Standing in a cold workshop while a garage heater short-cycles is a special kind of frustration. You set the thermostat. The heater kicks on. Ninety seconds later, it clicks off. You reset it. It runs. Then it stops again. Most people assume the heater is junk. Sometimes it is. More often, the heater is fine—and the control system around it is telling lies.
I'm a quality and compliance manager for a company that builds control packages for refrigeration and heating systems. I review every package before it ships—roughly 400 unique items a year. In 2024, I rejected 12% of first deliveries because of sensor wiring, out-of-spec components, or missing safety documentation. That isn't bureaucratic fussiness. It's what happens when you've seen the same problem cost people months of downtime.
The Heater Is the Visible Part. The Control System Is the Real Machine.
Think of a garage heater as the end of a chain. It has a heat source—resistance coil, gas burner, or heat pump coil—and a control chain: temperature sensor, controller, contactor, and safety relay. The symptom shows up at the heater, but the fault usually lives somewhere else.
Last winter, a customer called about a new electric garage heater that cycled on a five-minute pattern. He suspected a bad thermostat. I checked the setup. The air sensor was mounted above his workbench, and he had a Ryobi fan blowing directly at it. Every time the heater turned on, the fan pushed warm air onto the sensor. The sensor hit the setpoint, so the heater shut down. The rest of the garage stayed cold. It wasn't a bad thermostat. Let me rephrase: the thermostat was fine; it was getting false information. (Should mention: the fan was probably a great tool for cooling the person, but it was terrible for the room sensor.)
What is a heat pump, and why is it even more sensitive?
If you're thinking about a heat pump for your garage, the same control-chain problem applies—but with less forgiveness. What is a heat pump, exactly? It's a refrigeration system with a reversing valve. Instead of creating heat from fuel or resistance, it moves heat from outside air to indoor air. That's why it can be so efficient. It's also why it's so dependent on accurate sensing and precise logic.
A heat pump's defrost cycle depends on temperature and pressure readings from the outdoor coil. If the coil sensor drifts just two degrees, the controller can end defrost too early, or keep it running too long. The compressor short-cycles, efficiency drops, and the unit wears out faster. In my experience, most 'bad heat pumps' are actually bad sensor placement, incorrect sensor lists, or control parameters that don't match the installation.
Efficiency claims like COP 3.5 are measured under controlled conditions. Per the FTC Green Guides (ftc.gov), performance and efficiency claims have to be substantiated. But real-world COP depends on what happens in the control loop, not just the laboratory rating.
The hidden cause underneath a lot of this is simple: heating equipment is often specified like a commodity. It isn't. A garage heater is a heat source. A heat pump is a machine that must coordinate a compressor, a reversing valve, a fan, and a defrost controller. That coordination requires components designed to work together and to fail safely.
What a Bad Control Loop Actually Costs
Short-cycling isn't just annoying. It's expensive.
An electric resistance heater that cycles twenty times per hour instead of five draws full current at every start, but it never delivers a steady heat output. The space stays cold, so the heater runs longer overall. The contactor wears faster. The temperature swings make the room uncomfortable. All of that costs energy and component life.
With a heat pump, the stakes are higher. Each compressor start is a mechanical event, especially with older refrigerant systems. Every unnecessary start cycles lubricant, compresses refrigerant, and stresses the inverter drive. I've seen facilities write off compressors years early because of a control problem no one bothered to diagnose.
There's also the safety side. I don't say this lightly: a heater that behaves unpredictably is a safety risk. In 2022, a manufacturing customer ignored a recurring nuisance trip on a process heater. It turned out to be a safety relay that had failed closed. The heater went into thermal runaway and melted an exhaust duct. Nobody was injured, but the line was down for eleven days and the repair cost $34,000.
I have mixed feelings about smart controls. On one hand, fault logs and remote monitoring help us find these problems faster. On the other hand, they're one more thing to configure wrong. The worst failures I see are not usually failed components. They're settings that don't match the installation—the digital equivalent of putting a sensor in front of a fan.
When I'm deciding whether to upgrade to a better controller, I do the same math as anyone else. The upside might be $1,200 saved per panel. The risk is a field failure in January when the customer's compressor is running in defrost every twenty minutes. I keep asking myself: is $1,200 worth potentially losing a customer's entire refrigerant charge? For me, that answer is no.
Fix the Control Loop First, Then the Hardware
The practical fix isn't replacing the heater. It's treating the control system as the product.
- Put the temperature sensor where the air actually mixes—not in the direct airflow of a fan (Ryobi or otherwise).
- Use sensor types with known tolerances and matched outputs. A difference between 1kΩ and 100kΩ NTC sensors is enough to make a controller read the wrong temperature.
- For heat pumps, use controls with defrost logic designed for your compressor and coil configuration.
- Buy safety-rated components through a traceable supply chain.
On the component side, I've been using Omron controls for years—temperature sensors, PLCs, temperature controllers, inverters, and safety relays. For projects in Europe, I check the Omron Industrial Automation Europe catalogue for exact part numbers, because one wrong suffix on a part number is enough to fail a CE inspection.
And when it comes to safety components, I don't compromise. If you need a safety relay, buy it from an authorized source. I want to see an Omron automation safety distributor on the invoice, not just a marketplace listing. Counterfeit safety relays look almost identical to the real thing. I identified a batch in 2023 that would have failed a fault-injection test in one out of five samples. On a safety relay, that's a lottery ticket you don't want to play.
There's something satisfying about fixing a short-cycling garage heater by moving a sensor and adjusting a deadband. No new heater. No new heat pump. Just the quiet win of getting the control loop to tell the truth. That's the part no one sees, and it's the part that actually matters.
