The Ele(fan)t in the Room (Part 2): Supply Air Temperature Reset
Fan energy is the highest electric consuming component in commercial HVAC. The supply air temperature setpoint is one of the single biggest factors in determining fan energy and HVAC performance.
Fan energy is the highest electric energy consuming component in commercial buildings, surpassing chillers, pumps, lighting, DX compressors, office equipment, cooking, refrigeration, and computers. Fan energy falls behind only heating in overall commercial building energy consumption according to EIA data (Note 1). The last post presented reasons why fan energy is such a high percentage of building electric consumption and introduced the three main HVAC optimization approaches that are used to minimize fan power:
Supply Air Temperature Reset
Static Pressure Reset
In this post, I am going to describe the considerations and recommendations for the implementation of the second of these optimization opportunities: Supply Air Temperature Reset.
Why the supply air temperature setpoint is so important
The supply air temperature setpoint is one of the most important factors in determining fan energy because the amount of airflow needed to cool a space is proportional to the difference between the supply and return temperature. Increasing this difference by making the supply air colder, in a VAV system, reduces the amount of air that is required to be moved by the fan. Since fan power varies proportional to the cube of air flow rate (CFM), and proportional to the supply and return air temperature difference (Note 2), a small change in supply air temperature has a massive impact on fan energy, making supply air temperature optimization one of the most important controls optimization strategies and ways of reducing building energy use.
This is not a trivial optimization opportunity. This is a major and relatively easily implemented approach for reducing building energy consumption.
So then why are there so many different control approaches for supply air temperature setpoint determination? If it is such an important value, core to the function of air conditioning, what is the right setpoint? What is the optimal supply air temperature in terms of energy use? Is the optimal to maintain it at the design value? Vary it based on return air temperature? Vary it based on outside air or trim and respond?
This post looks at the factors for this setpoint, highlights the importance that fan energy plays into the overall optimization, considers the temperature and moisture considerations (psychrometrics), and identifies the utility that the ASHRAE Guideline 36 (“G36”) approach has for this optimization challenge.
Supply air temperature reset optimization isn’t simple
Supply air temperature reset is one of the more complicated optimization challenges because of the multitude of factors at play, from multiple energy flows (fan energy, cooling coil energy, reheat energy), economizer hours of operation, to indoor humidity control and psychrometrics impacts. Other reset strategies simply strive to make the setpoint as low as possible while maintaining the desired control function. Such is the case for static pressure reset, which attempts to make the setpoint as low as possible while allowing the VAV boxes to deliver the required airflow. This is not the case for supply air temperature reset. Optimal supply air temperature reset must look at both interior and exterior conditions and consider the complicated interactions.
Additionally, supply air temperature setpoint is often the go-to variable to override when spaces are too cold or too hot. Manually overriding this important setpoint and not using a best practice reset strategy can both cause the fan energy to skyrocket and cause interior comfort conditions to degrade, leaving the occupants feeling hot and clammy.
Thankfully, ASHRAE has funded research into the energy and operational dynamics of supply air temperature reset and integrated the control sequence recommendations into ASHRAE Guideline 36 (“G36”), which makes implementation simple.
Supply air temperature reset is code required
Most jurisdictions in the US reference energy standards based on the requirements of ASHRAE 90.1. Supply air temperature reset has been a requirement of ASHRAE 90.1 since the 2010 version. If your VAV multizone AHU system does not have an automated reset of supply air temperature in response to building loads or outside air temperature, it likely does not meet the requirements of current or recent energy codes.
The psychrometrics of 55°F supply air
The modern air conditioning industry was created when Willis Carrier determined that controlling the dewpoint of the supply air allowed for control of the space’s relative humidity, which is the principal function of air conditioning. He described this and other relationships of moisture and temperature in air in his 1911 “Rational Psychrometric Formulae” (Note 3). In modern air conditioning, control of the dewpoint is primarily done through supply air temperature setpoint control. The selection of that setpoint by the design engineer determines the remainder of the system design parameters: airflows, fan sizing, duct sizing, coil sizing, electrical infrastructure, chilled water temperatures, pipe sizing, etc. In operation, the supply air temperature determines the amount of airflow that is needed to cool the space and the ability of the HVAC system to perform its fundamental temperature and moisture control.
A standard value that HVAC design engineers use for cooling coil leaving air condition is often between 53-58F; sometimes colder, sometimes warmer, but this is the most common range, with a typical value being 55°F. Air systems are sized for their peak load requirements, which for the vast majority of systems occur in the summer when the outside air is humid.
A supply air of 55°F saturated cooling coil leaving air represents a condition that, after accounting for fan heat (which is a function of fan total static pressure, approximated as +2F in the below figure), when drawn on a psychrometric chart falls on a line known as the Sensible Heat Ratio line (SHR approximated at 0.75 in the below figure) which also falls on a standard typical interior space temperature design condition of 78°F/50% Relative Humidity (and also 72°F/58%, and 75°F/54%). The below figure shows this plotted on the psychrometric chart. The SHR ratio is calculated based on the peak occupancy of the space during the peak cooling condition and represents the amount of moisture generated in the space that must be removed by the air conditioning relative to the sensible heat. The greater the amount of moisture generated, the colder the air needs to be (per CFM) to absorb that moisture.
When the outside air is wet (summer), removal of moisture via the cooling coil is required. In this example, cooling the air to 55°F achieves the right balance of sensible and latent cooling, falling on the Sensible Heat Ratio (SHR) line.
Another way of describing this is that cooling warm humid air down to 55°F brings the dewpoint of the air down to approximately 55°F which wrings out just enough moisture to meet the dehumidification requirements of a moderately occupied space at ASHRAE comfort design conditions.
The thermodynamics do not prevent the operation from supplying colder air. Colder supply air would simply result in a dryer room, which is a good thing in the majority of summer conditions. It would also, as stated above, reduce the required airflow, savings fan energy.
You cannot supply warm air when the outside air is humid
In the summer, the outside air is often more humid than the inside space (i.e. the "dewpoint" or "absolute humidity" of the outside air is greater than that of the interior space) and the air must be cooled to remove moisture. Supplying air at a higher dewpoint than the space dewpoint will inject moisture into the space. If you don’t cool it to the space dewpoint, moisture is added to the space. Eventually, this will increase the relative humidity of the space and the occupants will end up feeling "clammy" because the space’s relative humidity increases, eventually outside the boundary for standard comfort conditions.
This was the principal point of Willis Carrier’ contribution to the industry of Air Conditioning. In short, when it is humid outside (summer) you cannot supply warmer air or your space will become humid. The below figure demonstrates this on the psychrometric chart.
When the outside air is wet (summer), removal of moisture via the cooling coil is required. In this example, cooling the air to 61°F does not remove enough moisture from the air stream, and injects moisture into the space which will result in a humid, clammy space.
You can supply warm air when the outside air is dry
In the winter and shoulder seasons, outside air is often less humid than the inside space (the "dewpoint" or "absolute humidity" of the outside air is less than that of the interior space) and the air does not need to be cooled to remove moisture. The outside air acts as a “sponge” to absorb any moisture generated in the space. Therefore, air can be supplied at any temperature colder than the room without concern for dehumidification, albeit with an impact on the amount of airflow needed for sensible cooling. In this condition, depending on the sensible load of the space, additional mechanical cooling may not even be required, in which case the air is in 100% economizer mode.
When the outside air is dry (winter), the cooling is sensible-only and therefore the supply air temperature setpoint can be increased without concern of moisture control.
Supply air must be colder than the space temperature in most commercial systems
Another important concept in the design and operation of commercial air systems is that at least one space, and often most spaces, require cooling even on the coldest days of the year. In the winter, the heating load is handled by the perimeter and/or perimeter zone terminal heating systems to offset envelope heat losses. Typically, the interior spaces do not have a heating load since they have no outside walls but do have heat-generating equipment and occupants that must be cooled. This cooling is performed by supplying colder air to the space than is removed by the return grills, and as such the supply air from VAV AHUs will always be colder than the return air, except in cases such as morning warm-up or another non-typical condition that requires the central AHU to heat the space rather than the terminal box or perimeter heating system. The difference between the return air and supply air (Air "Delta T") determines how much air is required to cool that space. Hence, colder air requires less airflow and less fan energy.
How cold is too cold?
There are operational impacts that limit how cold of air can or should be supplied to the space.
Overcooling: Depending on the parameters of the downstream terminal boxes, too cold of supply air can over cool the space. Specifically, the minimum terminal box flow parameter (Vmin) set during start-up/balancing, in my experience, is often arbitrarily set at values ranging between 30% and 50% of max flows. This is one of the biggest culprits of waste in air systems and buildings as a whole. Terminal boxes, which are likely already oversized for their actual peak design condition, are forced to deliver an arbitrary amount of air to a potentially lightly occupied or unoccupied space. If there is reheat at the box, it will engage. If there is no reheat, the space will over-cool. For this reason, it is imperative to deal with the Vmin opportunity in conjunction with supply air temperature optimization. This was described in greater detail in a previous post.
Dumping: The colder the air, the less buoyant it becomes, and too cold of supply air can “dump” from the supply diffuser and create a cold draft. This is one of the values of a Fan Powered Box, as it mixes return air with the supply air and mitigates the concern over air dumping.
Coil Approach: The equipment simply may not be able to meet too low of a setpoint, whether at peak conditions; or, depending on what else is happening in the system elsewhere, may not be able to supply colder air at off-design conditions either. Setting too low of a supply air temperature setpoint may cause the chilled water valve, for example, to go 100% open and cause issues up and downstream on the chilled water system including starving neighbors of flow and impacting the overall system CHW Delta T. Coils that are sized for a peak design load should be able to provide a better approach at part load conditions, but there is a limit to how low that approach can go and supplying air much colder than the design value may not be realistic.
Duct Insulation Condensation: For very cold supply air, the ductwork needs to be insulated to handle colder air so as not to “sweat”. As long as the supply air is not much colder than the design condition, this should not generally be a problem.
What is the optimal supply air temperature?
The optimal supply air temperature setpoint achieves the desired cooling, dehumidification, and desired space temperature setpoint at the lowest combined system energy cost. The discussion that follows is for a typical VAV HVAC system with variable speed fan and modulating terminal boxes.
The total combined system energy is made up of the summation of these three energies:
Fan Energy: Fan energy goes down with the cube (Note 2), so, the colder the air, generally the less fan energy. In a system with long duct runs and high total fan static pressure, this energy component becomes larger.
Cooling Coil Energy: When the supply air is cooled, the ventilation air is also cooled, and that increase in ventilation air cooling load is not negligible. The impact is greater with a higher percentage of outside air. Also, when using an air-side economizer that compares return air enthalpy to outside air enthalpy, the dryer space condition can reduce the hours of economizer operation and therefore increase cooling load, but this impact is not typically the primary driver.
Reheat Energy: When the terminal box has reached its minimum airflow setpoint (aka "Vmin"), if equipped with reheat, the reheat must engage to meet the space temperature setpoint, even on the hottest day of the year, depending on the occupancy condition. The higher the minimum terminal box setting, the greater the impact the reheat energy has on this overall balance. For lab buildings, where the minimum airflow is a function of required air-change rates rather than thermally driven, the reheat energy can be substantial, and this cost can be the primary driver.
Each of these energies has a different cost rate and is weighed differently. This can make it a very complicated optimization challenge.
Standardization: Or, How I Learned to Stop Worrying and Trust ASHRAE G36
The above discussion identifies the number of considerations that go into the challenges with supply air temperature reset. The actual optimization of this controlled variable requires an unrealistically complex analysis of the costs of these various energies. Thankfully ASHRAE has created a guideline, ASHRAE Guideline 36: High-Performance Sequences of Operation for HVAC Systems (aka “G36”) that includes a description of a best-in-class supply air temperature approach, as per the ASHRAE Control Theory and Application Technical Committee. Although it may not be the 100% “optimal” reset strategy since it does not calculate all of the above energy dynamics, it does create a simplified reset approach that is understandable, repeatable, and can be tuned by the controls vendor, and/or operator to fit the requirements of the space. For these reasons, it has my full support.
The specific supply air temperature approach is described in section 5.16.2 for multizone variable air volume systems. The approach varies the supply air temperature setpoint from a user-definable minimum to a user-definable maximum as the outside air varies from a user-definable minimum to a user-definable maximum. When the connected terminal boxes are also programmed with ASHRAE G36 sequences, the reset is "trimmed" based on feedback from the boxes. For the majority of existing systems, the terminal boxes are unlikely to have the resets programmed, and as such the maximum supply air temperature reset will be purely a function of outside air, which is an acceptable alternative since the reset values can be adjusted. The guideline gives the user the freedom to set these reset parameters based on design and experience. In my opinion, it is not the perfect approach, but a simplified sequence that works, and is well worth the benefit of being a standard approach that can be employed across a variety of systems and is commonly understood.
ASHRAE Guideline 36 (G36), section 5.16.2 Supply Air Temperature Control, describes the rest of supply air temperature based on outside air conditions with a trim-and-respond approach based on terminal box cooling requests.
Implementation of Supply Air Temperature Reset
The fundamental supply air temperature reset strategy described in ASHRAE G36 is not complicated and can be implemented with a tweak to the programming of the AHU building automation code. The trim and respond logic, however, requires that the terminal boxes have the programmed variables for “cooling calls” which is most often not the case for existing boxes (hopefully adoption of G36 will become more standard and terminal boxes will come standard with G36 programming). For existing VAV systems, in absence of a full G36 upgrade, the return air temperature could substitute for the terminal box calls, but this is not the G36 recommended approach; it is a shortcut.
It is important to note that resetting the supply air temperature setpoint exclusively on the return air temperature, without concern for outside air can sacrifice space relative humidity control and also be unnecessarily wasteful of fan energy. This is a common approach for supply air temperature setpoint reset and should be avoided.
Ideally, the upgrade to the unit controls would come with a full controls system upgrade. A well-implemented control system upgrade that includes G36 sequence can fit the definition of a perfect energy project, which is one that:
Improves the performance of an existing system, and ideally one that is nearing its end of useful life and in need of replacement anyway.
Leverages energy rebates to partially or fully fund the project
Has a lasting reduction in energy costs, that results in a highly attractive rate of return.
In my experience, a controls upgrade project, when coupled with optimization, meets this criterion for many sites and should be carefully considered.
Conclusions
Fan energy is one of the largest sources of energy waste in the built environment and a major opportunity for optimization.
Optimal supply air temperature reset is a complicated dynamic of fan energy, reheat energy, cooling energy, each with a differently weighted energy cost.
Supply air temperature is often used improperly as the primary control variable for space comfort rather than airflow. This both leads to humid “clammy” spaces and unnecessarily high fan energy. Resetting the supply air temperature setpoint based on return air alone without factoring in outside air conditions is not recommended.
I recommend consideration for implementing ASHRAE Guideline 36: High-Performance Sequences of Operation for HVAC Systems for both the AHU and terminal boxes.
When paired together, along with the implementation of a controls upgrade and overall system optimization, the project can achieve a high rate of return.
Notes and References:
Energy Information Administration (EIA)- Commercial Buildings Energy Consumption Survey (CBECS)
The fan affinity laws describe the relationship between power, pressure, flow rate, and speed for a variable flow centrifugal fan/pump system. In a typical real-world VAV system, where VAV boxes respond to duct supply air pressure by opening and closing dampers and the fan efficiency does not remain constant, the affinity laws are an approximation, but the order of magnitude impact is close. For example, a 20% decrease in airflow or increase in air temperature difference results in a ~50% reduction in fan power.