traps that make testing slow or unreliable.

What FPT verifies

Functional testing checks that equipment, controls and systems operate together according to the owner's project requirements, the basis of design and the sequences of operation. A thorough test covers:

• Start-up and shutdown sequences, including time delays and interlocks.
• Normal operation across occupied, unoccupied, warm-up and cool-down modes.
• Part-load operation, such as staging of chillers, boilers and fans, and reset strategies.
• Setpoint control and stability of control loops.
• Safeties and interlocks: freezestats, high static pressure limits, smoke detector shutdowns and flow switches.
• Alarms: generation, priority, routing and clearing.
• Failure modes: lead and standby changeover, response to sensor failure, loss of communication.
• Power failure and restoration: behaviour on emergency power and orderly restart.
• Integration with other systems, such as fire alarm, lighting, access control and generators.

Prerequisites

FPT should not begin until systems are ready. Starting early wastes everyone's time. Before testing, confirm that:

1. Construction checklists are complete and issues from them are resolved.
2. Equipment start-up is complete and documented by the manufacturer or contractor.
3. Testing, adjusting and balancing is complete, or at least sufficiently advanced for the systems being tested.
4. Controls point-to-point checkout is complete: every sensor reads correctly and every output moves the right device in the right direction.
5. Controls programming and graphics are complete and match the approved sequences.
6. The controls contractor has performed its own pre-functional verification and documented it.

The single most common cause of failed functional tests is incomplete controls work. Requiring the controls contractor to submit a completed self-test before the commissioning provider witnesses anything saves a great deal of time.

Writing test procedures

Test procedures, often called test scripts, are prepared by the commissioning provider, sometimes with input from the contractors, and reviewed by the design engineer. A good procedure is specific enough that any qualified person could run it and get the same result.

Structure of a test procedure

Expected responses should be measurable. "Supply air temperature controls to setpoint" is less useful than "supply air temperature stabilizes within plus or minus 0.5 K of setpoint within 15 minutes, without sustained oscillation."

Testing methods

Manual testing with overrides

The most common method is to simulate conditions by overriding inputs at the BAS: changing a sensor value, forcing a schedule or failing a piece of equipment. The tester observes the system response at the BAS and in the field. Overrides must be released and recorded at the end of each test.

Trend-based testing

Some sequences are best verified by trending relevant points over days or weeks of normal operation. Trend data shows how reset strategies, staging and economizer control behave under real conditions rather than simulated ones. Trend analysis is efficient for verifying many terminal units at once and for confirming stable control. See trend logging and using BAS data.

Sampling

On large projects with many identical units, such as hundreds of VAV boxes, a representative sample may be tested rather than every unit, provided the specification defines the sample size and what happens when a sample unit fails. A common approach is to expand testing if failures exceed a set threshold.

Running tests efficiently

1. Schedule realistically. Allow enough time and coordinate the trades needed: controls, mechanical, electrical and sometimes the manufacturer.
2. Test in logical order. Test central plant before air handlers, and air handlers before terminal units, because downstream systems depend on upstream ones.
3. Have the right people present. The controls technician must be able to make adjustments on the spot; operations staff benefit from witnessing.
4. Record as you go. Fill in results during the test, not from memory afterwards.
5. Log issues immediately in the issues log with responsible parties.
6. Retest promptly once corrections are made.

When a test fails

A failed test is not a disaster; finding problems is the point. The failure is recorded, the cause is investigated and corrected by the responsible party, and the test or the failed portion is repeated. Most specifications allow for one retest at the contractor's cost and charge further retests to the contractor, which encourages proper preparation.

Some failures reveal design issues rather than installation errors, such as a sequence that cannot work as written. These must be resolved by the design engineer with the owner's agreement. See common deficiencies found during commissioning for typical examples.

Seasonal limitations

Some tests cannot be done under realistic conditions at the time of turnover. Cooling plant performance is hard to verify in a Canadian winter, and heating and freeze protection are hard to verify in summer. Overrides can simulate many conditions, but full-load performance tests are often deferred to the appropriate season. See seasonal testing and deferred commissioning.

Frequently asked questions

Who performs functional tests?

Contractors operate the equipment and controls, and the commissioning provider directs and witnesses the tests. Operations staff should attend where possible.

How is FPT different from TAB?

TAB measures and sets flows. FPT verifies that controls and equipment respond correctly across operating modes. TAB must be largely complete before functional testing can produce valid results.

Key takeaways

• Functional testing verifies systems in every mode, including part load, failures, alarms and power restoration.
• Do not start until checklists, start-up, TAB and controls point-to-point checkout are complete.
• Test procedures should have measurable acceptance criteria tied to the sequences and OPR.
• Combine manual override testing with trend analysis for efficient, realistic verification.
• Plan seasonal tests for conditions that cannot be created at turnover. A control system verification applies the same methods to existing buildings.