Last updated: October 5, 2026. By Dan McKenna, founder, Elite Energy Management.
A rooftop unit that runs nearly all the time yet still holds setpoint is doing more work than the job requires, and the usual causes are a short list: lost cooling capacity from dirty coils or a slow refrigerant leak, restricted airflow, an economizer stuck open and pulling in hot air, a setpoint or schedule that asks too much, or a unit simply undersized for a kitchen's heat load. Because the space stays comfortable, nobody complains, which is exactly why constant runtime is worth catching: it is the quiet phase before a failure, and it is burning energy the whole time.
Why is constant runtime a problem if the space is comfortable?
Three reasons, in order of how fast they bite. First, energy: a unit that runs 90 percent of the time to do what it used to do in 60 percent is paying for that extra runtime on every utility bill, often for months before anyone notices. Second, wear: compressors and fans have finite hours, and a unit that never rests is spending them faster. Third, margin: a unit at full runtime on an ordinary day has nothing left for the hot one, so the first heat wave of the season is when "fine" becomes "down." The comfort is real; the unit is just borrowing against the next failure to deliver it.
What are the common causes?
| Cause | What usually points to it in the data |
|---|---|
| Lost capacity: dirty coils, slow refrigerant leak, aging compressor | Runtime creeps up week over week for the same outdoor conditions, and the space cools more slowly per minute of runtime than it used to |
| Restricted airflow: filters, belts, blocked returns | Similar runtime creep, often with longer recovery after doors open and sometimes short cycling as the unit overheats |
| Economizer stuck open | Runtime jumps on hot days specifically, because outside air is being pulled in while the unit tries to cool it |
| Setpoint or schedule too aggressive | High runtime from the day a setting changed, with no gradual trend before it |
| Undersized for the load | High runtime from day one, worst in the kitchen zone, and no change after maintenance |
The first two look alike at a glance and both start with a planned maintenance visit; the last three are settings and sizing questions that no amount of coil cleaning will fix. Telling them apart comes down to the shape of the runtime trend over weeks against the unit's own weather-normalized baseline, which is the comparison the methodology page describes, and the companion post on losing cooling capacity covers the gradual version in depth.
How much energy is it wasting?
Enough to measure. Elite Energy Management estimates each unit's expected energy use from its own history at the current weather and compares it to what the unit is actually doing, so a unit running constantly shows up as a dollar figure of waste rather than a hunch. The energy waste page covers how that estimate works. For context on the scale, the Department of Energy has estimated that fault detection and diagnostics could deliver up to 50 percent HVAC and refrigeration energy savings if widely adopted, and constant-runtime units are a large share of where that waste lives.
What should you do about it?
Start with the cheap and reversible: confirm the setpoint and schedule have not changed, then schedule a planned visit for filters, coils, belts, and the economizer. Watch the runtime after the visit. If it drops back toward the unit's baseline at the same outdoor conditions, the cause was maintenance and you are done. If it does not move, the conversation becomes sizing or a refrigerant and compressor check, and you are having it on a scheduled day rather than during the first heat wave.
Frequently asked questions
My thermostat shows the unit is maintaining temperature. Why would I worry?
Because the thermostat is reporting the outcome, not the cost of the outcome. A unit can hold setpoint by running twice as long as it should, and the thermostat will call that success. The runtime trend is where the problem is visible, and thermostats do not keep that history.
How do I know if it is capacity loss or just a hot month?
By comparing to the unit's own history at the same outdoor temperature. More runtime in July than May is weather. More runtime this July than last July at the same outdoor temperature, or more this week than three weeks ago at the same conditions, is the unit.
Can monitoring tell which cause it is?
It narrows it. The shape of the trend separates gradual degradation from a sudden change, and hot-day-specific runtime points toward the economizer. The technician on site confirms, and the data afterward confirms whether the fix worked.
What does this cost?
Elite Energy Management runs $5 to $12 per rooftop unit per month on standard tiers, with no hardware and no installation, full pricing here. A single unit running constantly for a summer typically wastes more than that in a month.
If your locations already have an energy management system or connected thermostats, a free 30-day pilot will show you which rooftop units are running far more than their history says they should: read-only access to your existing systems, first health scores within days, no hardware, no commitment.
Related reading: how do I know when a rooftop unit is losing cooling capacity, how much energy do rooftop units waste, and how detection and validation work.