Walk-In Cooler Load Calculation: The Four Components, Worked, With the Altitude Derate

A specialty food producer near the mouth of a canyon in Draper called us in August about a walk-in cooler that would not hold 38°F. On the hottest afternoons the box drifted into the mid 40s, the compressor ran without ever cycling off, and the product logs were failing their own temperature audit. The unit had been bought as a package off a national vendor’s cubic-footage calculator, dropped on a slab, and the condensing unit was mounted outside on a west wall that took full afternoon sun and a steady feed of canyon grit. Nothing on the equipment had broken. The box had simply been sized and sited as though it lived in a mild, humid, low-elevation warehouse. It sits at roughly 4,700 feet in a high desert that runs single-digit afternoon humidity and swings 30 to 40 degrees between a July afternoon and the following dawn, and every symptom traced straight back to that.

A walk-in cooler is the most calculated and least understood box in the food business. The free walk-in calculators and vendor sizing charts hand you a BTU number in thirty seconds and never show the arithmetic behind it. They also share a baseline: four inches of urethane at R-25 and, in one sizing guide’s words, “an ambient temperature of 95°F outside your cooler” (American Mortuary Coolers, Beginner’s Guide to Walk-In Cooler Sizing Charts). U.S. Cooler prints the same insulation assumption under the chart it publishes (U.S. Cooler, Size Your Refrigeration System). The R-25 is fine, it is the federal floor. The missing input is elevation. No chart states one, and none of them corrects for one, and nothing in the Mountain West sits at sea level. This page publishes the whole calculation instead: four components, computed separately for a real box, every figure cited to a document you can open, and then the step none of the calculators run.

The four components, and why no calculator publishes them

A refrigeration load is not one number. It is the sum of separately computed terms, each with its own formula and its own behavior across a day (Refrigeration Load Calculations for Cold Storage and Food Processing):

Then a design margin of 10 to 20 percent, from the same source. Then, in Utah, the step nobody outside the Mountain West runs: the altitude derate.

Separating the terms is not academic. Each behaves differently. Transmission is constant. Infiltration scales with how often the door opens. Product pull-down is enormous for half a day and then exactly zero. A single cubic-footage number cannot tell you which of those is your problem, which is why the Draper box could be “correctly sized” and still fail its audit.

And the shortcut formula in circulation is dimensionally wrong. The most widely syndicated field estimate is volume x 0.24 x temperature difference, offered with the worked example “1,000 cubic feet x 0.24 x 42°F = 10,080 BTU” (ServiceTitan, Refrigeration BTU Calculator). But 0.24 Btu per pound per °F is the specific heat of air per pound, and cubic feet are not pounds. Standard air weighs about 0.075 lb per cubic foot, which is exactly why the industry’s own sensible-heat constant is 1.08 = 60 x 0.075 x 0.24. Put the density back in and the same 1,000 cubic feet cooled 42°F holds 756 BTU, not 10,080. It is also a quantity of heat rather than a rate, so it was never a sizing number in the first place.

Worked example: a 10 by 12 walk-in cooler at 5,000 feet

Ten feet by twelve, eight foot ceiling. That is 960 cubic feet and 592 square feet of outside surface: 352 of wall, 120 of ceiling, 120 of floor. Holding temperature 38°F, which sits under the 41°F cold-holding ceiling the FDA Food Code sets for refrigerated produce. FDA’s retail guidance states it directly: “Cut leafy greens must be maintained at temperatures of 41ºF (5ºC) or less during cold storage and display (3-501.16)” (FDA, Recommendations for the Temperature Control of Cut Leafy Greens during Storage and Display in Retail Food Establishments). Design ambient 95°F, the value the vendor charts assume and also the 1 percent cooling design dry bulb that our Manual J permit page reproduces for Salt Lake City International from ACCA Manual J Table 1A. So dT is 57°F across walls and ceiling.

The box in the story sits at about 4,700 feet. The example runs at 5,000 feet, because that is an elevation the manufacturer correction table in step 4 tabulates directly. Nothing on this page is interpolated.

1. Transmission

Panels are R-25, and that is not a preference. It is the federal minimum: 10 CFR 431.306 requires “wall, ceiling, and door insulation of at least R-25 for coolers and R-32 for freezers”, plus “floor insulation of at least R-28 for freezers” and no minimum at all for a cooler floor (10 CFR 431.306). U is 1/R, so 0.040.

The floor gets a different dT, and this is the Utah credit. Mean summer soil temperature for the Salt Lake soil series runs about 60 to 62°F (USDA NRCS, Salt Lake Series Official Description), so a slab-on-grade cooler floor sees roughly 22°F of difference, not 57°F.

Notice what the federal rule does and does not require: R-25 on walls, ceiling and door, and nothing on a cooler floor. In a hot-ground market that omission is a liability. Here it is a design lever, because the ground you are coupling to is 35°F colder than your design ambient.

2. Infiltration

Air changes are the whole game, and the published range is enormous: 1 to 2 changes per day for a well-sealed, low-traffic cooler against 8 to 15 for a high-traffic one (Refrigeration Load Calculations). A produce-storage calculator published by the University of Vermont uses a different unit, changes per hour: “For long-term storage which is well-sealed with few door openings use 1. For a less well-sealed cooler or a walk-in with lots of traffic in and out, use something closer to 3 ACH” (UVM, Heating and Cooling Load Calculator for Produce Coolers and Warm Rooms). Check which unit a table is in before you borrow a number out of it, because 3 changes an hour is 72 a day, not 3. Everything below uses the per-day convention.

At 12 changes per day, 960 cubic feet is 11,520 cubic feet a day, or 8.0 CFM. Sensible: 1.08 x 8.0 x 57 = 492 Btu/h.

Same box, same panels, same setpoint: the door schedule alone moves this term by 534 Btu/h, which is 45 percent of the entire transmission load. It is the largest thing an operator still controls after the slab is poured, and a calculator that takes only dimensions cannot see it at all.

The latent term is where Utah bites, and its sign is the story. dW is the humidity ratio difference, and on the Wasatch Front in summer the infiltrating air is drier than the air inside the box. Every one of the twelve Utah stations in ACCA Manual J Table 1A carries negative design grains, which our Manual J permit page reproduces in full. So this term does not behave like a cooling load the compressor removes. It is moisture leaving your product: saleable weight off produce, and in a freezer, water vapor migrating inward, frosting the coil, and driving a heavier defrost cycle whose heat the system then has to remove again.

One honest caveat about 1.08. It is a sea-level constant by construction, 60 minutes per hour x 0.075 lb per cubic foot x 0.24 Btu per pound per °F. Denver’s station pressure runs about 12.1 psia against 14.7 at sea level, a 17.7 percent reduction (Denver HVAC Authority, High-Altitude HVAC Considerations). A calculation run at Wasatch Front elevations with an uncorrected 1.08 therefore overstates its infiltration line, and it does so in the opposite direction to the equipment derate in step 4. Both are real. They do not cancel, because they act on different lines of the calculation.

3. Product pull-down

This is the term the calculators skip most often and the one that fails the most boxes. For most produce a specific heat of 0.85 Btu per pound per °F is a reasonable assumption (UVM). A 500 pound delivery arriving at 60°F and pulled to 38°F is 500 x 0.85 x 22 = 9,350 BTU of stored heat. What that costs you depends entirely on how fast you want it gone:

Produce also keeps making heat while it sits there. Respiration is reported in milligrams of CO2 per kilogram per hour and varies by more than a factor of two between common commodities: apples in fall storage run about 5 mg/kg-hr between 32 and 41°F, cured potatoes about 12 mg/kg-hr at 41°F (UVM). Pull-down ends. Respiration does not.

4. The altitude derate

Add the steady-state terms: 1,183 transmission, 492 infiltration sensible, and 475 of internal gain (two 1/15 hp evaporator fan motors at 2,545 Btu/h per hp is 339, and 40 watts of LED at 3.4 Btu/h per watt is 136, both constants from the same source). That is 2,150 Btu/h. Apply a 15 percent design margin and the box needs 2,473 Btu/h removed, averaged over 24 hours.

Two conversions stand between that and a nameplate.

First, run time. A condensing unit is not sized to run continuously; it needs off time for defrost and reserve. The published default is 18 hours a day for a cooler and 16 for a freezer (Toolgrit, Walk-In Cooler/Freezer Heat Load Calculator). A 24-hour load of 2,473 Btu/h delivered in 18 hours of run time is 2,473 x 24/18 = 3,297 Btu/h of equipment capacity.

Second, altitude. And here the version everyone repeats is wrong. CellarPro publishes an altitude correction table for its refrigeration equipment, and it separates two things the trade routinely conflates (CellarPro, Altitude Correction Table):

At 5,000 feet airflow falls 11 percent and rated capacity falls 4 percent. Those are not the same number, and the reason is that reduced air density hurts the fan far more than it hurts the refrigeration cycle. CellarPro’s own explanation is that performance degrades “because of reduced mass flow rate of the lower density air”, which is an airflow statement, not a capacity statement. So the sea-level nameplate this box needs is 3,297 / 0.96 = 3,434 Btu/h.

One correction worth making loudly. Contractors routinely apply a 4 percent per 1,000 foot derate to refrigeration equipment. That figure belongs to combustion appliances, where it is applied above 2,000 feet of elevation (Denver HVAC Authority), and it is the number a Utah plan reviewer checks on a furnace, which our Manual J permit page works through line by line. Borrowed for a walk-in at 5,000 feet it produces a 12 percent derate where the manufacturer publishes 4 percent, three times the correction, and a 3,747 Btu/h nameplate against the 3,434 the table actually calls for. It is a category error that lands in a plausible-looking place, which is exactly what makes it durable.

What the numbers say

The spread is the finding. Same box, same panels, same setpoint, same elevation: 3,434 Btu/h of nameplate if it is a quiet walk-in with a disciplined door, 7,367 if it is a busy one taking a pallet a day. That is a factor of 2.1, and all of it comes from two inputs, door count and product turnover, that a cubic-footage calculator never asks about.

Which is why the vendor charts have two columns. U.S. Cooler’s guidance is to use the “Heavy Load” calculation “for walk-ins with glass doors or ones that have frequent traffic” (U.S. Cooler). The vendors know. The two-column chart is how they admit that volume does not determine the answer without having to publish the method.

Do not read the low end as permission to buy small. The 3,434 figure assumes the door discipline you say you have. If it is a restaurant line cooler, the honest recommendation is the number that survives your worst week rather than your best one. Size it deliberately, and write down which case you sized for. Choosing by accident is the failure mode.

Dry air makes infiltration and defrost the hidden loads

Every cold-room guide mentions infiltration. Almost none of them account for what infiltration means when the outside air holds almost no moisture. Summer afternoon relative humidity on the Wasatch Front routinely sits in the single digits to low teens. Two things follow that the national calculators miss.

First, the moisture that does exist inside the box, and the moisture the product gives off, is constantly pulled toward that dry outside air through every door cycle and every envelope gap. In a cooler that dries produce out and drops its saleable weight. In a freezer it means water vapor migrates in, freezes on the coil, and drives a heavier and more frequent defrost cycle, and every defrost dumps heat back into the box that the system then has to remove again. The defrost strategy, hot-gas versus electric, and its timing are a design decision here, not a factory default, because the frost load is larger and more seasonal than a humid-climate design assumes.

Second, the door and vestibule detailing carries more weight in Utah than the panel R-value does. Air curtains, strip curtains, tight sweeps, and a real vestibule on a high-traffic freezer are what keep the infiltration line on the load calculation from ballooning. A box with excellent panels and a sloppy door schedule loses the moisture and temperature battle at the threshold, every open, all day. The envelope also has to be a genuinely continuous vapor barrier: detailed like a humid-climate box, the assembly frosts inside the panel and loses R-value over years.

The diurnal swing and snow decide where the condenser lives

Utah’s high desert does not hold a steady outdoor temperature. A 30 to 40 degree swing between a hot afternoon and a cool pre-dawn is ordinary in summer, which means the condensing unit that is barely keeping up at 4 p.m. is oversized for the ambient at 4 a.m. A single-stage system built to the afternoon peak short-cycles at night, which wears compressors and controls. The design answers are variable or multi-stage capacity that can throttle down after dark, and floating head-pressure controls that let the system take free efficiency from the cold night air instead of holding an artificially high condensing temperature. Neither shows up on a cubic-footage tool.

Where the condenser physically lives is the same Utah decision that governs any heat pump or AC condenser in this state, and it is exactly where the Draper cooler went wrong. On a Park City or Snyderville site it faces snow load and drift and needs elevated, snow-cleared placement with clear airflow, because a coil buried in snowmelt ice or blocked drift loses capacity on top of the altitude derate. On an east-bench Draper or Sandy site it can sit at a canyon mouth and be fed grit and gusts that foul a coil fast. In St. George the priority flips to fine silt and extreme heat piling onto an already derated condenser. Condenser placement is part of the refrigeration design, not a spot the installer picks for convenience, and a fouled or heat-soaked condenser is the most common reason a correctly sized box still cannot hold its number. This is the same altitude-and-terrain discipline covered in our Utah climate HVAC design guide, and once sited correctly the unit has to stay on a Utah-aware service calendar, the coil-cleaning and derate-check routine in our guide to annual HVAC maintenance in Utah.

Earth-berming a walk-in: the transmission load Utah gives back

Transmission is the one load line a Utah site can genuinely reduce before any equipment is chosen, and on a commercial box it is worth real compressor capacity. The 106 Btu/h floor line in the worked example above is the whole argument in one number: the same 120 square feet facing 95°F air instead of 60°F soil would contribute 274 Btu/h, and a bermed wall behaves the same way.

For a commercial operator the consequence is capacity headroom, which is really inventory risk. A box whose transmission load is partly carried by the ground has margin left on the hottest afternoon of the year, which is exactly the afternoon a full cooler and a busy dock door will test. A slab-on-grade box on a west elevation spends that margin on the wall instead, then borrows it back from product temperature when the door opens. The siting decision, berm or no berm, north elevation or west, belongs at the schematic stage alongside dock layout and drainage, because after the slab is poured it is the one input nobody can revise.

The same four-component calculation at a much tighter residential setpoint, where humidity rather than product safety is the binding constraint, is worked line by line in our guide to sizing a wine cellar cooling unit, including the altitude derate.

How a Utah walk-in and cold-storage system should be specified

Pulled together, a walk-in cooler or freezer, cold-storage room, or process refrigeration system in Utah is a genuine engineering problem with four inputs the box calculators ignore:

• A real, component-by-component load calculation: transmission, infiltration sensible and latent, product pull-down and respiration, internal gains, plus a design margin, then converted to equipment capacity at a stated run time rather than read straight off a chart.
• Equipment sized on its derated capacity at the actual elevation, taken from the manufacturer’s own published correction table rather than from a combustion derate borrowed out of context, with the condenser placed for snow, canyon grit, or desert silt and heat.
• Infiltration and defrost treated as first-order loads, with door, vestibule, and vapor-barrier detailing sized to single-digit outdoor humidity instead of copied from a humid-climate spec.
• The cold Wasatch soil used as a heat sink through deliberate below-grade or bermed placement, so a modest derated unit can hold the band.

Done this way, the box holds 38°F through a St. George August and a Park City January alike, and the food-safety log passes. Done the calculator way, you get the Draper cooler: a unit that never rests and inventory at risk. The contractor model matters here too, and the case for one accountable team that owns the numbers from load calculation through maintenance is laid out in our piece on the design-build HVAC model in Utah.

A short note on ammonia, and the edge of our scope

At the industrial end of cold storage, ammonia (R-717) is a common and efficient refrigerant, and it changes what design means. Our scope there is narrow and worth stating plainly rather than burying: we are not an ammonia refrigeration installer, we do not fabricate, field-erect or service ammonia systems, and nothing here is an offer to. That work belongs to firms that specialize in it. Where we help is load and envelope arithmetic alongside your licensed ammonia contractor and your process-safety team, and one piece of standards awareness worth having early: ANSI/ASHRAE Addendum a to Standard 15-2016, approved June 28, 2018, adds Section 2.3, “This standard shall not apply to refrigeration systems using ammonia (R-717) as the refrigerant,” plus an informative note directing users to ANSI/IIAR 2 (ASHRAE, Addendum a to Standard 15-2016). A design conversation still leaning on ASHRAE 15 for an ammonia system is citing a standard that excludes those systems.

Get the number before you get the quote

If you are sizing a walk-in cooler or freezer, a cold-storage room, or a food-processing or beverage refrigeration space, the useful first deliverable is not a quote. It is the component-by-component load calculation above, run on your box, your door schedule, your product turnover and your elevation, with every input written down so a vendor’s chart can be checked against it. That is what our load calculation and system design services produce. If that is your project, send us the dimensions and the door count, and we will start with the numbers the system should be built on.