Repair or replace: how to actually decide

Almost everyone has heard some version of the $5,000 rule: multiply the repair cost by the age of the system, and if the answer is over five thousand, replace it. It is a memorable rule and it is a poor one, because it ignores the two variables that matter most — how much life the equipment has left, and what happens to your running costs either way.

A better framing is straightforward. A repair is worth doing when the cost of the repair, divided by the years of service you can reasonably expect afterwards, is less than the cost of replacement divided by the years the new system would give you. That is it. Everything else is refinement.

Working it through

Say a repair is $900 on a fourteen-year-old system. If that repair realistically buys three more years, it is costing you roughly $300 a year. A $9,000 replacement expected to run fifteen years costs about $600 a year. On that arithmetic the repair wins comfortably — and it keeps winning until either the expected remaining life drops or the repairs start stacking up.

Now change one number. If that same repair only buys one more year because two other components are visibly near the end, it is costing $900 a year, and replacement is the cheaper decision even though the sticker price is ten times higher. This is why the age of the equipment matters less than its condition, and why a technician who has not measured anything is not in a position to advise you.

The three cases where replacement genuinely wins

• The repair is a major component — compressor or evaporator coil — on equipment that is already past its typical service life. You are spending a large fraction of replacement cost on a system where everything else is the same age.
• The system uses a refrigerant that is being phased down. Availability and price for older refrigerants move in one direction, and a system with a leak in one of them is on a clock regardless of what you spend today.
• The equipment was wrong for the house to begin with. If it was oversized, or the duct system was never adequate, you can spend money on repairs indefinitely and never fix the complaint you actually have.

Two questions worth asking

First: what did you measure, and what were the readings? A recommendation to replace should be supported by numbers — amp draw, superheat and subcool, static pressure, capacitor values. If the answer is vague, the recommendation is a guess or a sales target.

Second: if I repair this instead, what would you expect to fail next, and when? A technician who knows the equipment can answer that specifically. It is also the question that reveals whether replacement is being recommended because it is right or because it is bigger.

One thing specific to this coast

Systems here accumulate run-hours faster than the same equipment would further north, and coastal installations deal with salt corrosion on top of that. A twelve-year-old system a few miles from the water is not in the same condition as a twelve-year-old system inland, and any honest assessment of remaining life has to account for where the equipment has been living.