Equipment · 7 Min Read

HVAC System Sizing: Why Tonnage Isn't Guesswork

How load calculations work, why oversizing is a common mistake, and what a properly sized system does for comfort and bills.

By the Superior Air team · CSLB #669421 · Updated September 8, 2026

Key points

A ton is 12,000 BTU per hour of cooling. The right number for your home comes from a Manual J load calculation, not from square footage or the size of the old unit.
Oversizing is the common mistake. An oversized system cools the thermostat quickly, shuts off, and leaves the house humid and uneven while wearing out the compressor.
Sizing does not stop at tonnage. Manual S matches real equipment to the load at San Diego design temperatures and Manual D sizes the ducts to carry the air.
Two San Diego homes of the same size can need very different systems. Coastal Climate Zone 7 and inland Climate Zone 10 have different design temperatures, and attic insulation, windows and orientation change the load more than floor area does.

The short answer

The capacity of an air conditioner or heat pump, its size in tons, should come from a load calculation of the house as it is today. The industry standard is ACCA Manual J, which adds up the heat that enters the house on a design day through the walls, the roof, the windows, air leaks, people and appliances, and arrives at the BTUs per hour the system must remove. The old unit's tonnage is a starting point for conversation, not a specification, because the builder may have guessed and the house has changed since. Square-feet-per-ton rules of thumb are worse, because they ignore everything that actually drives the load.

Getting it wrong in either direction costs comfort and money, but the common error in San Diego is oversizing, and it is common because it looks safe. It is not.

What a ton means

A ton of cooling is 12,000 BTU per hour. The term is a leftover from the ice trade: it is the rate at which a ton of ice absorbs heat as it melts over 24 hours. Residential systems run from 1.5 to 5 tons in half-ton steps, so 18,000 to 60,000 BTU per hour.

The nameplate is a nominal figure. A "3-ton" condenser delivers something like 34,000 to 36,000 BTU per hour at the standard rating condition of 95°F outdoors, and less on a 100°F afternoon in Poway, because capacity drops as the outdoor temperature rises. That is why equipment selection, covered below, looks at the manufacturer's expanded performance data at your design temperature rather than the number on the box. Heat pumps carry a heating capacity as well, rated at 47°F and at 17°F, and are sized so that both the cooling and the heating loads are covered.

Why square feet per ton gets it wrong

The rule of thumb is usually quoted as one ton for every 400 to 600 square feet. Here is what it does with two real kinds of San Diego house.

Two 1,800-square-foot San Diego homes with very different cooling loads. Illustrative; not a substitute for a calculation.
Factor	Home A: 1978 single-story, El Cajon	Home B: 2005 two-story, Carlsbad
Climate zone	Zone 10, inland. Design temperature in the mid-90s	Zone 7, coastal. Design temperature in the low 80s
Attic insulation	Original, thin by current standards; ducts in the attic	Modern depth; ducts partly inside the conditioned space
Windows	Single-pane aluminum, large west-facing slider	Dual-pane low-e, south exposure shaded
Duct leakage	20 to 30 percent of the airflow lost to the attic	Under 10 percent
Rule-of-thumb size	3 to 4.5 tons	3 to 4.5 tons
Load calculation result	Often 3.5 to 4 tons, and lower after duct sealing and attic insulation	Often 2 to 2.5 tons

The rule of thumb gives the same answer for both houses. A load calculation gives answers that can differ by nearly two to one. The reasons are in the rows: design temperature, insulation, window glass and orientation, and air leakage. Floor area is one input among a dozen.

The rule also fails within a single house. Two 1,800-square-foot homes on the same street in Santee can differ by a ton if one has a west-facing wall of glass and the other faces north, or if one owner added attic insulation and dual-pane windows and the other did not.

Manual J, Manual S and Manual D

Three ACCA procedures cover sizing from the load to the ducts. A proper design uses all three.

The three ACCA design manuals and what each one produces.
Factor	What it does	What goes in	What comes out
Manual J	Load calculation	Orientation, wall and roof construction, insulation, window area, type and shading, air leakage, occupants, appliances, and the local design temperatures	Heating and cooling loads in BTU per hour, split into sensible (temperature) and latent (moisture)
Manual S	Equipment selection	The Manual J loads and the manufacturer's expanded performance data at your design conditions	A specific outdoor unit and indoor coil or air handler combination whose real capacity fits the load
Manual D	Duct design	Airflow each room needs from its share of the load, the static pressure the blower can handle, and the routing	Duct sizes, layout and register selection that deliver the right air to each room

Manual J is the one most homeowners have heard of. Manual S is the one most often skipped, and it matters because equipment capacity at a Zone 10 design temperature is lower than the nominal rating. Manual S also caps how far above the load the selected equipment may go, roughly 15 percent over for cooling equipment with some extra allowance for heat pumps that must also carry the heating load, which is the formal reason "rounding up to be safe" is not allowed. Manual D is the one that decides whether the air the equipment produces actually reaches the back bedroom.

What oversizing does, and what undersizing does

An oversized system cools the air near the thermostat quickly and shuts off. That sounds like a feature. In practice it means:

Short cycling. Run times of eight or ten minutes instead of twenty or thirty. Each start is the hardest moment in a compressor's life, and an oversized system makes several times as many of them.
Poor humidity control. The indoor coil needs to run cold for a while before it condenses much moisture out of the air. A system that shuts off before that point leaves the house at the set temperature and clammy, which is why a coastal home on a humid September day can feel worse with a bigger unit.
Uneven rooms. Air only moves while the blower runs. Short cycles never push enough air to the rooms at the end of the duct runs, so the far bedroom stays warm while the hallway is cold.
Noise and drafts. More air through the same registers, in bursts.
Higher cost. Larger equipment costs more to buy, and the start-up inefficiency of short cycles eats into the SEER2 rating you paid for.

Variable-speed equipment softens the problem because it can throttle down, but it does not fix ducts sized for the wrong airflow, and it still costs more than the correctly sized unit would have.

An undersized system has the opposite, and milder, failure mode. On the hottest afternoons it runs continuously and the house may lag a degree or two behind the setpoint. The rest of the year it runs long, steady cycles, which dehumidify well and keep rooms even. Genuine undersizing, from an addition or a converted garage that the original system never accounted for, is a real problem and shows up as a room that never cools. But a system sized exactly to the load is not undersized, even though it runs longer than the old one did.

The design temperature explains the trade-off. Manual J uses the outdoor temperature that is exceeded only about 1 percent of the hours in a year, not the record high. Sizing for the record high guarantees oversizing during the other 99 percent.

Sizing in San Diego: coast, inland and the attic

On the coast, in Climate Zone 7, the marine layer keeps design temperatures in the low 80s and loads are small. A modest home in Pacific Beach or Encinitas often calculates to 1.5 to 2.5 tons. Oversizing is most common here, because installers carry inland habits to the coast and because "3 tons" sounds like a normal size. Coastal homes also have humidity to deal with in late summer, so the moisture-removal penalty of oversizing is felt most in exactly the places where it happens most.

Inland, in Climate Zone 10, design temperatures in Escondido, Poway, El Cajon and Santee sit in the mid-90s, and September heat waves go past 100. Loads are larger and the equipment loses capacity as the outdoor temperature climbs, so Manual S selection at the design temperature, not the nominal tonnage, is what keeps the house comfortable at 4 p.m. EER2, the efficiency at peak conditions, matters more here than anywhere else in the county; our SEER2 guide explains why.

The attic is the wild card in older housing. The 1970s tract homes across Mira Mesa, Clairemont, Santee and El Cajon have ducts running through attics that reach 130°F or more on a summer afternoon. Every foot of leaky, thinly insulated duct up there adds to the load. Sealing and insulating the ducts, or adding attic insulation, can drop the calculated load by half a ton or more, which is why we sometimes recommend duct work before equipment and why a load calculation should be run on the house as it will be, not as it was. Additions, enclosed patios, converted garages and replaced windows all move the load too, in both directions.

Heat pumps are sized on both loads. In San Diego the cooling load usually governs, and a heat pump sized for it has more heating capacity than the winter needs, which is why backup heat strips rarely run. The heat pump comparison covers the winter side.

Ducts: the half of sizing that gets skipped

Equipment capacity is meaningless if the ducts cannot carry the air. A 3-ton system needs somewhere around 1,050 to 1,200 cubic feet per minute of airflow across the coil. Ducts sized for 2 tons choke it: static pressure rises, airflow falls, the coil can freeze, the blower motor works harder and fails sooner, and the rooms at the end of the runs starve. The most common single problem we find in older San Diego homes is a return duct that is too small, which no new condenser can fix.

That is why the evaluation includes a static pressure measurement and a duct leakage test before equipment is selected, and why a proposal may include duct modifications or replacement designed under Manual D. It is also why California's energy code requires duct leakage verification on many changeouts. Our ductwork page explains what an evaluation involves and what sealing or replacement looks like.

How Superior Air approaches this

Measure the house. Room by room, with window sizes and orientation, insulation levels, attic conditions and construction type. We ask about additions and window replacements that are not obvious.
Run Manual J in software. Not a rule of thumb, not the old nameplate. The report shows the sensible and latent loads for the house.
Test the ducts. Static pressure and leakage are measured during the evaluation. If the ducts are the limiting factor, we say so and price the fix separately.
Select with Manual S. Equipment is chosen from the manufacturer's performance data at your design temperature, so a Zone 10 home gets the capacity it needs at 95°F, not just at the nameplate.
Show you the number. The load report goes with the written options. If it says 2.5 tons and the old unit was 4, we explain why, and you see the reasoning rather than a tonnage on a line item.

The AC replacement page covers the service from start to finish, and the replacement process guide shows where sizing fits in the sequence. To have your home measured, schedule a consultation.

My old system is 4 tons. Shouldn't the new one be 4 tons too?

Not necessarily. The old unit was sized by whoever installed it, often by rule of thumb, and the house has changed since: new windows, attic insulation, a converted garage, or a duct system that has been leaking for decades. We run a load calculation on the house as it is today. If it comes out at 3 tons, a 4-ton system would short-cycle and control humidity poorly. If it comes out at 4, the old installer got it right and the calculation confirms it.

Is a bigger air conditioner better for hot inland summers?

No. A system sized correctly for Escondido or El Cajon already accounts for the design temperature there, which is in the mid-90s. Going larger than the calculation adds short cycling, uneven rooms and compressor wear on the other 99 percent of the hours. What does matter inland is choosing equipment with the capacity the calculation requires at that design temperature, which is what Manual S checks, and a good EER2 rating for the hours at peak.

What information do you need from me for a load calculation?

Mostly nothing you have to look up. The technician measures the rooms, windows and orientation on site and inspects the attic and ducts. It helps to know when windows were replaced, whether attic insulation has been added, and whether any rooms were added or enclosed since the house was built, because those change the load and are not always visible.

Do I need to resize the system if I add a room or convert the garage?

Usually the load changes enough to check. An addition adds wall, roof and window area; a converted garage often has thin insulation and a big door that was never meant to be conditioned. The existing system may still carry it, or the addition may be better served by its own ductless unit. A load calculation answers that; guessing tends to leave the new room hot.

Ductless mini-splits for additions →

Why does my new system run longer than the old one?

Because it is sized correctly. A right-sized system on a hot afternoon runs for long stretches, which is exactly what removes humidity and keeps every room even. The old oversized unit cooled the thermostat quickly and shut off, which felt powerful and left the far rooms warm. Long, steady cycles are what a properly sized system is supposed to do.