Energy recovery ventilator (ERV) commissioning field guide for HVAC

What an ERV does, and why the code pushed it onto the job

An energy recovery ventilator (ERV) is an air-to-air device that recovers energy from the air a building is throwing away and uses it to pre-condition the fresh outdoor air coming in. The exhaust leaving the space in winter is warm and humid compared to the cold dry air outside. Run the two streams past each other in the right kind of device and the incoming air picks up that heat and moisture for free, so the heating coil downstream has far less work to do. In summer it runs the other way: the cool dry exhaust takes the edge off the hot humid outdoor air before it reaches the cooling coil.

The reason these show up on so many drawings now is the energy code, not preference. ASHRAE 90.1 and the IECC require ventilation air, and conditioning that outdoor air is expensive, especially the moisture. Recovery is how you bring in the outdoor air the ventilation standard demands without paying the full conditioning penalty for it. The device exists to make required ventilation affordable.

Here is the part that gets skipped. An ERV that is installed but never commissioned looks identical to one that works. The wheel turns, the fans run, air moves. Whether it is actually recovering 70 percent of the energy or 30 percent because the streams are unbalanced and the filters are loaded is invisible until someone measures it. The hardware is the cheap part. The commissioning is what turns it into the energy savings the model promised.

ERV or HRV: which one is on the job?

An ERV transfers both heat and moisture between the airstreams, so it is a total energy, or enthalpy, device. An HRV, a heat recovery ventilator, transfers only sensible heat, the temperature, and leaves the moisture in its own stream. That single difference decides which one belongs on a given building, and getting it wrong costs comfort or humidity control for the life of the system.

Climate drives the call. In hot humid climates the latent load, the moisture in the outdoor air, is a large share of what the cooling coil fights, so an ERV that keeps that moisture out earns its keep year round. In cold dry climates where the job is mostly heating and the indoor air is already on the dry side, an HRV that recovers heat without dragging moisture back can be the better fit, because an enthalpy device would return humidity you may not want and can be more prone to frost. Mixed climates usually land on the ERV for the summer latent benefit.

Do not assume from the nameplate. Plenty of equipment gets specified one way and ordered another, or a building changes use and the recovery type no longer suits the load. Confirm the device type against the schedule, the climate zone, and what the space actually needs for humidity before you commission to a target that assumes the wrong physics.

The four recovery devices and how each moves energy

Four device types do the recovery work, and the commissioning checks change with the type. Know which one you are standing in front of before you start.

The enthalpy wheel, or rotary exchanger, is a slowly turning matrix coated with a desiccant. As it rotates between the exhaust and supply streams it picks up heat from the warmer side and releases it to the cooler side, and the desiccant carries moisture across by adsorption, driven by the vapor pressure difference. A sensible-only wheel does the same thing for heat without the desiccant. The wheel is the highest-effectiveness device and the one with moving parts, seals, and a cross-leakage path to manage.

The fixed-plate core, sensible or membrane, has no moving parts. The two airstreams flow through alternating channels separated by thin plates, and heat passes through the plate without the streams mixing. A membrane plate also lets water vapor migrate through, making it a total-energy fixed core with very low cross-contamination, which is why hospitals and labs favor it. The heat pipe is a sealed finned tube charged with a refrigerant that boils on the warm end and condenses on the cool end, moving sensible heat with no moving parts and zero air mixing. The runaround loop puts a coil in each airstream and pumps a glycol solution between them, so the two ducts can be far apart, which suits retrofits and industrial layouts where the exhaust and intake are nowhere near each other. The runaround moves sensible heat only and has the lowest effectiveness of the four, traded for that flexibility.

What does ERV effectiveness mean?

Effectiveness is the ratio of the energy the device actually transferred to the most it could have transferred given the two airstreams. It comes in three flavors. Sensible effectiveness is how well it moved heat, latent effectiveness is how well it moved moisture, and total effectiveness folds both into one number for an enthalpy device. An HRV has a sensible number and essentially no latent number.

The rating you compare against comes from AHRI Standard 1060, which is the performance rating standard for air-to-air exchangers used in energy recovery equipment. It sets the test conditions and reporting for airflow, sensible, latent, and total effectiveness, and pressure drop, so one manufacturer's wheel can be compared honestly to another's. When a schedule lists 75 percent sensible effectiveness, that figure traces back to an AHRI 1060 certified rating at stated airflows and conditions, not to a marketing sheet.

Two things about that number matter on the job. First, it is tied to specific airflows and a specific balance between the streams. Run the device off its rated airflow or with the two streams unbalanced and the real effectiveness drops below the certified figure. Second, effectiveness is not efficiency in the COP sense. It tells you how completely the device closed the gap between the streams, which is exactly what you will measure in the field to confirm the install.

The four airstreams and four duct connections

An ERV has four duct connections and two airstreams that cross inside it, and mixing them up at startup is a real and common error. Outdoor air enters from outside, passes through the device picking up energy, and leaves as supply air toward the building or the downstream air handler. Return or exhaust air comes from the building, passes through the device giving up its energy, and leaves as the relief or exhaust air to outside.

So the four ports are outdoor air in, supply air out, return air in, and exhaust air out. The energy crosses between the supply path and the exhaust path inside the wheel or core. The streams never combine in a well-built device, or they combine only by a small, controlled amount you will deal with as cross-leakage.

The recovery only reaches its rating when the two airstreams are reasonably balanced in volume. A wheel rated at 75 percent sensible was tested at balanced flow. Pull more supply air than exhaust, or the other way around, and the device cannot transfer at its rated level because one stream cannot give or take what the other is moving. That balance is the heart of the commissioning, covered below, and it is why the airflow numbers come before the effectiveness numbers.

What is exhaust air transfer ratio (EATR)?

Exhaust air transfer ratio (EATR) is the percentage of the exhaust airstream that ends up back in the supply airstream, the carryover that contaminates the fresh air you are bringing in. On a wheel it comes from two paths: the small slug of exhaust air physically carried across as the matrix rotates from the exhaust side to the supply side, plus any leakage through the seals in that direction. A fixed-plate or membrane core and a heat pipe have far lower EATR because nothing rotates between the streams.

You care about EATR because the exhaust you are recovering from is dirtier than the outdoor air, sometimes much dirtier. Recover from a restroom, a kitchen, or a lab and any carryover drags those contaminants back into the building's fresh air. ASHRAE 62.1 limits how much transfer is acceptable based on the class of the exhaust air, and for the more contaminated classes the allowable EATR is low.

The wheel's answer is the purge sector, a small section between the seals where a bit of clean supply-side air flushes the matrix before it rotates into the supply stream, sweeping the carried-over exhaust back out. A properly set purge can pull EATR down toward 1 to 3 percent, in some cases near zero, but it costs fan energy and it has to be set to the actual pressure relationship across the wheel, which depends on fan arrangement. A purge set at the bench for the wrong pressure does little. Confirm the purge angle and the pressure across it against the manufacturer's setup data as part of commissioning, and never recover from a high-hazard exhaust without checking what EATR the application allows.

The ERV feeding a DOAS

On a lot of modern jobs the ERV sits at the front of a dedicated outdoor air system, a DOAS. The idea is to decouple ventilation from space conditioning: the DOAS handles all the outdoor air and the moisture that comes with it, delivering neutral or slightly cool dry air, while the zone equipment, fan coils, beams, or VRF terminals, handles the room sensible load. The ERV is what makes that affordable, because it pre-conditions the outdoor air before the DOAS coil finishes the job.

The recovery and the coil work as a team, and commissioning has to treat them that way. The ERV knocks down the bulk of the load, then the heating coil, cooling coil, or downstream dehumidification trims to the supply setpoint. If the recovery is underperforming because the streams are unbalanced, the coil quietly makes up the difference and the energy bill carries it, with no alarm. The whole point of the DOAS is the decoupling, so verify the ERV is doing its share rather than leaning on the coil.

Check the supply condition leaving the full DOAS against design, and separately confirm the recovery device is hitting its effectiveness. Those are two different measurements, and a DOAS can hit its discharge setpoint with a dead wheel by burning energy at the coil.

How do you control frost in a cold climate?

In a cold climate the warm humid exhaust gives up its moisture as it cools inside the device, and when the surface drops below freezing that moisture builds as frost. Frost blocks the airflow path, drives up pressure drop, and on a core can crack it. Above a climate threshold the device needs an active frost control strategy, and confirming that strategy works is part of commissioning, not an optional extra.

The strategies split into prevention and defrost. Preheat puts a heater in the outdoor air intake to lift the incoming air above the frost threshold before it reaches the device, which preserves effectiveness but spends heater energy. Bypass routes part of the cold outdoor air around the device when frost threatens, accepting reduced recovery to keep the surface above freezing. Recirculation pulls building air back through to warm the device. On a wheel, slowing the rotation with the VFD keeps the matrix in the warm stream longer and melts frost without stopping ventilation. Exhaust-only defrost stops the cold intake for a period while warm exhaust thaws the device, at the cost of building pressure during the cycle.

Test the frost cycle, do not assume it. Force the control into its frost mode, or wait for conditions, and confirm the preheater fires, the bypass damper strokes, or the wheel slows as designed, and that it returns to normal when the threshold clears. The frost sequence that was never tested is the one that lets a core ice over on the first hard cold snap, and the callback lands in January when nobody wants to be on the roof.

The economizer bypass: when not to recover

Recovery is not always what you want. When the outdoor air is cooler than the return air and the building needs cooling, you want that cool outdoor air delivered straight, not warmed back up by exhaust on its way in. Running the recovery device during free-cooling hours actively fights the economizer, which is why a well-designed ERV has a bypass that takes the device out of the air path when the economizer logic calls for it.

On a wheel the bypass is usually stopping the rotation, since a stopped wheel transfers almost nothing while air still passes through it. On a fixed core it is a physical bypass damper that routes the air around the core. Either way, the control has to know when free cooling beats recovery and switch accordingly, and the two sequences have to hand off cleanly instead of hunting against each other.

Commission the bypass and the economizer interlock together. Confirm the device drops out of recovery when the economizer opens for free cooling, and that it re-engages when the changeover passes. An ERV that keeps recovering through every mild shoulder-season day is heating the supply air right when the building wanted that free cool air, and it can erase a chunk of the savings the economizer was supposed to bank. The economizer and demand-control ventilation guide covers the high-limit and changeover logic that decides the call.

Filters on both streams

An ERV filters both airstreams, the outdoor air and the exhaust, and both filters matter. The outdoor air filter protects the building and the downstream coil. The exhaust filter protects the recovery device itself from the lint, grease, and dust the building throws off, because a wheel or core that loads up with that debris loses effectiveness and gains pressure drop.

Pressure drop is the number to watch. Loaded filters and a dirty device add resistance, the fans work harder, and at constant fan speed the airflow falls off, which pulls the two streams out of balance and drops recovery below the rating. Record the clean pressure drop across each filter and across the device at commissioning, because that clean baseline is what the owner's crew compares against later to know when to change filters. Without the baseline there is no honest way to call a filter dirty.

Confirm the filters are the specified class and are seated with no bypass around the frame. Air takes the easy path, and a filter that leaks around its edges filters nothing while reading like it is installed.

The controls: wheel, VFD, dampers, and BMS

The control package is where an ERV either runs as designed or quietly runs wrong. The wheel motor starts, stops, and on many units modulates speed through a VFD, which the controller uses both for frost control and to hold a supply condition. The bypass damper strokes open and closed for the economizer and frost modes. Isolation dampers on the outdoor and exhaust openings should close when the unit is off so the building does not breathe through a dead device.

Walk every output and confirm it does what the sequence says. Command the wheel and watch it start and reach speed. Stroke the bypass and isolation dampers through their full travel and confirm the actuators drive both ways and the end switches report true position, not just commanded position. A damper that reads open at the BMS but is mechanically stuck is a classic, and the only way to catch it is to look at the blades.

Then prove the interlocks. The ERV should start and stop with the air handler or DOAS it serves, not run on its own schedule, or you are ventilating an empty building or recovering against a system that is off. Confirm the BMS points read true, the alarms annunciate, and the sequence of operations on the drawings matches what the controller actually does, because the submitted sequence and the loaded program drift apart more often than anyone admits.

How do you balance an ERV's airflows?

Balancing the outdoor and exhaust airflows is the first real commissioning step, and the rated effectiveness depends on it. Measure the supply (outdoor) airflow and the exhaust airflow, confirm each hits its design value, and confirm the two are in the balance the design intends, which for energy recovery is usually close to equal unless the building deliberately runs a pressure offset. The recovery device cannot transfer at its rating if one stream is starved relative to the other.

Measure with a duct traverse using a pitot tube or a thermal anemometer, or with a powered flow hood at the device, and read the airflow at the device, not three fittings away where leakage has changed the number. The blowercfm tool runs the traverse math, including the velocity-pressure averaging, so the field reading turns into an airflow you can compare to design. This is straight TAB work, and the air balancing and TAB report guide covers the traverse, the proportional method, and the tolerances in full.

Set the fans, on most units through their VFDs, to land both streams on design and in balance, then lock the settings and record them. A common mistake is balancing only the supply because that is the air the occupants feel, and leaving the exhaust wherever it fell. Unbalanced streams are the single most frequent reason a commissioned ERV measures well below its rated effectiveness, and it is invisible until you put the instruments on both ducts.

How do you verify ERV effectiveness in the field?

Once the airflows are balanced, you confirm the device is actually recovering by measuring temperatures across it and computing the sensible effectiveness, then comparing that to the rated value. You need four temperatures: outdoor air entering the device, supply air leaving it, return air entering it from the building, and you can use the relief air leaving for a sanity check. Take them with the unit at steady state and the streams balanced.

Sensible effectiveness is the temperature rise of the incoming outdoor air divided by the full temperature difference between the return and outdoor air. Measure outdoor air at 10°F entering, return air at 70°F entering, and supply air leaving at 55°F, and the device closed 45 of the available 60 degrees, which is 75 percent sensible effectiveness. That should land near the AHRI 1060 rating at the tested airflow. A field number well below the rating points to unbalanced streams, a dirty or fouled device, a wheel not turning at speed, or bypass leakage, in roughly that order of likelihood.

Two cautions. The temperatures must be read at the same moment with the streams balanced, or the math is comparing apples to a different hour. And a field temperature check captures apparent effectiveness, which can read slightly high because it picks up fan heat and case gains that the lab method strips out. Treat the field number as a confirmation that the device is working and roughly hitting its rating, not as a certified replacement for the AHRI value. For a total-energy device, repeat the logic with humidity ratios or enthalpies to check latent and total, which takes accurate humidity readings on each stream.

t_OA

Outdoor air dry-bulb temperature entering the device, the cold side in winter

t_SA

Supply air dry-bulb temperature leaving the device toward the building or DOAS coil

t_RA

Return or exhaust air dry-bulb temperature entering the device from the building

The functional performance test

The functional test drives the ERV through every mode the sequence claims and confirms each one happens, instead of trusting the BMS graphic. This is where commissioning earns its name. You are not checking that the unit can run. You are checking that it does the right thing in every condition it will face.

Confirm the wheel rotates and reaches speed, and that the VFD modulates it where the sequence modulates. Stroke the bypass damper through full open and full closed and watch the blades, not just the command. Force the frost sequence and confirm the preheat, slow-wheel, or bypass response fires and clears. Drive the economizer changeover and confirm the device drops out of recovery for free cooling and re-engages after. Cycle the isolation dampers and confirm they close when the unit stops. Then confirm the interlock with the air handler or DOAS so the ERV starts and stops with the system it serves.

Log a pass or fail and the as-found condition for each mode, with the deficiencies that need a fix. A functional test with no failures recorded usually means nobody pushed the sequence hard enough, because new systems almost always have a damper end switch, a frost setpoint, or an interlock that was never right. Finding those before the building is occupied is the entire job.

The energy code requirement

Energy recovery is often not a choice the designer made for savings. It is a requirement triggered by the energy code. ASHRAE 90.1, in its energy recovery provisions, and the IECC require a fan system to include energy recovery once the supply airflow and the percentage of outdoor air at design cross a threshold that depends on the climate zone and the system's annual operating hours. Large airflows, high outdoor-air fractions, and systems that run long hours are the ones caught.

The thresholds live in tables, and they have tightened over recent code cycles. Earlier editions commonly triggered recovery around a 5000 cfm supply system with 70 percent or more outdoor air, with a required recovery effectiveness on the order of 50 percent enthalpy. Recent editions pull that threshold lower for high outdoor-air systems and adjust the required effectiveness. Do not quote a single number from memory. Confirm the supply airflow, the design outdoor-air percentage, the climate zone, and the operating hours against the specific table in the adopted edition of the code, plus any local amendments.

Why this matters for commissioning: when recovery is code-required, the device hitting its rated effectiveness is part of how the building demonstrated compliance. A wheel that measures at half its rating is not just wasting energy. It can put the building out of step with the basis the permit was issued on. The commissioning record is what shows the recovery the code required is actually there.

What the owner has to maintain

The ERV does not hold its commissioned performance on its own. It drifts down as the device fouls and the consumables wear, and the owner inherits a maintenance load that, skipped, quietly erases the recovery. Hand over what has to be done and how often, with the clean baselines to measure against.

The device surface is the big one. A wheel matrix or a plate core loads with dust and, near kitchens, grease, and a fouled device loses effectiveness and gains pressure drop. Wheels get cleaned per the manufacturer's method, which is often compressed air or a gentle wash, never a wire brush that wrecks the desiccant coating. Belt-driven wheels have a drive belt that stretches and slips, and a slipping belt means the wheel turns slow and recovery falls, so the belt is a check item. Filters on both streams get changed on the pressure-drop baseline you recorded. The seals around a wheel wear and open up the cross-leakage path, raising EATR over time, so they get inspected and adjusted.

A dirty wheel is the most common reason a recovery system that passed commissioning is performing at half its rating two years later. The owner who never cleans the wheel is paying to run the fans through it while getting a fraction of the energy back. Spell that out in the turnover, because the people who will run the building were not standing on the roof when you commissioned it.

Data center and makeup air ERVs

Two applications stress the ERV differently than a typical office, and they change what you check. High outdoor-air makeup units, the kind serving kitchens, labs, and manufacturing, run near 100 percent outdoor air, so the recovery is carrying a very large load and the airflow balance and effectiveness verification matter even more. The exhaust on these is often a Class 2 or worse airstream, which puts EATR and the purge front and center, because the carryover you can shrug off in an office is a contamination problem here.

Data centers use air-to-air recovery in a different way, often to reject heat or to support economizer-style cooling while keeping the white space sealed from outdoor contaminants and moisture. The thermal targets follow the ASHRAE TC 9.9 envelope for the IT equipment, and the recovery device is one piece of a cooling approach that holds tight temperature and humidity limits. Cross-contamination control is not optional here, since outdoor air carrying moisture or particulate into the data hall is a reliability risk, so the low-leakage devices, plates and heat pipes, tend to win over wheels.

In both cases the commissioning fundamentals do not change. Balance the streams, verify effectiveness, confirm the frost and bypass logic, and pin down EATR. The application just raises the cost of getting any of them wrong.

What to document

The commissioning record is the proof the recovery the design and the code called for is actually present, and it is the baseline the owner's crew measures against for the life of the unit. Capture it per stream and per device, not as a single pass or fail.

For each airstream, record the design airflow, the measured airflow, and the temperatures entering and leaving the device. From those, record the computed sensible effectiveness against the rated value, and for an enthalpy device the latent and total checks. Add the clean pressure drops across each filter and across the device, the purge setting and the pressure across it, the as-left fan and wheel speeds, and a pass or fail on each functional mode. The table below is the minimum a reviewer needs to reproduce the result.

Common mistakes

The failures on ERV commissioning repeat, and they are almost all things that look fine until someone measures them. These are the ones that come back.

• Balancing only the supply air and leaving the exhaust wherever it fell, so the streams are unbalanced and effectiveness falls below the rating.
• Skipping frost control verification in a cold climate, then icing a core or wheel on the first hard freeze.
• Ignoring EATR and cross-contamination, especially recovering from a restroom, kitchen, or lab without checking the purge and the allowable transfer.
• Leaving the wheel recovering during economizer hours because the bypass interlock was never commissioned, fighting the free cooling.
• Handing over a dirty-wheel maintenance plan that nobody follows, so recovery is at half its rating within two years.
• Skipping the filter baselines on both streams, so there is no clean reference to call a filter dirty.
• Confirming the wheel spins and calling it commissioned, without ever measuring airflow balance or effectiveness.

Field checklist

Standards and references

AHRI Standard 1060 is the performance rating standard for air-to-air exchangers in energy recovery equipment, covering sensible, latent, and total effectiveness, airflow, and pressure drop, with EATR among its rated values. It uses the ASHRAE 84 method of test as its reference. When a schedule lists an effectiveness, that number should trace to an AHRI 1060 certified rating at stated conditions.

ASHRAE 90.1 and the IECC carry the energy recovery requirement, triggered by supply airflow, design outdoor-air percentage, climate zone, and operating hours through tables that have tightened across recent editions. ASHRAE 62.1 sets the ventilation rates the outdoor air has to meet and limits the exhaust air transfer based on the class of the exhaust stream, which is what bounds acceptable EATR. ASHRAE TC 9.9 gives the thermal envelope for data center applications. The exact thresholds, table numbers, and effectiveness values shift between code cycles, so confirm them against the adopted edition and local amendments before citing a number on a submittal.

For the airflow measurement and balance, the TAB standards from AABC and NEBB govern the traverse method and tolerances. Above all of these, the manufacturer's installation and commissioning instructions and the project specification control the specific setup, the purge angle, and the cleaning method. Where the spec is tighter than the code, the spec wins.

Units, terms, and conversions

Energy recovery carries a handful of terms that read differently across a schedule, a submittal, and a controls drawing, so the same device can be described several ways.

Airflow is in cubic feet per minute (CFM) on most North American drawings and liters per second (L/s) in metric documents, where roughly 2 CFM is about 1 L/s. Pressure drop across filters and the device is in inches of water column (in. w.c. or in. wg) or pascals (Pa). Effectiveness is a percentage, reported as sensible, latent, or total. Enthalpy, the total heat content the ERV transfers, is in Btu per pound of dry air, and the moisture it moves is tracked as humidity ratio, the pounds of water per pound of dry air.

ERV / HRV

Energy recovery ventilator transfers heat and moisture (total energy); heat recovery ventilator transfers heat only (sensible)

Effectiveness

The ratio of energy actually transferred to the maximum possible, reported as sensible, latent, or total percent

EATR

Exhaust air transfer ratio, the percent of exhaust air carried back into the supply stream

Purge

A sector on a wheel that flushes carried-over exhaust before the matrix rotates into the supply stream

DOAS

Dedicated outdoor air system, which conditions all the ventilation air separately from the zone equipment

Enthalpy

Total heat content of the air, sensible plus latent, in Btu per pound of dry air

Related tools

Calculators and readiness checks for this work

Compare your options

FAQ

People also ask

• What is an air-side economizer?
• What is air balancing?
• What is a chilled beam?
• What is a DOAS?
• What is a fan coil unit?