Need help selecting the right controls? Talk to our specialists — response within 24 hours.

A Quality Inspector’s Checklist for Commercial HVAC Systems: What to Verify Before You Sign Off

Who This Checklist Is For

If you manage a commercial building, data center, or industrial facility and just had a new HVAC system—or a major upgrade—installed, you need this checklist. Not because your contractor is dishonest, but because even good installers miss things. I’ve been a quality compliance manager for Johnson Controls systems for over six years, and I review roughly 200 HVAC installations annually. In Q1 2024 alone, I flagged 18% of first deliveries for issues that would have cost the owner thousands in rework or energy waste. Below are the five things I check every time—plus one most people overlook.

Step 1: Check the Thermostat Integration (Especially Wi-Fi Models)

What to do: If you’re using a Johnson Controls Wi-Fi thermostat (or any zone controller), confirm it’s actually communicating with your building automation system—not just the local app. Walk to the unit, wait for the screen to show current readings, then open the back-end system (e.g., Metasys) and compare temperature and occupancy data.

Common miss: Installers pair the thermostat to the app but forget to register it in the central controller. You get a perfectly working local thermostat that can’t respond to demand limits or schedule changes. That’s like having a car with a fuel gauge that only works on the dashboard—fine until you need range data.

Checkpoint: Does the thermostat show “Connected to [System Name]” or a similar status? Can you change the setpoint from your central dashboard and see it update on the device within 30 seconds?

Step 2: Verify Oil Pressure Sensor Function

Oil pressure sensors on chillers and compressors are classic “silent killers.” They fail gradually, and by the time the alarm rings, you’ve already lost hours of runtime or damaged the compressor. Here’s the test:

  • Check the sensor’s baseline reading against the manufacturer’s spec (typically 15–25 psi for oil pumps).
  • Simulate a low-pressure condition—either by manually triggering the sensor (if safe) or by checking the diagnostic log for prior low-pressure events.
  • Confirm the controller shuts down the compressor below the threshold (usually 10 psi).

Real mistake I saw: A contractor replaced the compressor but reused the old oil pressure sensor. The sensor was drifting—reading 5 psi higher than actual. The system ran for three weeks without oil lubrication, and we caught it only during the first scheduled maintenance. Cost to redo: $14,000. The sensor itself was $120.

Checkpoint: The oil pressure reading matches the spec within ±2 psi. A proper shutdown occurs when pressure drops below the limit.

Step 3: Test the AC Compressor Properly

Everyone knows how to test an AC compressor with a multimeter—check resistance across terminals, look for continuity to ground. But that only tells you if the windings are dead. It doesn’t tell you if the compressor is actually moving refrigerant.

To verify performance:

  1. Measure suction and discharge pressures while the system is running.
  2. Check superheat and subcooling against the manufacturer’s target (usually 10–15°F for medium-temp systems).
  3. Listen for abnormal vibration or clicking. A healthy compressor hums; a failing one ticks or rattles.

Gut vs. data: I once saw pressure readings that looked perfect on paper—suction 68 psi, discharge 225 psi. My gut said something was off because the discharge line felt barely warm. Turns out the compressor valve was leaking internally; the refrigerant was just recirculating. A simple discharge temperature check caught it. The numbers told a story, but my instinct said check what the numbers didn’t show.

Checkpoint: Compressor amperage is within ±5% of rated full load amps. Temperature difference across the condenser coil is at least 15°F in cooling mode.

Step 4: Inspect Airflow and Fan Performance

Even the best compressor can’t condition a building if the air doesn’t move. Many technicians skip a proper airflow measurement and just assume the fan is fine. Do this:

  • Measure total static pressure across the supply and return. Compare to the fan curve.
  • Check that all dampers are open (I kid you not—two times last year a damper was left closed after installation).
  • If you have auxiliary fans (like Vornado-style units in server rooms), verify they’re wired to the correct fire shutdown relay. That’s a life-safety item, not a comfort item.

Checkpoint: Airflow (CFM) is within 10% of design. Fan motor amperage matches the nameplate.

Step 5: Validate Communication with Your Building Automation System

This is where Johnson Controls HVAC service really stands out—the ability to integrate chillers, heat pumps, and rooftop units into a single BACnet or N2 bus. But integration is only as good as the handshake.

Test: Send a command from your central system to override the setpoint by 5°F. Go to the unit and confirm the setpoint changed. Then adjust a local setpoint on the unit and see if the central system reflects it.

Transparency point: Some vendors quote a low integration price but exclude the cost of programming each device. I always ask, “What’s NOT included?” That question alone saved one client $18,000 on a 50-unit project—they realized the base quote only covered hardware, not commissioning.

Checkpoint: Every device shows up in your BAS with the correct name and point map. No “unknown” or “comm fail” statuses.

Final Risk to Weigh: The Overlooked Calibration Clause

One thing most facility managers skip: verifying the calibration certificates for sensors (temperature, pressure, humidity). Your installer may say, “They come calibrated from the factory.” That’s often true, but the factory calibration can drift during shipping. I’ve seen a batch of 20 pressure transducers that were all 0.5 psi off because of vibration during transit.

What to do: Ask for field calibration records, or better, spot-check two or three sensors against a known reference. The cost is about $50–150 per sensor. On a project with 100 sensors, that’s $5,000–15,000. But the cost of a false alarm or a misread that leads to an energy penalty—easily double that in a year.

To be fair, not every project needs 100% field calibration—it depends on the criticality of the space. But for data centers, labs, or hospitals, it’s non-negotiable.

One More Thing

If your contract includes a performance guarantee, check the fine print: some guarantees only kick in after a 12-month measurement period. That means you eat any energy waste for a year unless you verify everything upfront. Transparency builds trust. A vendor who shows you all the fees—calibration, integration, commissioning—before signing is more likely to deliver a system that actually works.

Leave a Reply