I manage HVAC service orders for a mid-sized commercial property. Nine years in, I still keep a mistake log, partly because I'm not as smart as I like to think, and partly because the lessons keep wanting to repeat themselves. This one is about an AAON supply air temperature sensor, a leaky condenser coil, a Midea dehumidifier, a Hisense dehumidifier, and a hot water heater I should have flushed years earlier. They're different machines, but the pattern is the same: I fixed the part that reported the problem instead of the part that actually caused it.
The Surface Problem: A Sensor That Wasn't Lying
The building automation system flagged the AAON supply air temperature sensor for reading 14 degrees below setpoint. The air leaving the rooftop unit felt lukewarm. I did the classic thing: ordered a replacement AAON supply air temperature sensor, swapped it, and watched the alarm clear for about eight days. Felt good. Then the high-pressure alarm hit, and a week later we found oil spots on the condenser coil.
At the time, I told myself those were two separate failures. Sensor first, coil second. Looking back, they were one failure with a delayed reveal.
The Deep Cause: The Sensor Was the Messenger
An AAON supply air temperature sensor is a thermistor. It measures the temperature of air leaving the coil. If the refrigerant side is sick, the leaving air temperature is sick too, so the sensor reading actually makes perfect sense. The problem was that our condenser coil, the aluminum coil on the outside of the unit, had developed a slow leak. The leak was small enough that the unit still had some cooling capacity, but the compressor was running hotter and the evaporator coil was starting to freeze in one corner. The sensor was reading a mix of cold air and, when the airflow split, slightly warm bypass air. It looked like a sensor problem. It was a refrigerant flow problem.
Here is the part I didn't get until later: the sensor was the only component telling the truth. I replaced it because I didn't want to believe that a coil on a six-year-old unit could fail. That was my problem, not the sensor's.
Condenser coils are the unsung workhorses of any rooftop unit. They sit outside and take the rain, the pollen, the roof grit, and in our case salt spray from a nearby road. The aluminum fins are delicate. If you power-wash them too close, you bend the fins and choke off airflow. If you never clean them, they develop hot spots that stress the metal. Our AAON unit had a pinhole leak in a return bend, and we only found it after pulling the panels and using an electronic leak detector. It was not a bad sensor. It was a tired coil that had been slowly dying for months.
According to our AAON service manual, the supply air temperature sensor resistance should be checked against the temperature-resistance table before replacement. I skipped that step. If I had done it, I would have seen that the sensor was within spec, and I might have gone looking for the real issue instead of ordering a part.
The Cost of Fixing Symptoms
That one incorrect diagnosis cost way more than the part. The sensor was around $220 as of January 2025, maybe $230 with tax, I'd have to check the invoice. The service call to swap it was $340. The next service call for the high-pressure alarm was $450. The temporary workaround, one of those just keep it running fixes, cost about $600 in labor. When I finally ordered the right parts, the quote for AAON replacement condenser coils was $2,800. Actually, $2,950 with freight. The entire episode added up to almost $4,800, and roughly half of that was because I fixed the symptom first.
The numbers said buy an aftermarket coil. The price was lower and it was in stock. My gut said stick with AAON replacement condenser coils because the original coil geometry was engineered for our unit's cabinet and fan curve. I went with my gut. Later, I found out the aftermarket option had different fin density and would have required adjusting the fan pressure settings. Not an impossible retrofit, but not the forty-five-minute swap the sales quote made it sound like.
What would have happened if I had ignored the alarm entirely? The leak would have gotten worse, the compressor would have cycled on thermal overload, and eventually the compressor would have failed. Replacing a compressor costs several times more than replacing a coil. Actually, I should say could cost several times more. It depends on the model. But every service manager I know has a compressor failure story, and none of them are cheap.
Same Pattern, Different Equipment
I made the same mistake in the same month with a Midea dehumidifier in our basement network room and a Hisense dehumidifier in the storage area. The Midea unit would run for hours and barely pull water from the air. I was convinced the humidistat was bad. When I opened the front cover, the filter looked like a gray felt blanket, and the coil was a block of frost. The machine wasn't broken. It was suffocating. I cleaned the filter, let the coil thaw, and it worked within an hour.
The Hisense unit had a similar story. It couldn't keep up, so I assumed it was undersized. It was, but not for the reason I thought. The coil was frozen because the air inlet was blocked by a box of packing foam. A larger unit would have wasted money. Removing the box fixed it.
I'm not here to bash either brand. Midea and Hisense both make decent budget dehumidifiers. But their performance depends on airflow, and most of the airflow problems are from filters and blocked inlets, not from dead sensors. When comparing dehumidifiers, I look at AHRI ratings, not just the pint number on the box. The rated pints are typically measured at 80 degrees and 60 percent humidity. A basement at 65 degrees will pull a lot less, and an undersized unit will freeze up even faster if the filter is dirty.
The water heater was the same lesson in a different tank. A tenant reported that hot water ran out in eight minutes. I thought the heating element was dying. In fact, the element was fine. The bottom half of the tank was full of sediment, so the element was trying to heat a layer of mineral crust instead of water. That is when I finally learned how to flush a hot water heater.
How to flush a hot water heater is not complicated: turn off power or gas, close the cold water supply, attach a garden hose to the drain valve, open a hot water fixture or the pressure relief valve, and drain it until the water runs clear. The catch is that you have to do this every year or two, not after the tank starts making popping sounds and spitting out rusty water. Sediment also causes the burner or element to cycle more often, which means higher energy bills and eventually a leak from the tank bottom. If I had known this five years ago, I would have saved one plumber's visit and a new heating element that was never actually broken.
What I'd Do Now
When a sensor or a dehumidifier gives a weird reading, I force myself to look at the whole chain before opening the wallet. For the AAON units, that means checking the supply air temperature sensor against the resistance chart, then checking condenser coil cleanliness and pressure readings. Replacing an AAON supply air temperature sensor is a legitimate repair when it fails, but it is usually not the first thing to blame.
For condenser coils, I order AAON replacement condenser coils from a distributor that looks up the exact revision number. I don't guess based on physical dimensions, because the coil geometry and fin spacing matter more than I used to think.
For dehumidifiers, I compare AHRI ratings and actual conditions, not the pint numbers on the box. A Midea dehumidifier can be a perfectly good unit if it is the right size and the filter is clean. A Hisense dehumidifier can be equally reliable, but it will not save you from a blocked inlet.
For the hot water heater, I put a reminder in our maintenance calendar. The total time to flush is about 45 minutes. The cost of not flushing is a tank that works worse, uses more energy, and eventually needs replacing.
I still kick myself for ignoring the sensor's message. A sensor is not a fortune teller. It's a measuring device. When it reads wrong, you ask why. I don't know why some AAON coils last twelve years and ours lasted six. My best guess is that the local stormwater runoff and salt spray around the roof did things to the aluminum that we didn't plan for. If someone has a better theory, I'd love to hear it.
