How to keep smart lighting from being stupid
As smart lighting becomes more commonplace, it is important for engineers to understand the pitfalls and problems that may come with it
- Review the role of energy conservation codes in the adoption of smart lighting.
- Examine the functional requirements of smart lighting
- Explore factors that influence the cost of smart lighting.
Smart lighting insights
- Smart lighting technologies, while offering potential energy savings, require clear user understanding and proper training to prevent confusion and ensure effective use.
- Rising energy prices and the need to address climate change could make smart lighting more economically viable and contribute to grid capacity optimization.
The most efficient light is the one that is turned off when it is not needed. The least efficient light is the one what is still turned on when it is not needed.
This simple guiding principle for lighting control systems has been at the core of energy conservation codes since their inception. However, with the emergence of new lighting control technologies and their wildly expanded functionality, engineers often get lost in the minutia of the various available control solutions and lose sight of this guiding principle.
Frequently, the result is a control system that is not cost-effective, performs erratically and does not meet the client’s needs. In some cases, it may be that the controls did not perform as expected because there was not a clear understanding about what controls could and couldn’t do.
How the code evolves
Changes in technology have allowed for dramatic reductions in energy use, such as the replacement of incandescent and fluorescent light sources with LED. For example, 20 years ago, the 2003 International Energy Conservation Code (IECC) and ASHRAE Standard 90.1: Energy Standard for Buildings Except Low-Rise Residential Buildings mandated a maximum allowable lighting power density (LPD) of 1.3 watts (W) per square foot for commercial office buildings. Now, in the 2021 IECC, the LPD number is 0.64 W per square foot.
This represents almost a 51% reduction. Such reductions were easily achieved with the transition to LEDs. However, these types of improvements require revolutionary changes in technology and are usually the exception. Instead, incremental, marginal gains in lighting system efficiency and energy savings with each energy code revision are usually the norm. Without some other type of transformational change like the adoption of LEDs, the only way to achieve persistent incremental improvements is by increasing the controllability of lighting — to turn it off when it’s not needed.
Or, in the parlance of current design trends, to increase granularity of control (i.e., be able to adjust light levels to the minimum intensity needed and to perform that control only in the exact area where it is needed). The ideal solution would be to bring that controllability down to individual lighting fixtures – to have luminaire level lighting controls (LLLC). This LLLC concept where anyone can turn individual fixtures on or off was introduced as an additional efficiency package option in the 2015 IECC (C405.2 and C406.4).
What is smart lighting?
The term “smart lighting” is used and abused. Perform a search for smart lighting on the website of any big-box home improvement store, and there will be a multitude of products, usually with wireless control. However, does simply having that capability make something “smart”?
Beyond the singular characteristic of having wireless control, the functionality of those products can vary widely. Some are standalone devices, others are networked. Some have cloud-based controls while others rely on a direct Bluetooth connection to an iOS/Android app. Some allow for automation, others do not. Without a consistent definition to set clear, realistic expectations about the functional capabilities of smart lighting, confusion among the general public is unavoidable.
Absent standardization by the manufacturers, the driving force in dictating smart lighting functionality requirements, has been the prevailing energy conservation codes. While the term smart lighting is not used in IECC or ASHRAE 90.1, the IECC does have a basic framework for lighting control functionality that could be considered smart.
IECC lists the following definition for LLLC:
“A lighting system consisting of one or more luminaries with embedded lighting control logic, occupancy and ambient light sensors, wireless networking capabilities and local override switching capability where required.”
IECC Section C405.2 Lighting Controls and C406.4 Enhanced Digital Lighting Controls further expanded on this definition with specific functional requirements. These requirements essentially push the required granularity of control down to small groups, or individual light fixtures as in the case of LLLC. A notable omission from functional requirements listed for “Enhanced Digital Lighting Controls” is a lack of a requirement for wireless communication. Wireless capability is specifically called out in the description for LLLC. We will see that the incremental cost associated with adding LLLC functionality is a significant barrier to adoption and eliminating costly physical infrastructure, such as network cabling in lieu of wireless, is critical.
How does the industry define networked lighting controls and LLLC?
Beside industry wide standards from manufacturers, what is the driving force for developing consistent functionality standards for smart lighting? The answer is money.
The DesignLights Consortium (DLC) is an association of utility and regional energy efficiency organization through the U.S. and Canada. Its members are the same groups that control utility energy efficiency rebate programs across the country. One of its major goals is to create rigorous criteria that substantiates the inclusion of new lighting technologies in energy efficiency incentive programs.
Utility rebates are frequently used to help enable projects that otherwise would not have been financially viable. DLC’s requirements have successfully accelerated the adoption of LEDs by making compliance a condition of utility rebates. It is expected that this same implementation model will be applied to networked lighting controls/LLLC.
DLC has developed formal technical requirements for networked lighting control systems. Version 5 (NLC5) of these requirements was released in 2020 and updated in June 2023. The technical requirements include both “required” and “reported” capabilities. While reported capabilities are not currently required inclusions, their presence or absence needs to be documented by the manufacturer.
Many of these requirements have already been incorporated into the IECC. NLC5 also has some provisions for interoperability, but those are typically limited to communication with other systems beyond lighting.
Any mention of requirements for nonproprietary implementations of industry standard communication protocols (i.e., ZigBee, Bluetooth, DALI2, etc.) for individual control components are notably absent. This lack of standardization is perceived as being a significant barrier to the wider adoption of NLC5/LLLC technology.
The cost of energy and potential benefits of LLLC
In 2022, the U.S. Energy Information Administration (EIA) estimated that lighting represents 11% of total electricity use for the commercial building sector. While increasing light fixture efficiency and controllability can significantly reduce that, what is the value of that electricity? Unfortunately, the primary barrier to LLLC adoption is that the incremental cost associated with increased lighting control granularity is not fully offset by energy cost savings.
The first challenge is being able to quantify potential energy savings in a statistically defendable manner. The U.S. Department of Energy is required, per the Energy Conservation and Production Act of 1976, to make a determination as to whether the latest version of consensus-based building energy conservation standards will improve energy efficiency as compared to the previous edition. As part of this determination, an economic analysis to quantify the associated energy cost savings is also performed.
Nationally aggregated energy cost index (ECI) savings for the 2021 IECC code revision were estimated at 10.6% (reduction from $1.32 a square foot per year to $1.18 a square foot per year). Although not specifically quantified, it is understood that the ECI savings associated with lighting are only a fraction of that total. A key point was that the analysis excluded enhanced digital controls since it did not have quantifiable impact through energy modeling.
While studies have been performed by third parties, they have not yet been able to quantify LLLC energy savings relative to code baselines in a statistically significant way. A Northwest Energy Efficiency Alliance (NEEA) and DLC joint study attempted to quantify the potential energy savings associated specifically with LLLC. The study identified an average energy savings of 63% during normal business hours compared to a “do nothing” scenario, which does not meet code. As such, the quoted savings are not directly comparable to a building with a lighting control system that met the minimum requirements of one of the prevailing energy conservations codes.
Additionally, there were significant limitations in the NEEA/DLC study. The total savings were based on a sampling size of only 98 buildings with LLLC. Of the total energy savings, 37% was attributed to high-end trim, or the capability to reduce the maximum light output of a light fixture at the time of installation or commissioning, and had nothing to do with the adjustments associated with lights turning on-off and dimming during the day.
Beyond these issues, the primary downfall of this study was that the buildings included in this study were not a random sampling and the overall sample group was relatively small. While the average savings of the sample group may be valid, the savings from building to building varied widely. It is generally suggested that there are potentially significant energy savings, but a broader study with direct comparison to minimum code compliant buildings is needed to demonstrate defendable statistical trends.
The next question is: What is the material cost associated with LLLC? In 2021, NEEA published a study examining the incremental cost associated with LLLC. The study found that while costs were steadily dropping year over year, they were still substantial. The incremental cost of adding “clever” or “smart” functionality compared to the code minimum installation ranged from $29 to $70 per light fixture.
The 3/30/300 rule
As seen above, smart lighting is not a compelling investment, based on solely on energy cost savings. There are other, potentially more cost-effective energy efficiency solutions, such improving building envelope thermal performance. However, there is a real estate concept known as the 3/30/300 rule that may provide a more compelling reason for smart lighting adoption.
This concept states that there is an average order of magnitude between a company’s costs for utilities, rent and payroll per square foot per year:
$3 for utilities.
$30 for rent.
$300 for payroll.
While the exact values will vary, it is expected that the relationships and orders of magnitude difference between these three items are generally valid. Based on this relationship, an incremental percentage improvement in employee productivity can have an outsized impact compared to a similar percentage improvement in energy consumption.
Numerous studies have examined the impact of tunable lighting on circadian rhythms and educational outcomes. The consensus is that controllability is a positive influence on those outcomes. However, similar evidence-based research for commercial office environments is lacking — for now.
A trend toward higher energy prices could make LLLC viable
Why would a utility company actively encourage its customers to use less electricity if its revenue is tied to selling electricity? The simple answer is that the capacity that one customer does not use can be sold somewhere else. A rebate program makes sense if the utility company can buy that capacity through energy efficiency rebates at a lower cost than what would be required to purchase that capacity from somewhere else.
EIA studies have demonstrated that economic growth in the U.S. decoupled from energy usage long ago. EIA projected that between 2019 and 2050 gross domestic product will grow at an annual rate of 1.9%, but that energy consumption will have an annual average growth of 0.3%. The net result has been reduced investment in the grid.
It is also expected that carbon dioxide emissions will continue to increase, furthering the associated risks of climate change. To this end, the federal government has passed legislation that included decarbonization efforts. So, while overall energy usage growth across the entire U.S. economy may be limited, the fuel mix (coal, gas, solar, nuclear, wind, etc.) will change to emphasis electrification and utilizing renewable generation sources.
The change in the country’s fuel mix will have severe consequences. The intermittent and limited duration nature of wind and solar generation means that replacing thermal generation with renewables is not a one for one swap. Typically, multiple megawatts of renewables are needed to replace a single megawatt of traditional thermal generation. As existing thermal generation assets are retired, it is unclear if the new renewables will come online at a sufficient rate to make up this shortfall.
PJM, a regional electrical transmission organization responsible for coordinating the reliable transmission of electricity between generation companies and local utilities, issued a study that examined this shortfall issue, and the conclusions were not encouraging. While demand response and distributed energy storage are important tools in helping bridge the gap, additional resources are still needed.
Making this shortfall worse, the problem extends beyond the commercial and residential building market. The EIA typically separates energy usage into four broad categories: industrial, transportation, commercial and residential. While a significant percentage of the total energy consumption in the commercial and residential sectors already comes from electricity, that is not the case for industrial and transportation sectors. If those markets take significant steps toward electrification, the electrical capacity situation will worsen.
The solution to demand outpacing supply in a market-based economy is to raise prices. If the cost for electricity increases, some technologies such as smart lighting may become more economically viable. The logical extension of DLC’s efforts to develop NLC/LLLC guidelines, is standardizing functional requirements, so it will be easier to quantify the associated energy savings. While energy costs may increase, implementation of smart lighting may also result in energy savings for consumers and recovery of some valuable grid capacity for utility companies.
Examples of real-life problems with smart lighting
Lighting control concepts may seem natural to an engineer, but they often are not for the general population. Most engineers assume that building occupants generally want more control over their environment. However, that ability to have extensive controllability is often more confusing than enabling. As such, in a well-designed lighting control system, most functions are automatic, and the need for direct user interaction is kept to a minimum. Where interaction is required, clear and consistent identification of what the controls are supposed to do is critical.
In the first example, a U.S. Green Building Council LEED Gold corporate headquarters project was designed with lighting system sensors controlling 50% of plug loads within offices areas. Plug load control is commonly integrated into lighting control systems per 2021 IECC and ASHRAE 90.1. The associated receptacles were properly marked “controlled” in accordance with NFPA 70: National Electrical Code Article 406.3(E).
A short time after the client moved into their new offices, complaints started. Many of the executives rarely worked in the office. As they started to use their offices, they would plug laptops, phone chargers and other equipment requiring constant power into whatever receptacle was most readily accessible.
The key problems were:
The occupants had no previous experience with lighting-sensor controlled receptacles.
Occupants did not know what the markings meant.
The receptacles were usually located under desks and behind other obstructions where markings were not clearly visible.
This situation could have been made better by improved training and supplemental labeling or color coding to help the occupants to determine that these receptables were different from what they are used to.
In a second example, a networked wireless lighting control system for a commercial office space had been in service for multiple years. After what was assumed to be a power surge, the network controller rebooted unexpectedly. After the reboot, multiple wireless sensors started to function erratically and could not maintain consistent communication with the controller. The controller reported no errors after the reboot. In addition, some wireless occupancy sensors started to report low-battery conditions and it was unclear exactly which sensors were dying.
The key problems were:
“Control persistence” was not present. Control persistence is defined as the ability to execute three energy saving strategies (occupancy sensing, daylight harvesting and high-end trim) in the absence of communication with the next higher networked element in the system. This functionality is a reported capability of network lighting systems as defined by DLC’s NLC5 requirements.
The individual wireless devices did not have an obvious method, such as a different colored blinking light on the device, to identify which one had a low-battery condition.
The naming convention used to identify devices in the lighting control software interface was convoluted and did not necessarily reflect where the devices were located.
Lack of knowledge about how to interpret error messages and interact with the software made troubleshooting difficult.
The first two problems reflect possible limitations with the lighting control equipment itself. Hopefully as standards evolve, functional requirements will become more consistent across the industry. The second two problems are more indicative of insufficient training and interaction with the client during the system setup and commissioning process.
In the last example, a tenant noted issues with trying to turn lights on/off in its offices while in the process of moving in. The lighting control system was straightforward with a dedicated button assigned to each control zone. However, the contractors were rushing to complete their work before move-in and it was noted during the punch list that only some of the control stations were properly labeled.
Once the tenant realized that all the unlabeled buttons throughout their space had something to do with lights turning on and off, they resorted to button mashing to figure out what each button did. In addition, the contractor had programed a placeholder time-of-day schedule for certain common area lighting that was not approved by the tenant. As such, the lighting within the space would turn on or off at time that didn’t match their normal business hours.
The key problems were:
Installation, commissioning and training activities did not take place in a timely manner.
Labeling of controls, which would have made the intended functionality of the system more readily apparent to the occupants, was inconsistent leading to confusion.
As lighting control systems become more complicated with additional granularity of control, these last two problems will become a reoccurring theme.