Why upstairs is always hotter, and why a bigger air conditioner will not fix it

In July we get the same call fifteen times a week. Downstairs is fine. Upstairs is eight degrees hotter and nobody can sleep in the back bedroom. The person calling has usually already been told by somebody that they need a bigger air conditioner.

They almost never do. Here is the order we actually work through it, and why bigger is usually the expensive way to make it worse.

Heat rises, but that is not the whole story

Yes, warm air is buoyant, and yes, a two-storey house will always tend to be warmer upstairs. But that tendency is small compared with the three real drivers, which are: how much cool air actually reaches the upper floor, how much heat that floor is gaining from the roof and the walls, and whether the system can move air at all.

A house with adequate airflow and a properly insulated attic will sit within two or three degrees floor to floor. If yours is eight or ten degrees apart, something specific is wrong, and it is findable.

First: is there any return air upstairs

This is the single most common finding in Minneapolis and Saint Paul. An enormous number of houses here have exactly one return-air grille, in the hallway or the stairwell on the ground floor, because that is how they were built or how the gravity furnace was converted.

A forced-air system is a loop. Air can only be pushed into a room as fast as it can get back out and return to the air handler. If the only path back is down the stairs, the upper floor is pressurised slightly, the supply registers up there deliver less than their rated flow, and the room does not get conditioned regardless of how big the outdoor unit is.

You can test the crude version of this yourself. Close an upstairs bedroom door with the system running and hold a tissue at the bottom gap. If air is streaming out under the door hard, that room has more supply than return and it is fighting you. A jumper duct, a transfer grille, or in some houses a proper second return, fixes it.

Second: what filter did you put in

People buy the best filter on the shelf, which is a completely reasonable instinct, and then wonder why the house got worse. A MERV 13 filter in a one inch slot designed for MERV 8 can cost a quarter of your airflow. In an already marginal system, a quarter of your airflow is the difference between two degrees and eight.

This is not an argument against good filtration. It is an argument for putting good filtration in a place that can accommodate it: a four or five inch media cabinet gives you MERV 13 with a fraction of the resistance, because the filter area is several times larger. That is a real fix, and it costs a small fraction of a new air conditioner.

Third: measure the static pressure

Total external static pressure is the reading that tells you whether the ductwork is fighting the blower. Most residential air handlers are rated for around 0.5 inches of water column. We routinely measure houses at 0.9 or 1.1, which means the blower is spending most of its effort pushing against the duct system rather than moving air into rooms.

High static comes from undersized returns, crushed flex duct in a joist bay, a filter that is too restrictive, a coil packed with dust, or a duct system that was designed for a smaller furnace than the one somebody fitted later. All of those are fixable and none of them is a new condenser.

Fourth: look in the attic and behind the knee walls

If the upper floor is a converted attic or a one-and-a-half storey bungalow, there are almost always knee walls: short vertical walls behind which sits an unconditioned triangular void, open to the attic. If those walls are uninsulated, or insulated with batts that have slumped, that bedroom is being heated from the side by a 130 degree attic all afternoon.

Air sealing and insulating that void does more for the room than any equipment change, and it works in winter too.

Why a bigger unit makes it worse

This is the part that surprises people. An oversized air conditioner satisfies the thermostat faster. It runs shorter cycles. During a short cycle, the coil never gets cold enough for long enough to condense much moisture out of the air, so the house ends up cool and clammy rather than cool and dry, and clammy at 74 feels worse than dry at 76.

Shorter cycles also mean less total air movement through the house per hour, which makes the floor to floor difference worse, not better. You have paid more, to be less comfortable, and to wear the compressor out faster.

When it genuinely is a capacity problem

Occasionally it is. If the static pressure is fine, the returns are adequate, the attic is properly insulated and the charge is correct, and the system still cannot hold setpoint on a 92 degree day, then the equipment is undersized and we will say so. A Manual J tells us which situation we are in rather than leaving it to argument.

And in a house where the upper floor genuinely cannot be reached by the existing ductwork, which is common in this city's older stock, the answer is oftena ductless head with its own thermostat rather than more central capacity that still cannot get up there.

What to ask for

If somebody quotes you a bigger air conditioner without having measured static pressure and looked at your returns, get a second opinion. It does not have to be from us. But the measurements should come before the quote, every time, and any contractor who is comfortable with that sequence will not mind being asked.

Our cooling page lists the six things we check, in order, anda diagnostic visit leaves you with the readings whether or not you buy anything.