Last Thursday, a returned thermostat landed on my bench in the Johnson Controls quality lab. White housing, intact LCD, warranty sticker still readable. The service note said: "Cooling mode won't engage. Customer requests replacement."
I'm the person who reviews these returns before we hand out credits. Roughly 200 units cross my desk annually, and in 2024 I've rejected about 12% of first inspections because the issue was in the installation, not the product. (That figure comes from our internal tracking—nothing formally audited, but consistent across the three years I've done this work.)
This unit looked clean. No burn marks, no water damage, no bent pins. So I pulled the logs.
A Thermostat Can Only Report What It Senses
Through the Metasys building automation gateway, the thermostat communicated without a single dropout for weeks. It reported room temperature drifting between 72 and 74 degrees every morning, holding right at setpoint. On paper, it was doing its job.
So why was the customer convinced it was broken?
I called Dan, the facilities manager at a 45,000-square-foot office building in the suburbs. I recognized his voice immediately—that tone people use when they've already tried everything and started doubting themselves.
"We've got three oscillating fans running in the southeast corner all day," he said. "The techs keep telling me the AC is fine, but that corner hits 80 degrees by 2 p.m. The thermostat display says 72. It's not cooling."
I asked him to photograph the thermostat installation before I approved anything. Over the years, a photo has explained a surprising number of so-called defects.
The photos arrived thirty minutes later, and there it was. The thermostat sat directly above a ceiling supply diffuser—cool discharge air blowing straight onto the sensor before it mixed with room air. The system short-cycled: it satisfied the sensor quickly, shut off, and never cooled the room down. The display said 72. The occupants felt 80. (You could see the diffuser in the frame, maybe eighteen inches to the left of the thermostat.)
ASHRAE Standard 55, the industry standard for thermal comfort, is clear about sensor placement: away from supply air paths, away from heat sources, typically 48 to 60 inches above the floor on an interior wall. The installing contractor, unfortunately, didn't read that chapter.
"Your thermostat isn't broken," I told Dan. "It's lying to itself. So is your fan strategy."
Ceiling Fans, Oscillating Fans, and a Question I Refused to Answer
I drove to the site the next morning. I brought a replacement thermostat in case, but I was maybe 80% sure I wouldn't need it.
Dan met me in the southeast corner, surrounded by humming oscillating fans aimed at desks and doorways. "Should we switch to ceiling fans?" he asked. "Beats replacing these every couple of years."
That one I could answer. Ceiling fans move a larger volume of air at lower speed, which helps whole-space circulation. Oscillating fans are for spot-cooling people, not for fixing a room that's hot because the HVAC system is under-delivering. But neither fixes a thermostat that's reading supply air. Fans are comfort tools. They're not ductwork.
Then Dan asked the question I'd been expecting: "We're replacing the rooftop units next year. Heat pumps or straight air conditioners? You're the Johnson Controls guy—you'd know."
And here's where I had to be honest, even though it felt a little like turning down a sale.
"I know our heat pump line well. I've reviewed lab data on coil failures, compressor warranty returns, and service intervals. But I'm a quality inspector, not a mechanical systems engineer. Whether heat pumps make sense for your building depends on your climate, your utility rate structure, how much winter heating you actually need, and whether you're planning for electrification. That's a load calculation and a financial model. If I answered off the top of my head, I'd be doing you a disservice."
What I could offer was a framework:
- Heat pumps and air conditioners use the same refrigeration cycle for cooling. The difference is the reversing valve—heat pumps can also move heat into the building during winter.
- In mild climates, heat pumps often handle both heating and cooling year-round. In cold climates, performance drops as outdoor temperature drops, so a cold-climate-rated unit or backup heat may be needed.
- The efficiency numbers to compare are SEER2 and HSPF2 under the DOE's updated test procedure, effective January 2023. But ratings don't tell you which option saves more money in your specific building.
- AHRI 210/240 certifications matter for matching real-world system performance. Utility rebates can tip the economics, but they shouldn't be the deciding factor.
What I didn't do was pick a winner. I told him to call our applications engineering team—they run the models and give recommendations based on the building, not a sales target.
How to Adjust a Johnson Controls Thermostat (and When Not To)
His thermostat did go back into the no-fault-found bin. We replaced it with one that supports a remote sensor and mounted the sensor on an interior wall, away from the diffuser. Problem solved without a credit.
If you have a Johnson Controls thermostat that seems to be reading wrong, the adjustment is straightforward:
- Press the Menu button (the gear icon) on the thermostat display.
- Scroll to Setpoints using the arrow keys.
- Adjust the Cool and Heat values with the up and down arrows.
- Press Done. The thermostat will hold the new setpoints.
But here's the advice I'd give any customer: if you change the setpoint and the space still feels wrong, check the sensor before you blame the thermostat. If it's near a supply diffuser, in direct sunlight, above a heat source, or behind a bookshelf, it will read the wrong air. No amount of button-pressing fixes that.
The Lesson: Know Where Your Expertise Stops
A week later, Dan had a ceiling fan installed in the southeast corner, and the oscillating fans went into storage. The applications engineers ran the numbers for his rooftop replacement: heat pumps for the west wing, straight AC with gas heat for the north wing. I don't have the full report in front of me, and I won't pretend to understand every variable in their model. That's the point—they're the engineers. I keep the thermostats honest.
This is a lesson I learned the hard way. In 2021, I approved a batch of controllers for a customer's retrofit without escalating to our engineering team. The controllers tested fine in our lab. The building's electrical environment wasn't compatible. That mistake cost us a $22,000 redo and delayed the customer's launch by six weeks. "I think this will work" is not a specification.
What this job has taught me:
- Most "defective" products aren't defective. I don't have hard data on industry-wide failure rates, but based on three years of reviewing returns, my sense is at least half of the "won't cool" thermostats I see have installation or system-level causes, not hardware faults.
- The most credible sentence I said to Dan all day was "I'm not the person to ask." B2B customers know when they're being sold to. A specialist who acknowledges boundaries earns more trust than a generalist who's confident about everything.
- My experience comes from roughly 600 inspections in our region over three years. If you're in a different climate or facility type, your troubleshooting path might look different. That's okay—the principle still holds: fix the measurement before you blame the instrument.
If your Johnson Controls thermostat feels wrong, check where it's mounted, look at the setpoints, and confirm the system is actually running. If it's still not working, call us. We'll figure out whether you need a controls specialist—or an engineer. Usually, it's the second one.
