kW/ton, EER, SEER2, COP and IPLV: HVAC Efficiency Metrics Explained

Every HVAC efficiency number answers the same question: how much cooling do you get for each unit of electricity? The confusion comes from the fact that the industry asks that question in at least eight different ways, under different test conditions, for different kinds of equipment. A facility manager comparing a chiller quote (in kW/ton), a rooftop unit spec sheet (in IEER) and a homeowner-style split system (in SEER2) is looking at three numbers that cannot be compared without translation.

This guide defines each metric, shows how to convert between them, explains what changed when the U.S. Department of Energy (DOE) moved to SEER2 in 2023, and closes with a worked example of what a change in kW/ton costs in a real plant.

The two building blocks: the ton and the Btu

Cooling capacity in the U.S. is measured in tons of refrigeration. One ton equals 12,000 Btu per hour of heat removal; DOE's Federal Energy Management Program (FEMP) uses the same definition in its O&M guide. Electricity is measured in kilowatts (kW) of power and kilowatt-hours (kWh) of energy, and the U.S. Energy Information Administration lists the conversion as 1 kWh = 3,412 Btu. Every metric below is built from those two facts.

The metrics at a glance

How to convert between kW/ton, EER and COP

Because the units are related by fixed constants, three formulas cover nearly every translation you will need:

• kW/ton = 12 / EER (12,000 Btu/h per ton divided by 1,000 W per kW)
• EER = 3.412 x COP (3,412 Btu per kWh, or 3.412 Btu/h per watt)
• COP = 3.517 / kW/ton (one ton is about 3.517 kW of heat removal)

The "typical context" column is illustrative. The conversions are exact; the context depends on the equipment and the conditions at the moment of measurement.

SEER2, EER2 and HSPF2: what changed in 2023

On January 1, 2023, DOE's new efficiency standards for central air conditioners and heat pumps took effect, expressed in three new metrics: SEER2, EER2 and HSPF2. DOE created the new metrics because it changed the test procedure (from Appendix M to Appendix M1) to better reflect real installations. According to AHRI, the M1 test raises the external static pressure, the resistance of ductwork that the indoor fan must push against, by up to five times. Higher static pressure means more fan power, so the same equipment earns a lower number under M1.

DOE is explicit that SEER2, HSPF2 and EER2 "are not directly comparable" to the old ratings. AHRI notes that even though the minimums only rose by about one SEER point numerically, minimally compliant products are approximately 7% more efficient than before. A 2015-era 14 SEER unit and a 2023 14.3 SEER2 unit are not almost identical; the second one cleared a tougher test.

2023 minimums that matter in Texas and the South

DOE divides the country into North, Southeast and Southwest regions. Texas is in the hot-humid Southeast region. In the South, the standards for split-system air conditioners are based on the date of installation, so a unit installed on or after January 1, 2023 must meet the new levels regardless of when it was built.

DOE describes EER as a peak-load measure taken at 95°F outdoors, and SEER as a seasonal measure. For a Gulf Coast building that spends long stretches above 90°F, the EER or EER2 value is the better predictor of how the unit behaves on the afternoons that set your demand charge.

IEER: the rooftop unit number

Commercial unitary equipment of 65,000 Btu/h (about 5.4 tons) and larger is rated in EER at full load and in IEER, a single part-load figure calculated per AHRI Standard 340/360. IEER blends EER measured at 100%, 75%, 50% and 25% of capacity, with outdoor temperatures that fall as load falls (95°F, 81.5°F, 68°F and 65°F, per AAON's summary of the method). For reference, the California Energy Commission's compilation of federal minimums lists 11.2 EER and 14.8 IEER for air-cooled air conditioners from 65,000 to under 135,000 Btu/h, and 11.0 EER and 14.2 IEER from 135,000 to under 240,000 Btu/h (values as of July 2025).

IEER replaced IPLV for unitary equipment. It is a good tool for comparing new units, but it is a lab-condition number. A rooftop unit with a 14.8 IEER nameplate can perform well below that once it has a few years of airflow, refrigerant charge and heat-transfer problems.

Chillers: full-load kW/ton, IPLV and NPLV

Chillers are rated under AHRI Standard 550/590. Two numbers appear on almost every chiller submittal:

• Full-load kW/ton at standard rating conditions (44°F leaving chilled water).
• IPLV (Integrated Part Load Value), a weighted average of efficiency at four load points. The weights are 1% at 100% load, 42% at 75%, 45% at 50% and 12% at 25%, with entering condenser water temperature falling as load falls. When efficiency is expressed in kW/ton, the weighted average is taken on the efficiency values (EER or COP) and then converted, which is why IPLV formulas in kW/ton are written as "one over" a weighted sum.

If a chiller will run at conditions other than the AHRI standard ones, for example a different chilled water temperature or a higher condenser water temperature, the same calculation is called NPLV (Non-standard Part Load Value). NPLV at your design conditions is more useful than IPLV at someone else's.

FEMP's federal purchasing requirements (updated October 2024) give a sense of current benchmarks. For a 300 to 399-ton water-cooled centrifugal chiller, the full-load-optimized path requires 0.544 kW/ton at full load and 0.520 kW/ton IPLV; the part-load-optimized path requires 0.595 kW/ton at full load and 0.370 kW/ton IPLV. FEMP also lists air-cooled chillers in EER: for example, 10.890 EER full load and 13.700 IPLV for full-load-optimized units under 150 tons.

Rated versus measured efficiency

Every rating above is measured on new equipment in a test lab. Your equipment is not new, and your roof is not a test lab. That is why operators of larger systems track field kW/ton:

1. Measure power with a true-RMS power meter or the chiller's onboard kW reading.
2. Measure cooling delivered. For a chiller, tons = gpm x chilled water temperature difference (°F) / 24, which comes from 500 Btu/h per gpm per °F of water divided by 12,000 Btu/h per ton. For air-side equipment, cooling output requires airflow plus entering and leaving air enthalpy, which is harder to measure accurately.
3. Divide kW by tons, and log the result together with outdoor temperature, condenser water temperature and load.
4. Compare like with like. A kW/ton reading is only meaningful against a baseline taken at similar load and similar condensing conditions. See how HVAC energy savings are measured for how professionals build that baseline.

Worked example: what 0.1 kW/ton is worth

Take a 400-ton water-cooled chiller that averages 240 tons of load over 3,500 cooling hours a year, or 840,000 ton-hours. The numbers below are illustrative; substitute your own load, hours and electric rate.

The same arithmetic works for a split system. A 14.3 SEER2 unit has a seasonal equivalent of about 12 / 14.3 = 0.84 kW/ton under test conditions. If that unit loses a meaningful share of its heat transfer ability, the same tonnage costs proportionally more to deliver, and on the hottest days it may not deliver the tonnage at all. For the full savings arithmetic, including demand charges, see how to calculate HVAC energy savings.

Which number to track for your equipment

Where efficiency goes after the rating

Ratings drift for physical reasons: dirty or fouled heat-exchange surfaces, incorrect refrigerant charge, reduced airflow, controls that drift out of sequence, and wear inside the compressor. Our guide to why HVAC efficiency declines with age covers each mechanism, and diagnosing efficiency loss shows how to tell them apart with field readings.

One mechanism that a nameplate never captures is the film of compressor oil that can coat the inside of evaporator and condenser tubes. CryogenX4 describes its product as a one-time treatment designed to lift that oil film back to the compressor sump and condition internal heat-transfer surfaces, and the company reports energy savings of up to 30%, with results varying by equipment condition. Whatever measure you consider, kW/ton measured before and after, at comparable conditions, is the fairest way to judge it. The how it works page describes the company's stated mechanism.

Next step

Pull the nameplate and AHRI ratings for your three largest cooling systems, convert them to kW/ton with the formulas above, and compare them with a week of measured kW/ton from your building automation system or a temporary power logger. A gap of more than about 10% at similar conditions is worth a diagnostic visit.