24th May 2021
Part 1: Understanding Unmet Load Hours (UMLHs)
In this, Part 1 of our Unmet Load Hours series, we dive into the fundamentals of UMLHs—much more so than the usual one-paragraph definition. If you already have a truly strong understanding of UMLHs, you may want to skip this part; however, you may also find it informative in some unexpected way!
An unmet load hour is any hour of any day of the year when one or more zones in the building are outside of their temperature control range. Typically, this translates to an hour in which one or more zones is outside of the thermostat setpoints plus or minus one half of the temperature control throttling range. Any hour with one or more zones with an unmet cooling load or unmet heating load is defined as an unmet load hour.
The fundamental concept underlying the UMLH test is that the ability to maintain space temperatures within the throttling range is an indication that the simulated heating and cooling equipment and zone airflow are adequately sized and appropriately controlled. If they are not, this will result in UMLHs. The same will be true with respect to appropriate zoning and other factors that will be covered in subsequent sections of this UMLH user guidance.
Temperature Control and Throttling Range
The overall temperature control range for a space is a composite of the deadband between heating and cooling setpoints and the throttling range for each of these. The throttling range defines the temperature range over which zone-level heating and cooling controls operate. Typical throttling ranges for heating and cooling are about 2¬¬–4°F or 1–2°K where metric units are used. In classic proportional control schemes, these are normally centered upon their respective setpoints; however, some control logic places the entire cooling control loop or throttling range above the cooling setpoint and the entire heating control loop or throttling range below the heating setpoint. Either way, this allows for a range of space temperatures over which the HVAC supply airflow rate and temperature, as well as water flow rates for coils and hydronic units, can be modulated to account for changes in demand. The graphic below illustrates this in terms of a dual-maximum control sequence for a variable-air-volume terminal unit with 4°F (2.22°K) heating and cooling throttling ranges.
Control sequence and throttling range for dual-maximum VAV airflow, cooling supply air temperature reset, and zone reheat coil leaving air temperature (IP and Metric units):
The graphic above describes a “dual-maximum” control sequences used in VAV terminal units wherein both cooling and heating airflow rates are variable (thus separate maximum airflow rates for each). The heating control in this case firstly ramps up the supply air temperature (SAT) as the zone temperature drops, and follows this with an increase in supply airflow if the zone temperature continues to fall. Conversely, the cooling control sequence is to firstly ramp up the airflow as the zone temperature rises, and then to follow this with a request for colder air (SAT reset from the air handler) if the zone temperature continues to rise above the zone setpoint.
Note: Whereas the term “deadband” is commonly used to describe the temperature range between the heating and cooling setpoints, which may or may not overlap the heating and cooling throttling ranges, this graphic defines deadband more narrowly as the range of zone temperatures for which the HVAC airflow will be fixed at its minimum and the supply temperature will not be modulated (the thermostat will not request either warmer or cooler air).
The overall control range of concern for UMLHs encompasses not only the heating and cooling setpoints, but also space temperatures within the throttling range above and below those setpoints. Heating and cooling setpoints are typically about 5–6°F or 3–4°K apart during occupied hours, and the throttling ranges extend that another 1–2°F or 0.5–1°K below and above their respective setpoints. Thus, in the example illustrated above, the relevant control range for UMLHs would extend from 67°F to 77°F or 20°C to 25°C in the metric version of the graphic. This is the overall range of zone temperatures in which the system will be taking corrective action to return the space to the setpoints or deadband between them.
In the ApacheHVAC module of the IES Virtual Environment, the default throttling range is 4°F when working in inch-pound (IP) units and 2°K when working in metric units (as in the graphics above, this is a unit-specific default value, not a unit conversion). This throttling range encompasses both airflow and temperature modulation, sequentially, in a variable air volume (VAV) system or similar. In the example of cooling operation, the flow through a single-duct VAV terminal unit would modulate from its minimum flow to maximum flow over a range of 2°F or 1°K, and the supply air temperature (SAT) would then reset over a range of 2°F or 1°K. Some actual controls may use smaller throttling ranges, which can be modeled, or may use PID loops, which are more difficult to model, to provide tighter or more stable control. In any case, it is generally considered best practice to sequence the modulation of airflow and temperature without overlap, and this sequencing must occur over some range of change in the zone temperature.
Energy codes and green building rating systems requiring a check for UMLHs typically assume a tolerance of 1¬¬–2°F or 0.5¬¬–1°K outside the zone temperature range bounded by heating and cooling setpoints. Unfortunately, this assumption may not be stated*, and where it is the description can be vague. Ideally, the tolerance allows just for the portion of the heating and cooling throttling ranges that lie outside of the range bounded by the heating and cooling setpoints. Thus for the example illustrated above, once again, the control range for UMLHs including these tolerances would extend from 67°F to 77°F or 20°C to 25°C.
*Unlike most other compliance standards, the California Title-24 Energy Code Alternative Calculation Methods Manual includes requirements for control throttling range and associated UMLH tolerance. See Appendix A: Specific UMLH Requirements of Codes and Standard Supported by IESVE, below.
In most cases, the tolerance is simply a way to inform the UMLH calculation of the amount of the cooling throttling range above the cooling setpoint and heating throttling range below the heating setpoint. So, why not just have the software always assume a tolerance of one-half the throttling range? There are a number of reasons for this:
Some energy codes, such as CA Title-24, specify the control throttling range and disallow any tolerance beyond that, making clear that the tolerance is intended to cover just the throttling range.
For most building simulation software tools there is either a control range or a default tolerance, which may or may not be editable (for software that doesn’t actually model the dynamics of control bands, this tolerance can be a proxy for the control band). In any case, the tolerance should be reasonably close to the limits of the overall control throttling range and must be the same for Baseline and Proposed models.
As will be described in Part 2 of this guide, the IES Virtual Environment provides means of setting or adjusting that tolerance in keeping with requirements of a given energy code or rating system or particular user needs, such as when modeling a system with broader or narrower throttling ranges.
Graphic example of Unmet Load Hour Analysis
The plot above provides an example of the various factors considered in the UMLH analysis for a particular HVAC zone. The light blue (cyan) line indicates the relevant time period on this day, as this is when the space is assumed to be occupied and the HVAC system is scheduled to condition the space within the occupied-hours setpoints. Whereas there may be occupants present during other times (e.g., a small number of maintenance staff staying on after the building is closed), or there my be times when spaces are meant to be fully conditioned in spite of being unoccupied, it is important that the UMLH analysis is constrained to the time period when the HVAC system is scheduled to meet the occupied space setpoints.
The occupied-hours heating and cooling setpoints in the example results plotted above are 70 and 75°F (21.1 and 23.9°C), respectively. The +/- 2°F (+/- 1.11°K) tolerance values allow for the control throttling range extension below the heating setpoint (responding to room temperatures down to 68°F / 20°C) and above the cooling setpoint (responding to room temperatures up to 77°F / 25°C). With ample cooling airflow that is not topping out on this day in July (magenta line does not go flat at the top), and zone temperature (blue line) maintained just below the cooling setpoint (green line), the VAV box is not yet requesting cooler air from the system. As this is true for all 15 HVAC zones on the system, the system supply air temperature (red line) is not required to go below the warmest reset value (60°F / 15.6°C).
The second plot shows that during the 7:30 AM to 5:00 PM test period for UMLHs (the light blue/cyan line labeled “Cooling occupied and available” on the previous plot) all zones on the system are maintained above the 70°F / 21.1°C heating setpoint and below the 75°F / 23.9°C cooling setpoint. Thus, there are no UMLHs on this particular day, even without including the tolerances for control throttling ranges. In fact, as a result of system autosizing, this generic U.S. Department of Energy (DOE) “medium office building” model with standard code-compliant constructions, pre-defined VAV system with parallel fan-powered boxes (system 08a), and located in Chicago, IL, has zero UMLHs for the entire year, even with the tolerance reduced to +/- 1°F (+/- 0.55°K). This leaves room for either more diverse and perhaps more realistic load profiles associated with building use patterns or types of zones not included in this generic office model.
Most energy codes and green building rating systems allow for some number of UMLHs. The assumption is that equipment and airflow sizing will account for the majority of possible conditions, but not the most extreme outliers, and again that both real HVAC systems and computer simulations are less than perfect. Therefore, in light of appropriate sizing to design loads at near-peak conditions and imperfect controls, it should be acceptable for space temperatures to occasionally stray outside the control ranges.
It is also common for energy codes and green building rating systems to emphasize either a difference in unmet load hours between the proposed design model and a prescribed baseline model or a maximum value that must be applied to both models. The point of this is to ensure that the equipment sizing, airflows, and controls in the proposed design model are not significantly worse than those in the baseline. This comparison is somewhat independent of the absolute accuracy of the modeling.
For many years, ASHRAE 90.1 Appendix G and USGBC LEED Green Building Rating System EA Credit 1 stipulated that the difference in UMLHs between the Baseline and Proposed be no more than 50 hours. This language was included up through LEED v2.2, but later removed, as the 300 maximum UMLHs for both models covers this quite effectively, and allows for the possibility of the Proposed design being significantly better than the Baseline (i.e., having significantly fewer UMLH). For example, while it would not have been allowed under earlier rules, it would be acceptable under the new rules for the Proposed model to have fewer than 100 UMLHs while the Baseline model has 250. This would suggest that either the Proposed design has achieved a reduction in transient loads, the Proposed HVAC system is better suited and/or controlled to meet the loads, or both.
Spaces for which UMLHs are counted
UMLHs are normally counted only for regularly occupies spaces in the building. Thus unoccupied spaces, such as plenums, voids, and crawl spaces, and other spaces only occasionally occupied, such as storage rooms, stairwells, restrooms, mechanical rooms, parking garages, etc., are not counted. Both Appendix A and Part 2 of this guide provide more detail regarding which spaces are excluded from the UMLH analysis.
This is important, as inadvertently including irregularly occupied spaces that are only semi-conditioned can throw off the number of UMLHs for the entire project. In the example below, a restroom conditioned only by transfer air entering under the door to make up for the exhausted air has been inappropriately but intentionally (just to illustrate the point) included in a manual test for UMLHs.
As can be seen in the manual UMLH test above for a building model in Orlando, FL, the restroom that is semi-conditioned only by exhaust fan and transfer air tends run a bit on the warm side. Additional tests confirmed that the restroom ranged from 0.5 to 3 degrees outside of the cooling control band for the adjacent fully conditioned spaces. This is not very much, and most occupants would not be bothered by it, but it is enough to make the UMLHs value for the entire building unacceptable. Thus, it is essential that unconditioned and semi-conditioned spaces are excluded from UMLH tests.
When UMLHs are counted
Most green building rating systems and energy codes, such as LEED and California Title-24, count UMLHs only during the occupied hours of building operation when HVAC systems are operating continuously. In other cases, UMLHs are required to be counted whenever the building is conditioned, including nighttime setback, when the building is unoccupied but still conditioned to a relaxed setpoint. The manual UMLH test described in Part 2 of this guide can be used when either UMLHs are required to be counted 24 hours a day, seven days a week or when it is desirable to do so as means of informing the design.
Specific UMLH Requirements of Codes and Standards
Appendix A to Part 1 of this UMLH Guidance document, below, summarizes key requirements of the energy codes and standards—and thus indirectly green building rating systems such as LEED—supported by the IES Virtual Environment. This includes definitions, rules, limits, and space type exclusions for UMLH test compliance for each code or standard covered by the IES-VE software.
Continuing on to Parts 2–9 of this Unmet Load Hours guide
Part 2 of this guide will explain how UMLHs are determined within the IES Virtual Environment. This includes which spaces are tested, which hours are tested for those spaces, the setpoint values used, the settings for control Throttling Range, and the related UMLH tolerance value. This section describes where the associated user inputs are located within the software and how these affect the outcome.
Parts 3–9 will provide guidance on how to locate the sources of UMLHs, how to minimize UMLHs though appropriate zoning, loads analysis, system sizing, and simulation settings, addressing the use of transfer air, specialized building types, and multiple systems serving the same spaces.
Part 3: Identifying the problematic rooms/zones using VistaPro, including manual UMLH tests
Part 4: Minimizing unmet load hours through appropriate zoning
Part 5: Minimizing unmet load hours associated with zone loads and system sizing
Part 6: Best practice simulation settings to minimize unmet load hours
Part 7: Unmet load hours in semi-conditioned spaces
Part 8: Minimizing UMLHs in complex buildings, such as healthcare and laboratories
Part 9: Avoiding UMLHs when multiple sources of heating and cooling serve the same zones
Appendix A: Specific UMLH Requirements of Codes and Standard Supported by the IESVE
The following appendix to Part 1 quotes and describes specific Unmet Load Hour requirements of codes and standards (and thus also indirectly green building rating systems, such as LEED) that are supported by the IES Virtual Environment software. This is not meant to be comprehensive, nor should it be used as a substitute for the complete language provided within the references codes & standards. Rather, it’s meant to highlight requirements and differences in the context of the discussion of UMLHs in Part 1, above.
ASHRAE Standard 90.1–2016 and 90.1–2019
An unmet load hour is defined as an hour in which one or more zones is outside of the thermostat set point plus or minus one-half of the temperature control throttling range. See part 1, above, regarding the term “throttling range”. Any hour with one or more zones with an unmet cooling load or unmet heating load is defined as an unmet load hour. There are further specific rules for ECB vs. PRM, as described below.
Energy Cost Budget method (ECB):
Unmet load hours for the proposed design or baseline building designs shall not exceed 300 hours. The unmet load hours for the proposed design shall not exceed the unmet load hours for the budget building design. Unmet load hours exceeding these limits may be approved by the building official, provided that sufficient evidence is provided to demonstrate that the accuracy of the simulation is not significantly compromised by these unmet loads.
Appendix G Performance Rating Method (PRM):
Unmet load hours for the proposed design or baseline building design shall not exceed 300 (of the 8760 hours simulated). Unmet load hours exceeding these limits shall be permitted only by approval of the rating authority, provided that sufficient evidence is provided to demonstrate that the accuracy of the simulation is not significantly compromised by these unmet loads.
The total number of hours that space conditioning loads are not met must be reported. The number of unmet heating and cooling load hours in the Proposed Design must not exceed the number of unmet heating and cooling load hours in the Standard Reference Design.
For both the proposed and reference buildings, the number of hours during which the heating loads for each thermal block are not met shall not exceed 100 hours in a simulated year.
The number of hours during which the cooling loads for each thermal block of the proposed building are not met shall not differ by more than ±10% from the number of hours in a simulated year that the cooling loads of the reference building are not met.
Where the requirements above are not met, the capacities of the primary and secondary systems of the proposed or reference building shall be incrementally increased until those loads are met. As with the other codes & standards, the modeler/designer must address any other causes of unmet load hours in the proposed design model, such as inappropriate zoning, mismatched schedules, and erroneous inputs.
California Title 24 – 2019
The CA Title-24 Alternative Calculation Method (ACM) Reference Manual offers the most comprehensive and also most stringent definition of unmet load hours. Whereas much of this provides a useful reference for compliance with other codes & standards, some of the requirements are specific to CA Title-24. The following summarizes or paraphrases the key points of the ACM Manual section 2.4 Unmet Load Hours.
As with other codes & standards, the concept of unmet load hours applies to thermal zones but is summed for hours whenever any conditioned thermal zone in the building has unmet loads. For a thermal zone, this represents the number of hours during a year when the HVAC system serving the thermal zone is unable to maintain the setpoint temperatures for heating and/or cooling. A thermal zone is considered to have UMLH if the space is outside the throttling range for heating or cooling.
The throttling range is defined as the space temperature difference between no cooling and full cooling, or between no heating and full heating. For CA Title-24, the throttling range is fixed at 2°F for simulating both the standard design and proposed design. Furthermore, it is assumed that the cooling and heating setpoints are “centered” on the throttling range so that a cooling setpoint of 75°F results in an acceptable temperature band of 74°F to 76°F.
The test for UMLH shall not be applied to thermal zones that contain only the space types listed below:
No other HVAC zone may exceed 150 UMLH within the annual simulation.
The UMLH test applies only to periods when the HVAC system is scheduled to operate.
The CA Title-24 ACM Manual suggests that, if the problem is heating, the size of the boiler or furnace may need to be increased, and if it is cooling, the size of coils or chillers may need to be increased. And that, in some cases, adjusting zone airflow sizing may also solve an UMLH issue. These ‘supply-side’ remedies have their place, however, they may not address issues with inappropriate HVAC zoning, excessive or highly transient loads (e.g., resulting from a significant user-input error relating to internal gains, envelope properties, or glazing performance), incorrectly specified controls, excessive ventilation, mismatched schedules, etc. It is up to the designer to adjust equipment sizes and address other causes of unmet load hours in the proposed design.
When any given zone within the building has no cooling system, and does not meet Title-24 requirements to be classified as a “heating-only” zone, and therefore may have UMLHs as a result of intentionally having no mechanical cooling, Title-24 allows users to “add a phantom cooling system” to the Proposed building design. This is added as a minimum efficiency (minimally compliant) single-zone AC system serving each space that otherwise lacks cooling, and must be noted and explained to the code official in the compliance documentation. As the SZAC cooling equipment added to the compliance model will, by virtue of being added, have capacities that differ from the building plans, its cooling capacity must be represented as such and explained to the code official within the compliance documentation. For spaces conditioned by the “phantom” cooling equipment, this forces users to comply with other relevant aspects of the Title-24 energy code.