HVAC Airflow Problems: Why Most San Jose AC Repairs Blame The Wrong Thing

I am going to make a claim that will annoy a chunk of this industry, then back it up.

Most air conditioning problems diagnosed as refrigerant problems in San Jose are actually airflow problems. Not some. Most. The gauges show low suction pressure, somebody reaches for a jug, and the real fault, a duct system trying to breathe through a straw, goes untouched for another three years until the compressor quits.

It happens for a boring reason. Gauges are fast and dramatic. Airflow measurement is slow and requires drilling holes. One of those fits neatly into a twenty minute service call. The other one is correct.

Air Is The Whole Point Of The Machine

An air conditioner does one thing. It moves heat from inside your house to outside. Refrigerant is the delivery vehicle. Air is the cargo.

The evaporator coil can only pull heat out of air that actually touches it. Cut the volume crossing that coil and you have cut the machine's capacity, no matter how perfect the charge is. You bought four tons. If the blower delivers the air of a two and a half ton system, you own a two and a half ton system with a four ton electric bill.

Design target for conventional residential equipment is roughly 400 CFM per ton, meaning 400 cubic feet per minute for every 12,000 BTU per hour of capacity, with a workable range of 350 to 425. A four ton system wants somewhere near 1,600 CFM. If it is moving 1,000, refrigerant will not fix that, and adding refrigerant makes it worse.

The Five Usual Suspects

• The filter. Number one, forever, not close. This valley loads filters faster than the box assumes: remodel dust, spring oak and grass pollen, a couple weeks of wildfire ash every August. A one inch pleated MERV 13 in a house with pets is spent in six to eight weeks. And here is the trap, a high MERV filter in a system whose ductwork was never sized for the added pressure drop is a self inflicted airflow problem on day one, brand new.

• Undersized or crushed ductwork. The structural one and the expensive one. Huge numbers of 1960s and 1970s tract homes in San Jose and Santa Clara were built with ducts sized for two or two and a half tons, then somebody sold the homeowner a four ton unit and left the original ducts in place. Add flex sagging between joists, kinked at the takeoff, crushed under Christmas decorations, and the system physically cannot deliver rated airflow.

• Closed and blocked registers. Somebody shut two bedrooms to push more air to the living room. That is not how a duct system works. The blower does not redirect air, it fights higher pressure and moves less air everywhere.

• A dirty blower wheel. The invisible one. A squirrel cage has dozens of small curved blades, each one a tiny airfoil. Dust packs into those cups and turns a precision fan into a spinning drum. It looks fine and it is moving a third less air. Nobody catches this without pulling the blower assembly, which is exactly why nobody catches it.

• A dirty evaporator coil. The inlet face is wet whenever the system runs, and wet aluminum is a superb dust magnet. What builds up is a felt mat, and it sits on the back side of the coil where the access panel does not show it. Even a homeowner religious about filters can have a fouled coil, because most filter racks leak air around the edges.

Total External Static Pressure Is The Blood Pressure Of Your Duct System

Here is the measurement that separates a diagnosis from an opinion.

Static pressure is the resistance the blower has to overcome, measured in inches of water column. Total external static pressure, TESP, is the sum of all of it. The blood pressure comparison is not a flourish, it is how the number behaves. High blood pressure does not tell your doctor which artery is narrowed. It tells them a narrowing exists and the pump is straining.

Measuring takes five minutes. Two 3/8 inch test holes, one in the return before the blower, one in the supply after the blower and the heat exchanger. Magnetic pressure tips facing upstream into the air. A dual port manometer. Run it in cooling so the coil is wet, because that is the condition the manufacturer rated.

Then subtract, and here is where people fumble. The return reading is negative because the blower is pulling. Supply 0.31 and return negative 0.38 is 0.69 inches of water column, not negative 0.07. The negative side adds, it does not cancel. I have watched techs report a healthy TESP because they had that backwards.

One boundary before anybody gets ambitious with a cordless drill. Drilling static pressure test ports and changing blower speed taps are technician procedures, not homeowner projects, because both put you inside a cabinet with line voltage wiring, a gas valve or a heat kit, and components that do not forgive a hole in the wrong place.

Compare against maximum rated TESP on the rating plate. Most residential furnaces and air handlers are rated at 0.50 inches of water column. Actual has to land under maximum or the blower cannot deliver rated CFM. That is not a suggestion, it is the condition under which the published airflow tables are valid at all.

A 0.90 reading, which I find regularly, is roughly double the design ceiling. That blower is running at a load it was never built for and the coil downstream is a strong candidate to ice on the next hot afternoon. That is the mechanism behind most of the calls in the article on an AC freezing up in San Jose.

Once TESP proves a problem exists, component pressure drops say where it lives. High on the return side points at the filter or return ductwork. High on the supply side points at the coil or supply ductwork. That one fork cuts the search in half.

Measuring Actual CFM By Temperature Rise

TESP says a restriction exists. It does not tell you delivered airflow. For that, the cheapest accurate field method borrows the heating side as a known heat source.

Know exactly how much heat you are adding to an airstream, measure how much the air temperature rises, and you can solve for how much air is moving. More air, smaller rise. Less air, bigger rise.

On electric resistance heat the relationship is CFM equals volts times amps times 3.414, divided by 1.08 times the temperature rise. The 3.414 converts watts to BTU per hour. The 1.08 is the sensible heat constant for standard air. On gas equipment you substitute input BTU per hour times combustion efficiency in the numerator.

Say a 240 volt heat kit draws 40 amps, and after ten minutes the supply duct reads 96 degrees F against a 71 degree return, a 25 degree rise. Numerator: 240 times 40 times 3.414 is 32,774. Denominator: 1.08 times 25 is 27. That is about 1,214 CFM. On a three ton system that is roughly 405 CFM per ton. Airflow is not your problem, go look somewhere else.

Two things ruin this test. The blower has to be on its cooling speed tap, then set back afterward or you trip the furnace limit switch in December. And the supply reading goes a few feet downstream past the first turn, because radiant heat inflates it at the unit and understates your CFM.

California Title 24 Agrees With Me, Where It Applies

This is not one contractor's opinion, it is state energy code, and it is worth being precise about the boundary. Where Title 24 requires airflow verification, and under the 2025 standards that means heat pump installs statewide plus cooling only systems in certain climate zones, the floor is 350 CFM per ton verified by measurement. Not estimated. Not assumed from a speed tap setting. Santa Clara County is climate zone 4, so a cooling only replacement here often does not trigger it. That changes who is legally obligated to measure. It changes nothing about the physics, and 400 CFM per ton remains the design target regardless.

Sit with what that implies. The energy code treats measured airflow as pass or fail wherever it applies, yet a huge share of service calls on existing systems in San Jose never involve a single airflow measurement. The code caught up. Much of the trade did not.

Why A Contractor Who Never Measures Static Pressure Is Guessing

Airflow gets verified before charge gets judged. Every time, and the order is the entire argument.

A refrigerant reading taken on a system with bad airflow is garbage, because low airflow drives suction pressure down and the gauge looks exactly like an undercharge. The tech who skips airflow sees low suction and adds refrigerant. Now there is too much refrigerant in a system that still cannot breathe, and the fingerprint the next honest tech finds is unmistakable: low suction pressure sitting next to very high subcooling. That is not a fault the equipment produced. That is a signature somebody left behind.

Worth knowing which metering device you have, too. On a fixed orifice system, low airflow drives superheat toward zero, and zero superheat means liquid refrigerant is arriving at a compressor engineered exclusively for vapor. A TXV masks the same fault, holding superheat near normal while the coil quietly ices, which is why on a TXV system subcooling is the number that tells you the truth and superheat will smile at you and lie.

So when a contractor opens the panel, hooks gauges, and hands you a refrigerant answer in twelve minutes without a manometer ever leaving the truck, understand what happened. It might be right sometimes, the way a coin is right sometimes.

Static pressure takes five minutes and a fifty dollar drill bit. There is no excuse.

Where To Go From Here

To screen your own system with a cheap thermometer first, start with the AC delta T test. If it runs but the air is not cold at all, work through the AC blowing warm air diagnosis. And if trouble only shows up on the hottest afternoons, the outdoor side is the likely culprit, covered in the piece on a dirty condenser coil in valley heat. All of them point back here.