Five loads, and they are not the same size
A refrigeration load is built from separate pieces: heat conducted through the panels, heat carried in on air when the door opens, heat taken out of product that arrives warm, heat generated inside by lights, fans and people, and an allowance for defrost. Adding them up is easy. The value is in seeing the proportions, because they are not what most people assume.
Run the defaults — a ten by eight by eight box at R25, 45 degrees of difference, sixty door openings a day at twenty seconds. Conduction through all six panel surfaces gives about 806 BTU/hr. The door gives about 1,418 sensible plus 851 latent, so roughly 2,268 in total, which is nearly three times the panels. Four hundred pounds of product coming down from 70 to 35 over a day gives about 525. The evaporator fans on their own give about 1,365, which is also more than the walls.
Two things in that list surprise people. The first is that a door open twenty minutes a day is the largest single load in the box, well ahead of the insulation everyone argues about. The second is the fans: they sit inside the insulated envelope, they run continuously, and every watt they consume ends up as heat the same system has to remove again.
The door is a habit, not a component
Door load scales linearly with how long the door stands open. Sixty openings at twenty seconds is twenty minutes a day of open doorway. Get it to ten seconds and you have removed half of that load without spending anything. Strip curtains and air curtains do the same job less politely, and the manufacturer will give you a reduction figure for them.
This is also where the moisture term lives. Warm air rolling in over the threshold carries water, that water condenses on the evaporator coil, and in a freezer it freezes there and has to be defrosted off. In a humid climate the latent part can exceed the sensible part, which is why the share is an input on this page rather than a constant.
Product pull-down is a schedule problem
Four hundred pounds of product spread over twenty-four hours is a modest load. The same four hundred pounds required to be at temperature within six hours is four times the load, and it lands at the worst moment, on a delivery morning, in a box that has also had its door open for a solid ten minutes. Sizing on a daily average and then operating on a six-hour deadline is a common way to end up with a box that works all winter and fails in July.
Whether the product actually has to be down within a given time, and what temperature it has to reach, is a health department question and a product question. It is not answered here — it is the input you type in.
Reading the answer honestly
The output is a load and an implied capacity at a run time you choose. It is not an equipment selection. A condensing unit is chosen against a suction temperature, an ambient temperature and an evaporator, and its capacity at your conditions is not its nameplate capacity. Take the load figure to a refrigeration contractor and let them do that part. What you have gained is the ability to tell whether their assumptions about your door traffic and your delivery schedule match reality.
For the household version of the same question the refrigerator size calculator is a different problem entirely — volume rather than heat. If the box is part of a wider build, the heat loss calculator uses the same conduction arithmetic on a building, and what you keep in the box connects to prep batch sizing and portion yield.
Questions people ask
What size walk-in cooler do I need for a small restaurant?
Volume and refrigeration capacity are two separate questions and this page answers the second. For volume, work from cases rather than covers: count what a delivery actually brings, allow the aisle you need to reach the back shelf, and remember that a box packed solid stops circulating air and develops warm corners. Once you have the dimensions, this page tells you what refrigeration that box needs given your door traffic and your delivery pattern.
Why does my walk-in struggle in summer but not in winter?
Three things move together. The kitchen around the box is hotter, so the panel load rises with the temperature difference. The air coming through the door carries much more moisture, so the latent part grows faster than the sensible part. And the condensing unit itself makes less capacity when the air it rejects heat into is hot. A box sized with no margin passes in February and fails in August, which is why the run time input on this page defaults to eighteen hours rather than twenty-four.
How much difference does a strip curtain make?
It reduces the door exchange, and the manufacturer will quote a percentage. Treat that quote as an upper bound achieved with curtains that are complete, clean and hanging straight, which is not the state most of them are in after a year of hand trucks. The bigger lever is the seconds figure. Cutting average open time from twenty seconds to ten halves the same load and costs nothing but a habit.
Do the evaporator fans really matter?
On the defaults here they are the single largest internal load, larger than conduction through all six panel surfaces. They run continuously, they are inside the insulated envelope, and every watt they draw is a watt of heat the system removes again. It is worth knowing what the fan motors in a box actually draw, because on a small cooler it is a meaningful share of the total and it is one of the few loads that responds to an equipment choice rather than to behaviour.
Can I use this for a freezer?
The structure works, but three inputs change substantially and one is easy to forget. The temperature difference is much larger, the latent share of the door load goes up because everything that comes in freezes on the coil, and the system efficiency figure drops a long way at low suction temperatures. The one people miss is the latent heat field: product being pulled through the freezing point releases a large amount of heat that has nothing to do with its temperature change, and leaving that field at zero will understate a freezer load badly.