A cold store runs at minus eighteen and the product in it sits at minus eighteen, because a box of food cannot be colder than the air it has been sitting in for a fortnight. In bay 14 the probe reads minus twenty three, and the electricity to do that is not on the bill.
I am a refrigeration engineer and I look after about forty industrial cold stores.
What the measurements are, because this gets confusing fast if you do not separate them.
Air temperature is what the store's controller reads, off sensors in the return air. Minus eighteen is the legal figure for frozen food and the store is alarmed if it drifts.
Product temperature is different and it is what an environmental health officer actually checks. You measure it with a probe between two boxes, or properly by driving a needle probe into a product, and a product at minus fifteen in air at minus eighteen is a pallet that came in warm and has not pulled down yet.
Product cannot be colder than the air. That is not a regulation, it is the second law: heat moves from the warmer thing to the colder thing, so the coldest thing in a room ends up at the room's temperature and no further. The air is what is being refrigerated. The boxes are along for the ride.
And the whole plant is metered. Electricity at the switchboard, suction and discharge pressures logged, run hours on each compressor, and the heat rejected at the condenser is calculable from all of it.
Bay 14 in a store in the Midlands holds product at minus twenty three.
The air in that aisle is minus eighteen point two, which is what the controller says and what my own calibrated probe says at four heights.
Between the boxes in bay 14, it is minus twenty three.
I have measured that with three instruments over six years: a needle probe into a block of product, an infrared gun on the outside of a box, and a logger left in the bay for a week between two pallets.
The logger is the one that finished the argument with myself. A week of ten minute readings: minus twenty two point six to minus twenty three point four, flat as a road.
What the obvious answers are.
A cold spot from the evaporator throw. A store is not the same temperature everywhere, and the air coming off a fan can be four or five degrees below the setpoint, which is why you measure return air and not supply air.
I have smoke tested the air movement in that store. The throw from the nearest evaporator goes down the main aisle and bay 14 is off it, in the lee of a rack, in the deadest air in the building.
Dead air is warmer. Everywhere else in the world, the bay in the lee of the rack is the bay that fails an inspection for being at minus fifteen.
The product. It arrives on a lorry at minus twenty or so, because a good transport chain runs colder than the store to allow for loading. So a fresh pallet is briefly colder than the room.
Briefly. A tonne of frozen product in air five degrees warmer than itself comes up to air temperature in about two days, and you can watch it happen on a logger.
Bay 14's product arrives at minus twenty and goes DOWN to minus twenty three over about thirty hours, and stays.
The electricity, which is why I cannot let this go.
Taking a tonne of product from minus eighteen to minus twenty three is heat you have to remove and then keep removing, because the room is constantly leaking heat back into it.
I worked it out twice and had a colleague check the second one. To hold that bay where it sits, against that temperature difference, with that much product, takes somewhere between nine hundred watts and one and a half kilowatts, continuously.
The store's electricity is metered at the board and submetered per plant room. The consumption matches the calculated load for a store at minus eighteen, to within about three percent, which is as good as this sort of arithmetic gets.
There is no kilowatt and a half. The plant is not doing it.
So something in that bay is removing about a kilowatt of heat and not using any energy I can find to do it, which is a sentence I have typed into a service report and then deleted.
What is there.
Concrete floor on insulation on a heated slab, which is standard to stop frost heave under a store, and the slab heating in that area works and is monitored, and if it had failed the floor would be lifting, and it is not.
Racking is galvanised steel, same as the rest. No pipework in that bay, no duct, no wall. Bay 14 is in the middle of the floor.
What happens about it.
Nothing needs to happen, and that is the reason this has gone on for six years.
Product at minus twenty three passes every inspection there has ever been. The customer's quality team are delighted with bay 14 and have asked me twice why the rest of the store cannot be like it.
I have written it up once, in 2020, as an observation on an annual service report, in plain language, with the logger data attached.
The site manager's reply was a one line email thanking me and asking whether we could do the same in bay 15.
What I do now.
The bay is on my list as monitored, and I drop a logger in it once a year because I want the record to exist.
The one thing I have changed is how I do that.
You need to be in the store for about four minutes to place a logger properly, and minus eighteen in a freezer suit is fine for twenty.
Bay 14 was the place in that store where I would stand to write my notes, because it is out of the draught and out of the way of the forklifts.
I write them in the lobby now. The reason is small and I have told nobody at that site: in 2021 I stood in bay 14 for about six minutes filling in a sheet, and when I came out my pen had stopped working, and the ink in it was frozen solid, and ink does not freeze at minus eighteen, and I was holding it.