A single central return serving the whole house
Common in older properties and almost always insufficient for cooling airflow once internal doors are closed overnight.
Central cooling is only as good as the path the cold air takes to reach people. The equipment produces capacity at the coil, and the duct system decides how much of it arrives in the bedroom at the far end of the house.
That is why a central air project begins with the distribution rather than the condenser. In homes that have never had cooling, and in homes where the ducts were designed for heating alone, the existing system frequently cannot carry the airflow that cooling demands.
A central air project covers more than a condenser and a coil. It covers the trunk and branch layout, register sizing and placement, return capacity, the condensate path, electrical supply and the physical siting of the outdoor unit.
Treated as a whole, it produces a house that is evenly comfortable. Treated as an equipment purchase, it produces a house with one cold room, one warm room and an owner who concludes the new system was a disappointment.
Before any figure is discussed it is worth agreeing what the finished system is actually meant to achieve, because central cooling projects are judged against expectations that frequently go unstated.
Whether every room is expected to hold the same temperature, or whether the household is content for bedrooms to be cooler than a rarely used dining room. Whether humidity is the real complaint rather than temperature, which changes the equipment recommendation considerably. Whether the system needs to be quiet enough for a bedroom directly above the air handler. Whether an addition, a converted attic or a garage workshop is inside the scope or outside it.
Those answers shape the design rather than decorate it. A house where even temperature across every room is the priority may need zoning or a second return upstairs. A house where humidity is the complaint benefits far more from two stage or variable capacity equipment than from extra tonnage. Establishing this at the survey means the proposal answers the actual requirement, and it means there is a clear standard to commission against at the end.
Warm air rises and cold air falls, and a duct system designed only to deliver heat can rely on that. It can also run at a lower airflow, because the temperature difference between supply air and room air is much larger in heating than in cooling. Cooling needs considerably more air moved to deliver the same comfort.
The result is that a duct system entirely adequate for a furnace can be too restrictive for a cooling coil. Static pressure rises, the blower struggles, the coil runs colder than it should and sometimes freezes, capacity falls and the rooms furthest from the air handler never catch up.
Return capacity is the most commonly undersized element of all. Many older homes have a single central return sized for heating, with bedroom doors that close over it. Air that cannot get back cannot be supplied, so closed rooms pressurize, airflow collapses and comfort follows it.
Assessment is straightforward. Total external static pressure measured on the existing system, register and return areas totalled against the airflow the new equipment will require, and the trunk sized against a target friction rate. Those numbers determine whether the existing distribution is reusable, improvable or genuinely needs replacing.
Where a property has never had forced air, the design starts from scratch and the constraints are architectural rather than mechanical. Available routes through floor and ceiling cavities, whether an attic or basement can carry the trunk, ceiling heights, and how much soffit or bulkhead the household is willing to accept all shape what is realistically possible.
Where a conventional duct system genuinely cannot be routed without unacceptable disruption, that is said plainly and the alternatives are laid out honestly, including ducted systems using small diameter high velocity distribution and zoned ductless arrangements. Recommending the approach that fits the building matters more than recommending the one already in the catalog.
Installation ends at commissioning, but a central system is not finished until it has been balanced. Balancing is the process of adjusting how the available airflow is divided between rooms so that each one receives the share its load actually requires, rather than whatever the path of least resistance delivers.
Without it, air takes the easiest route. The rooms nearest the air handler, fed by short straight branches, receive far more than they need. The rooms at the end of a long run with two elbows receive far less. The household compensates by setting the thermostat lower to make the worst room tolerable, which overcools everywhere else and raises the bill for the privilege.
The work itself is unglamorous. Airflow is measured at each register with a balancing hood or an anemometer, compared against the room by room design figures, and branch dampers are adjusted in sequence. Because closing one damper redirects air to every other branch, the process is iterative rather than a single pass, and the readings are taken again after each round of adjustment.
Transfer air is the other half of it. A bedroom with a single supply register and a door that closes overnight becomes pressurized, which chokes off the supply into it and simultaneously starves the central return. Jump ducts, transfer grilles or undercut doors give that air a path back. It is a small intervention that regularly fixes the one room nobody could ever get comfortable.
Second floors deserve particular attention. Heat accumulates upstairs, the load profile differs markedly from the ground floor, and a system balanced for an average across both will satisfy neither. Depending on the building this is solved by damper adjustment, by a dedicated return upstairs, or by zoning.
The balanced figures are recorded per room and handed over with the commissioning sheet. That record matters later: if a room starts underperforming in three years, there is a known good baseline to compare against, which turns a vague complaint into a specific diagnosis.
| Assessment | Healthy finding | What a poor finding requires |
|---|---|---|
| Total external static pressure | Within the equipment design value | Trunk, branch or return enlargement before new equipment |
| Total return area | Sufficient for the design airflow | Additional returns, or transfer paths for closed rooms |
| Trunk sizing against friction rate | Correct for the target airflow | Resizing, or accepting reduced capacity at the far end |
| Register count and placement | Matched to room by room load | Added or relocated registers to balance the house |
| Duct sealing condition | Joints sealed, boots sealed to drywall | Sealing works before capacity is committed to the design |
| Insulation in unconditioned space | Adequate for the location | Upgrading, or the losses appear directly in the running cost |
Static pressure, return area, trunk sizing and sealing condition are measured before any equipment conversation happens.
The load per room determines branch sizing and register placement, which is what makes the house even rather than merely cold somewhere.
Trunk and branches sized to a target friction rate, returns sized for the airflow, and transfer paths planned for rooms that close off.
Capacity selected against the calculation, and a blower capable of the required airflow at the static pressure the design produces.
Airflow verified at the equipment and at registers, the system balanced across rooms, and the measured values handed over.
Common in older properties and almost always insufficient for cooling airflow once internal doors are closed overnight.
A branch sizing or trunk sizing problem. New equipment makes the cold rooms colder and leaves the far rooms exactly as they were.
Each of those adds resistance that the blower has to overcome. A long sagging run can cost more airflow than a badly undersized branch.
Uninsulated or leaking ducts in those spaces lose a substantial share of the capacity before it reaches anybody, and gain heat on the way.
Rooms added later are frequently fed by a branch tapped off whatever was nearest, with no recalculation of the trunk feeding it.
Noise at a register is velocity, and velocity that high means the opening is too small for the air being pushed through it. It is an audible symptom of a sizing problem.
Extensions are frequently served by tapping the nearest branch. That borrows air from rooms that were already balanced and unbalances the whole system at once.
These are the questions that decide whether the finished system feels even or merely adequate.
Often yes, but rarely without modification. Heating ducts are sized for a lower airflow than cooling requires, and return capacity is usually the binding constraint. The assessment measures static pressure on the existing system and totals the supply and return areas against what the new equipment will need. Sometimes the answer is that the trunk is fine and two extra returns solve it. Sometimes the trunk itself is the limit. You get the measurement either way.
The blower works against a higher static pressure than it was designed for, so it moves less air. Less air across the coil means the coil runs colder, which reduces capacity and can freeze it solid. Run times lengthen, the far rooms never satisfy, the blower motor wears out early and the efficiency rating on the equipment becomes irrelevant. It is the most common reason a new system disappoints its owner.
It depends entirely on the routes available. A property with a full basement or an accessible attic can often be ducted with modest disruption. A two story house on a slab with finished ceilings throughout is a different proposition and may need bulkheads, closet chases or a different distribution approach. The survey establishes which of those you have, and the disruption is described honestly before you commit rather than discovered during week two.
Zoning is worth considering where parts of the house have genuinely different loads: a second floor that overheats, a west facing room, or a space used only part of the time. It is not a substitute for correct sizing or adequate distribution, and adding zone dampers to a restrictive duct system usually makes the static pressure problem worse. Fix the distribution first, then decide whether zoning adds anything.
Somewhere with the clearance the manufacturer specifies on all sides and above, because the unit rejects heat by pulling air through the coil and it cannot do that in a tight alcove. Beyond that, the considerations are noise relative to bedroom windows and neighboring properties, protection from roof runoff and falling ice, a level pad that will not settle, and a service path. The siting is agreed with you during the survey rather than decided on installation day.
Frequently, though it depends on the building. Basements and accessible attics give a trunk somewhere to live and branches somewhere to run, and in those properties the visible result is registers and occasionally a boxed section in a utility area. Finished ceilings throughout, a slab foundation and no attic access is the harder case, and there the options are closet chases, bulkheads along corridors, small diameter high velocity distribution, or a zoned ductless arrangement. The survey establishes which category your building falls into and the disruption is described honestly before anything is committed.
Central cooling projects almost always involve the ducts as much as the equipment.
A survey establishes whether your existing distribution can carry cooling, and what it would take if it cannot.
That answer determines the whole project, and it is far cheaper to learn before the equipment is ordered than after it is running.