The Night a Tower Fan Tried to Replace a GEA Polacel Cooling Tower

The call that starts every bad story

At 11:47 p.m. on July 11, 2024, my phone rang while I was three bites into a gas-station sandwich. I remember the time because nobody calls a refrigeration engineer at 11:47 at night to catch up. At that hour, a call means something has already gone wrong.

I'm a service engineer with GEA Industrial. For the past eight years, I've worked on refrigeration systems for food warehouses and processing plants and taken more than 200 emergency calls, including overnight turnarounds for frozen-food clients. This one still shows up in my training presentations.

The caller was Mike, the facility director at a cold-storage warehouse in central Iowa - roughly 480,000 square feet and 42,000 pallets of meat, vegetables, and prepared meals for two national grocery chains. We had installed two GEA Polacel cooling towers on the engine room roof in 2021, and I had been part of the commissioning team. So when Mike said Cell 2 tripped and wouldn't reset, I started calculating how fast a freezer building could lose ground in an Iowa heat wave.

An air movement situation

While Mike talked, I heard a whirring sound in the background. Not the heavy hum of industrial equipment. The thin, oscillating whir of something with a plastic grill. I asked what it was. He paused, then said, 'We have an air movement situation.' Those words are still funny to me, in a painful way.

When I got to the engine room, I found four residential tower fans lined up in front of the ammonia condensers, running at full speed. On a rolling cart sat one of those countertop arctic air cooler units - the kind marketed for nightstands and dorm rooms - quietly evaporating water from its small reservoir and blowing a damp puff of air at a condenser shell.

Mike was proud of the setup. He said it was the same principle as the cooling tower: evaporative cooling.

Wrong. Not sort of wrong. Wrong by orders of magnitude.

Why fans don't scale up to cooling towers

I'm not going to make fun of Mike. He's a sharp facility director, and his logic wasn't insane. A tower fan moves air. An arctic air cooler uses evaporation. A cooling tower also uses evaporation. So why shouldn't they help?

Because of the difference between cooling a person and cooling a process.

A person gives off only a few hundred Btu of heat per hour. That's why a 40-watt fan can make you comfortable: it evaporates sweat from your skin. Steel doesn't sweat. A condenser shell doesn't feel a breeze. The heat in that engine room wasn't being carried away by air moving through a doorway. It was being carried in condenser water up to the roof, where the GEA Polacel cooling tower was supposed to evaporate hundreds of gallons of water per hour and reject millions of Btu. You cannot replace a 30-hp fan moving 150,000 cubic feet of air per minute with four 40-watt tower fans. It's like trying to empty a lake with a drinking straw.

What Mike's crew had done was the industrial equivalent of putting a desk fan next to a campfire and announcing that the forest fire was handled.

The immediate problem wasn't the air temperature in the engine room. It was condenser water temperature. If the tower couldn't shed heat, the water would come back to the ammonia condensers too warm. Discharge pressure would climb until the compressors tripped on high pressure. Once the compressors were off, the freezers would start warming, and 42,000 pallets of frozen food would be on a clock.

The motor that had been talking for weeks

At 2:45 a.m., we climbed to the roof and opened the fan section of the failed Polacel cell. The smell hit first: hot grease and burnt insulation. The drive-side bearing had collapsed, the shaft was scored, and the motor was seized solid. The VFD had done its job and tripped on overcurrent, which is why the cell wouldn't reset. This wasn't an electrical mystery. It was a mechanical failure that been building for a while. A maintenance log from two weeks earlier had noted 'unusual growl from Cell 2 fan deck.' In fairness to Mike, no one expects a cooling tower to fail in July, in the middle of a heat wave, in the middle of the night.

We needed a replacement motor. Not just any motor - the correct motor for that tower's drive arrangement, with the right mounting base, shaft dimensions, and sheave. A generic substitute from a local supplier might run, but it could vibrate, overheat, or destroy the new belt in a month. The right part was sitting in a GEA distribution center 190 miles away.

Standard freight would have meant Monday. The plant didn't have until Monday.

The freezer burn question

Around 6 a.m., Mike asked the question I hear from cold-storage operators more than I expected: 'Is freezer burn safe to eat?' His quality team had already flagged pallets near the door, worried about frost and dry spots on exposed cartons. If they had to throw those out, the loss would be enormous.

I'm not a food scientist, so I'm not going to lecture anyone on food safety regulations. But on this one, the USDA is clear. According to the USDA FSIS, freezer burn is a quality issue, not a food safety issue. It happens when air reaches the surface of frozen food and dries it out. The texture suffers, the flavor may change, but the food is safe. At worst, you trim the dry spots and cook it normally.

Freezer burn wasn't the thing Mike should have been scared of. The real food-safety risk was different: if the freezers climbed high enough for long enough, and the product thawed, then you have a temperature-abuse problem. That's the scenario where you start discarding pallets. Dry spots are cosmetic. A warehouse full of thawed meat is not.

The part, the courier, and the clock

By noon, the condenser water temperature was getting uncomfortably close to the high-pressure trip setting on the compressors. The remaining three tower cells were doing everything they could, but in 96-degree heat, they couldn't fully handle the load alone.

At 8:15 a.m., our parts coordinator confirmed that the exact replacement motor was in stock at the GEA distribution center 190 miles away. The courier fee was, I want to say, around $800 - don't quote me on that. Mike did the math faster than I could. The cost of one pallet of frozen beef was higher than the freight bill. He signed for the rush delivery.

The motor arrived at 1:10 p.m. in the back of a courier's van. My crew had it bolted down, aligned, and belted by 4:40 p.m. When we started Cell 2 and it ramped to full speed, the condenser water temperature started dropping within minutes. The warehouse never warmed past -4 degrees. Mike's quality team pulled a couple dozen suspect cartons, trimmed a few frost-damaged spots, and didn't discard a single pallet.

Not one.

What that night changed for me

GEA Polacel cooling towers aren't complicated machines. They move air, evaporate water, and reject heat. But the word 'cooling' covers too much territory. A fan cools your skin. A cooling tower cools an industrial process. The difference isn't size - it's purpose, and people who operate these systems deserve a clear explanation of that difference before something fails at midnight.

That night changed how I talk to customers. Now I'd rather spend ten minutes explaining heat rejection than show up at 2 a.m. to find four tower fans and an arctic air cooler aimed at a condenser. An informed customer asks better questions and makes faster decisions. Mike made a fast decision when the real stakes were clear.

The tower fans are still in the engine room, last I heard. I hope they ended up in front of a desk in the break room, where they belong. They're fine fans. None of them was ever a cooling tower.