Commercial HVAC replacement and installation in Birmingham.

Most rooftop replacements come down to three choices, and they are not equally good in every building.

The first is like-for-like: same tonnage, same fuel, same curb where possible. It is the fastest way back to conditioned air and the cheapest to engineer, and it carries forward whatever sizing mistake the old unit had.

The second is a high-efficiency unit of the same type. DOE's High-Performance Rooftop Unit Challenge asked manufacturers for units that cut energy use up to 50% against the standard in place at the time, and more than twenty models now exceed that specification. On a building that runs long hours, that difference shows up on the bill every month for the life of the unit.

The third is a heat pump rooftop unit. It is a close cousin of an air-conditioning unit but adds reverse-cycle heating, and it comes in dual-fuel, all-electric, and cold-climate versions. DOE's guidance treats today's heat pump rooftop units as an appropriate solution across most of the South and West, where they can handle most or all of the winter heating demand.

There is no honest universal installed price for commercial HVAC, because tonnage, roof access, and infrastructure swing the same nominal unit by a wide margin. What is stable is the list of lines a real quote contains. If one of these is missing, it has not been priced. It has been deferred.

The honest test for repair versus replacement is not the size of the last invoice. It is whether the unit can still hold the building at design conditions on the worst day of the year, and what it costs to do that.

Rooftop units condition close to half of all commercial floor space in the country, and most of them get replaced the same way: on failure, in season, at whatever the market charges that week. DOE's guidance is blunt about the cost of that habit. Units past about fifteen years waste substantial energy against newer models, and facility managers usually wait until the unit no longer conditions the space, needs constant service calls, or quits outright. Replacing proactively with a more efficient unit often beats both replacement-on-failure and a like-for-like swap at the end of the old unit's service life.

The practical version for a Birmingham building: if the unit is young and the failure is one component, repair it and put the unit on a watch list. If it is past fifteen years, needs a compressor or a coil, and serves a space with complaints it has never been able to fix, you are being asked to pay a large repair bill to extend a system that already loses money every month. Price the replacement before you approve the repair.

Two buildings can ask for the same tonnage and get quotes that are not close to each other. The gap is almost always access and infrastructure, not the box itself.

A useful site review answers the questions that decide equipment, not just the ones that decide price. DOE's own checklist for rooftop decisions runs the same list a good contractor walks: the heating load, the projected utility bill impact, the building type and how it is used, what equipment can actually be delivered, the infrastructure on site, and whether contractors nearby know the technology being proposed.

Infrastructure is where projects break, and three things get verified. First, electrical capacity. An all-electric unit with resistance backup heat can ask more of the service than the gas unit it replaces, and that gets checked at the panel rather than assumed. Second, physical constraints: curb footprint, clearances, structural capacity, and whether a crane can reach the position at all. Third, gas lines, which are kept, capped, or removed, and that is a scope decision with a cost either way.

The last item on the list matters more than it looks. Equipment that nobody nearby is comfortable servicing is a maintenance problem you buy on day one, and it belongs in the decision next to the efficiency numbers.

Two choices on a rooftop replacement follow the building long after the crane leaves, and both are far easier to get right before the purchase order than after.

The first is timing. Replacement-on-failure feels like it defers cost, but it hands you the equipment that happens to be available instead of the equipment you chose, at the worst point in the season, with emergency labor attached. The second is fleet consistency. A roof running mismatched units, mismatched controls, and mismatched refrigerants costs more to maintain than the same roof standardized, and that difference compounds across the full service life of the equipment.

Neither one shows up on a single quote. Both show up in the operating budget.

Most commercial roofs do not have one unit. They have a row of them installed within a few years of each other, which means they will reach the end of their lives within a few years of each other too. Phasing is how that stops being one enormous capital request.

Rank the units by age, service history, and the hours each one runs. The unit serving the space with the longest operating schedule is usually where a high-efficiency replacement earns back the most, so it goes first even when it is not the oldest unit on the roof. Units on short-hour spaces can often wait, or take advanced controls instead of new equipment. DOE's rooftop campaign treats efficient replacement and controls retrofits as two legitimate answers, chosen by which one makes better financial sense for that unit.

Procurement is the other half of the plan. Equipment availability changes what is realistically installable this quarter, so availability belongs on the quote next to the price rather than in a phone call after the purchase order. Check incentives before you sign, because program offers change and are usually tied to the efficiency tier you specify. And settle your refrigerant position once, for the whole roof, so the service contract and the spare-parts strategy are not quietly managing three platforms.

Two quotes are only comparable when they cover the same work. Ask each one to state these five items in writing, and the cheaper number usually explains itself.