No cold air, ice on the copper, or a system that stopped by itself. Three very different problems that all arrive as the same phone call.
People call it a diagnosis. It is not. Ice forms because the evaporator coil is running below freezing, and there are exactly two ways to get there: not enough air moving across the coil, or not enough refrigerant inside it. Both end in the same white lump on the copper, and the lump then makes it worse, because ice is a very effective way of blocking the airflow that may have been the original problem.
This is also why the first thing we do on an iced system is usually nothing at all. Fan on, cooling off, and while it thaws we look at the filter, the coil face, the return, the ducts and the drain. You cannot read pressures through ice, and any pressure taken during a thaw is a number that means nothing. Somebody who names a cause while it is still frozen has decided in advance what they are selling.
Both restrict air. The system does not know or care which one is doing the restricting: less air arriving at the coil means less heat arriving at the refrigerant, the refrigerant does not boil off as it should, the coil temperature falls, and moisture on the fins freezes. Identical mechanism, identical symptom, and one of them costs four dollars and the other is a couple of hours with a coil cleaner and a garden sprayer.
The difference matters because they are found in different places. A filter you can see. A coil face loaded with a gray felt of dust on the entering side is behind a panel, and it gets that way slowly enough that nobody notices the system running longer each summer. If the filter is clean and the split is still poor, the coil is the next thing we open.
And a filter can cause it while looking spotless. A one-inch pleated filter with a very high MERV rating, in a system built around low resistance, restricts air on the day it is fitted. It is a real cause of frozen coils in houses where somebody was trying to do the right thing.
An air handler pulls several gallons of water a day out of Florida air. That water leaves through a small PVC line, and that line runs full of warm damp air and biological slime, and it blocks. When it blocks, the pan fills, and if there is a float switch fitted — and there should be, on both the primary drain and the secondary pan — it opens the low-voltage circuit and stops the system on purpose.
So the call that comes in as “everything died, nothing works” is very often a working safety device doing precisely its job, and the several hundred dollars of drywall it just saved you never becomes visible. We clear the line properly, check the trap and the slope and the vent, test the switch by lifting it, and tell you why it blocked. If there is no float switch on the secondary pan, we will suggest one, and it is a small part.
A run capacitor genuinely is the most common failure on a residential system. It is an electrolytic component living on a hot slab in a hot state, it dries out, and it takes the compressor or the condenser fan with it in the sense that neither will start properly without it. Replacing one is quick and the part is cheap.
Which is the problem. A cheap, common, quick part is the perfect thing to fit to a system that did not need it, on a hot afternoon, to a customer who is relieved it is running again. The defense is a number: every capacitor is stamped with its rating and its tolerance, usually six per cent, and a meter reads it in ten seconds. Ours goes on the report whether or not it gets replaced. If somebody swapped yours, it is fair to ask what it read before it came out.
A hard-start kit adds a start capacitor and a relay to give the compressor more torque for the first fraction of a second. There are two legitimate reasons to fit one. The first is a system with a long line set or a marginal electrical supply, where the compressor is genuinely being asked to start against conditions it was not sized for. The second is a compressor that is tiring — drawing high locked-rotor current, hesitating on start — and where a kit will very likely get you through a season.
The second use is fine as long as it is said out loud. It is a walking stick, not a cure, and it is a sensible purchase if you know that and a dishonest one if you were told it was a repair. When we fit one for that reason, the report says so, in those words.
We will not add refrigerant to a system with a known leak and describe that as a repair. We will not quote a compressor from a symptom without amp readings and a capacitor check. We will not condemn a system on the doorstep, and we will not replace a part that measured inside its own printed tolerance because the invoice looked thin.
Request a visit and the office will call you back to agree a time. We measure before we price, and you keep the readings.
Scripted demonstration. It cannot book, hold or confirm a time, and it will not quote a price.