Commercial HVAC replacement and installation in Roswell.

Commercial buildings like yours are usually cooled by packaged rooftop units, the equipment type that conditions nearly half of all U.S. commercial floor space. That means the replacement decision normally comes down to three paths, and they should be priced next to each other rather than assumed.

The cheapest quote is not automatically the cheapest decision. A unit you keep for two decades spends far more on power than it cost to install, which is why a high-efficiency specification exists at all. Ask for each path with its installed price, its confirmed lead time, and what it changes on the utility bill.

There is a fourth answer that is sometimes correct. If the unit is young and mechanically sound, advanced controls can be retrofitted instead of replacing the box, which is a legitimate alternative when it makes better financial sense.

Nobody can quote a commercial rooftop replacement from a tonnage number alone. These are the lines that separate a clean swap from a project, and each one should appear on the quote as its own price rather than buried in a lump sum.

The visible part of the job is one morning. The crane arrives, stalls are coned, the old unit comes off, and the new one lands on a curb adapter measured weeks earlier. Everything that makes that morning uneventful happened before it: the rigging plan, the structural check, the gas and electrical confirmation, and the weather call the day before.

The part that decides whether you are happy in year twelve happens even earlier, at sizing. Two decades change occupancy, plug loads, lighting, and sometimes the envelope itself. Replacing off the old nameplate carries forward whatever the last sizing got wrong, and an oversized unit short-cycles its way through comfort complaints that no amount of service will fix. Ask for the load calculation, not the nameplate.

Waiting for failure is its own decision. Facility teams often delay until the unit no longer conditions the space, needs constant service, or dies outright. Replacing proactively usually beats replacement-on-failure, and often beats a like-for-like swap at the end of a unit's service life. On failure you take whatever is in stock, at whatever lead time, in whatever season. Planned, you pick the equipment, the efficiency tier, and the week.

Two rooftop replacements at the same tonnage can land far apart on price, and the reason is almost never the equipment. It is what the building makes you do around it. Here is how to tell which one you have before the quotes arrive.

A quote written from a photo of the nameplate is a guess with a number attached. A useful one starts on the roof and works down. DOE's checklist for rooftop replacement decisions runs through the same list a good contractor walks: climate and heating load, utility bill impact, building type and how it is used, equipment availability, electrical capacity, physical constraints, and existing gas lines.

Electrical capacity is the item that surprises owners most. A gas-heat rooftop unit draws power for the fan and compressor. An all-electric replacement adds heating to that same panel, and the backup electric resistance element sizes the worst case. If the panel or the service cannot carry it, the honest answer is that the electrical upgrade is part of the project cost, not a change order in week three.

Physical constraints matter almost as much. Curb dimensions, roof access, refrigerant line routing, and whether gas piping stays or gets capped all change the labor. So does contractor familiarity with the equipment you choose, which is its own listed consideration, because unfamiliar equipment gets commissioned badly more often than it gets commissioned well.

Commercial mechanical replacement is permitted work in Georgia, and the permit language is blunt: a contractor who alters, repairs, replaces, or removes a mechanical system must obtain a permit before the work starts. The exemptions are narrow, covering portable equipment and replacement of a minor part that does not change the equipment's approval or make it unsafe. A rooftop unit is neither.

Two details matter for scheduling. Emergency replacements can proceed, but the permit application is due to the building official the next working business day, so the paperwork follows the crane rather than replacing it. And a permit application goes stale: it is treated as abandoned if the work has not commenced within six months of filing. Confirm the submittal path and the inspection sequence with the Roswell building department before the crane date is locked.

The install is a few weeks of attention. The consequences run for as long as the unit stays on the roof, and two of them are worth deciding on purpose rather than by default.

The first is what the unit costs to run. Old, inefficient rooftop units waste between $900 and $3,700 per unit each year, which turns the efficiency tier into arithmetic instead of opinion. Across organizations tracked through DOE's rooftop campaign, more than 43,000 upgraded units returned a combined $37 million a year in energy savings.

The second is refrigerant. EPA rules under Section 608 govern how refrigerant is handled, recovered, and recycled in stationary cooling equipment, and new units ship on newer low-GWP refrigerants. That changes nothing about day one and quite a lot about how you stock parts and write service contracts for a mixed fleet.

Buildings rarely have one aging unit. They have five, installed the same year, aging at the same rate. Replacing all of them in one season is clean but capital-heavy. Replacing them one failure at a time is cheap per event and expensive over a decade, because every emergency swap buys whatever is available instead of what you chose.

The middle path is a ranked plan. Rank by age, service history, and how much waste the unit is producing. The ones past fifteen years with recurring service tickets are usually both the worst performers and the most likely to quit on the hottest day. Replace the top of that list on a planned schedule and let the rest run on maintenance and controls until their number comes up.

Procurement follows the same logic. Ordering two or three units together often improves both pricing and crane economics, since a single mobilization can set more than one unit. Lead time governs everything else: a unit ordered in spring for a summer outage is a plan, and the same unit ordered in summer is a hope.

Write the multi-year sequence down, even loosely. It turns a series of emergencies into a line in the capital budget, which is where this belongs.

Two quotes are only comparable when they price the same scope. If one includes the curb adapter and the other does not, the lower number is not actually lower. Ask every bidder to break these out as separate lines.