Electronics and Networks
By Bruce Nordman, Evan Mills, and Jessica Granderson
High-tech companies are defined by the intellectual output their employees create with their IT devices and other electronics. These include computers, displays, network hardware, small power supplies, imaging devices vending and beverage machines, among others. New ways are emerging to control, monitor, and power devices.
Depending on the definition, these loads account for on the order of 20% of total energy use in commercial buildings. The range of typical to best practice diversified power density ranges from 2 to 0.25 W/ft2 at peak, contributing on average about 0.75 kWh/ft2-y to overall building energy use (Wilkins and Hosni 2011). The presence of even small server rooms significantly increases these values. Because of the speed of technology development in this area, capabilities can transition from speculative to widely available in just a few years, introducing a fundamentally different set of opportunities and challenges. Because network-based technologies derive much of their value from computation and communication, many capabilities can be added or upgraded in the field via software updates. The ability to communicate through open and standard protocols is an essential foundation.
Applications
Many companies have policies to purchase only ENERGY STAR(R) compliant products. This is a needed foundation, but must be accompanied by processes to ensure compliance, address product types not addressed by Energy Star, and be attentive to product-specific opportunities to go beyond this relatively modest "top-25%" threshold, to procure best-in-class products and components (e.g., using the Top-10 system depicted in Figure 1). Attention should also be paid to products with low standby power requirements. In some cases, timers can be deployed to ensure no after-hours and weekend operations.
Figure 1. Top-ten identifies products at the very high end of the efficiency spectrum (http://www.toptenusa.org).
Electronics and other miscellaneous loads of course have direct impacts on space-conditioning energy, and thus understanding and managing them is important at the building level. Maximizing efficiency as well as diversity are additional important determinants of design loads. Rules of thumb tend to over-estimate demand (historically 3 to 5 W/ft2 in many cases) and lead to HVAC over-sizing and consequent excess capital and operating costs.
As electronics’ energy use is a product of both power levels and operating patterns, assuring optimal device operation is a critical strategy. In many buildings, most energy used by PCs, for example, occurs when no one is present or when the user is not performing computational tasks. While many devices have the ability to power down--from electronics that can go to sleep, to coffee machines that can turn themselves off, to dimming of computer displays, to vending or beverage coolers that can turn off lights when not needed --these abilities are often crudely configured or disabled entirely. Food- and beverage-related devices can be particularly energy-intensive and so should be a target for savings. The IT network is a potential, already existing mechanism to make visible such problems, distribute better operational policies, and track their results.
Advanced power strips are a technology that has been successfully applied to manage the energy use of electric plug loads in commercial buildings. Three modes of control are offered, alone or in combination: schedule-based timers, load sensing, and occupancy/vacancy sensing. Many offerings exist on the commercial market. In addition to power strip controls, these loads can be reduced through explicit target setting, workplace policies, and design choices in new construction projects (Lobato et al, 2011): computer monitors can be specified to use LED backlit LCD monitors can be used instead of fluorescent backlight or CRT monitors; personal copiers, printers and fax machines can be centralized into common multi-user stations; laptop computers can be used instead of desktops, for employees who do not need maximum computational power; the number of break rooms and kitchens can be optimized to serve maximum numbers of employees, reducing unneeded redundancy in appliances; elevators can be equipped with occupancy-controlled high-efficiency lighting and fans, and in some building designs hydraulic models can be replaced by regenerative traction models.
An emerging feature of networked devices is their ability to track their own energy use and report it to the local network (Nordman 2014). This is called "Energy Reporting" and can be used to understand and track energy use and key performance indices. Energy Reporting has been available for many years in some data center and telecommunications equipment, since these are critical services and highly managed. Standards are emerging, for example ANSI CEA 2047 defines Energy Reporting for appliances or any other device that chooses to use it. Electronic devices can use existing network interfaces to report energy data. While some devices may include hardware to measure power levels directly, many can reliably estimate their consumption. Building owners will be increasingly able to track energy use of each device in a building with as fine a time resolution as desired. The market is also seeing the emergence of integrated monitoring and control solutions; two examples are Budderfly (which includes lighting and thermostats) and Enmetric. An illustrative example of monitoring and reporting offered in the Enmetric platform is shown in Figure 2.
In addition to providing energy use data, this capability also automatically provides a detailed and dynamically updated inventory of the devices in buildings, and can show their operating pattern over the course of a day, week, or year (Nordman et al., 2014). This can reveal the existence of savings opportunities, provide quantitative evidence of how much potential savings is at stake, and show the presence of energy-using devices that are no longer needed. This new information can enable more energy saving measures to be implemented, and more rapidly than would otherwise be the case. Energy use data reporting can be aggregated across an entire campus, for greater visibility, while maintaining desired detail such as disaggregation by device type, time of day/week, or business function.
Figure 2. Screenshot of a plug load monitoring and reporting application (https://www.enmetric.com/platform)
Another opportunity is Direct DC powering (Garbesi et al., 2011). This takes advantage of the fact that all electronics and most miscellaneous devices (and many others such as fluorescent and LED lighting) are already DC internally. Commercial buildings have used DC powering for years for niche applications such as phones, Wi-Fi access points, and small USB devices. Industry groups such as the Emerge Alliance are working to develop standards to facilitate increased uptake of DC power distribution systems in commercial buildings. Direct DC can avoid hardware and losses from multiple AC/DC conversions, enable simpler integration of local generation and storage, and reduce installation and maintenance costs. DC devices can be integrated incrementally over time in addition to large-scale introduction when a building is undergoing a significant retrofit. Since DC power distribution usually includes communications, it is synergistic with the other technologies discussed in this section. DC datacenters have been demonstrated.
The wire for network and phone communications in buildings is already suited to being used for DC powering. However, the future wire and cable types best suited to DC are not yet certain, so for new buildings or significant renovations, building owners should ensure that new wiring or other infrastructure can be readily added at low expense. Any time that local generation or storage are added to buildings is a good opportunity to consider DC powering to take advantage of and be integrated with these systems.
DC powering can be used effectively within an individual building, or portions of a building. However, it may be desirable on a campus level to establish DC links among buildings to more efficiently share local generation and storage resources, and for local reliability. Vehicles, both corporate and personal, are another excellent candidate for DC integration as they are internally DC.
Economics
The energy savings from strategies such as advanced power strips depend on the space types in which they are implemented, the types of loads that they are deployed to control, and the specific control strategy adopted. Savings can reach 25-50%, with schedule based control solutions particularly cost effective when appropriately applied; retail prices are approximately $20, possible paybacks of two-years or less (Metzger et al., 2013). While energy management techniques such as monitoring and reporting do not directly produce savings, they enable insight into consumption and efficiency opportunities. These approaches are most cost effective when integrated into a holistic organizational approach to energy performance monitoring. The ROI of emerging networks-based communications and sensing that can be leveraged by other applications is not yet known, and general-purpose DC-power based applications are not yet mainstream.
Other considerations
All of these technologies have many non-energy benefits that by themselves may justify the investment, significantly outweighing the benefit of energy savings. In addition, some high-tech companies may be able to integrate the more emerging technologies discussed into their own hardware and software products. Energy Reporting capabilities can automatically create inventories of devices present in buildings, and track them at no additional cost. DC powering offers safety and resiliency benefits, from widespread use of voltages less than 60V and the ability to power select devices during utility grid outages at low cost. These data may also have value in augmenting existing security systems.
Institutional requirements & capacity
While many of the strategies discussed are readily commercially available, those on the more bleeding edge of the spectrum would require in-house staff engagement and direct involvement to define many of the technology details. This should not be a problem for technology companies, given that IT is their core business. For example, large enterprises have the option to custom-build PCs in order to ensure that high-efficiency components are specified. This can be particularly impactful where high-performance PCs are in use (Mills and Mills 2015). EBay has instituted just such a policy for their server arrays, maintaining a degree of choice while helping standardize maintenance and enhancing purchasing power (Schuetz et al., 2013). Corporate and building-based facility management and operational staff who are designated as responsible for these end uses, can receive periodic reports and address changes in devices or technologies as they arise.
Documentation
Garbesi, K., V. Vossos, A. Sanstad, and G. Burch. 2011. “Optimizing Energy Savings from Direct-DC in US Residential Buildings,” LBNL-5193E.
Lobato C., Pless, S., Sheppy, M., and Paul Torcellini. Reducing Plug and Process Loads for a Large Scale Low Energy Office Building: NREL’s Research Support Facility. National Renewable Energy Laboratory, February 2011. NREL Report Number NREL/CP-5500-49002 .
Metzger, I., Sheppy, M., and D. Cutler. Reducing Office Plug Loads Through Simple and Inexpensive Advanced Power Strips. National Renewable Energies Laboratory, July 2013. NREL Report Number NREL/CP-7A40-57730.
Mills, N. and E. Mills. 2015. "Taming the Energy Use of Gaming Computers." Energy Efficiency (in press).
Nordman, B., K. Christensen, R. Melfi, B. Rosenblum, and R. Viera. 2014. "Using Existing Network Infrastructure to Estimate Building Occupancy and Control Plugged-in Devices in User Workspaces," International Journal of Communication Networks and Distributed Systems, 12(1):4-29.
Nordman, B. "Energy Reporting". 2014. Report to Northwest Energy Efficiency Alliance, Lawrence Berkeley National Laboratory.
Schuetz, N., A. Kovaleva, and J. Koomey. 2013. "eBay, Inc.: A Study of Organizational Change Underlying Technical Infrastructure Optimization." Stanford University. 27pp.
Wilkins, C.K. and M.H. Hosni. 2011. "Plug Load Design Factors." ASHRAE Journal. May, pp 30-34. https://www.ashrae.org/File%20Library/docLib/eNewsletters/Wiilkins-052011--01242013feature.pdf.