Designing electrical systems for higher education
College and university campuses depend on reliable, readily changeable, and easily maintainable electrical system networks to fulfill their academic and research missions. Regardless of cause, power disturbances can compromise and even invalidate scientific investigations, as well as disrupt and inconvenience the routine classroom functions of an institution.
To design such systems, the electrical professional must weigh a series of diverse immediate and long-term functional requirements—in addition to safety codes and standards—to design a reliable and durable electrical system for the entire campus and its individual exceptional high-performing buildings. Indeed, thorough consideration of infrastructure, reliability, backup systems, metering, changeability, and maintainability illustrates the extent to which the complex design requirements of higher education facilities exceed code minimum guidelines.
Infrastructure and reliability
The electrical network infrastructure supplying the higher education campus must provide reliable and safe power to its individual components. To do so, the incoming electric utility service must be distributed in such a way as to facilitate restoration of power during an outage in a safe and expeditious fashion. When requesting utility services, facility owners must weigh various factors in selecting the most suitable electric service to bring into the campus. Solutions vary in terms of geography. Campuses at the heart of large cities can rely on the available utility network to serve their buildings directly, while a more remote campus may have to manage its electrical infrastructure and distribute its own services internally. For the latter, the challenge lies in selecting the proper utility services and determining how to most effectively distribute them.
To analyze the reliability of the electric utility service, which, according to IEEE, is the largest contributor to both the failure rate and the forced-hours downtime per year at the 480 V point of use, engineers should refer to IEEE Standard 493-2007: Recommended Practice for the Design of Reliable Industrial and Commercial Power Systems. This standard provides valuable examples that prove useful in determining the reliability of single- and dual-utility sources, and in comparing various campus distribution techniques. The examples given conclude that increased reliability is obtained with a dual utility source arranged in a primary selective configuration.
IEEE 493-2007 compares the reliability of a simple radial system (one source), a primary selective system with manual throw over (9 min switchover), and a primary selective system with automatic selective throw over (5 sec switchover) (see Table 1). With automatic throw-over equipment at the primary switches, the number of failures per year is reduced by a factor of 6. IEEE 493-2007, Section 184.108.40.206 states, “The use of automatic transfer equipment that could sense a failure of one 13.8-kV utility supply, and switch over to the second supply in less than 5 sec would give a 6-to-1 improvement in the failure rate at the 480 V point of use.”
Based on these data, university campuses should request two utility sources and establish a primary selective system with automatic throw-over equipment for increased reliability at the 480 V building point of use (see Figure 2).
The transformer in secondary distribution systems is a very reliable component (at a low failure rate λ of 0.0062) but possesses the second highest outage time after the utility company (at 132 hr, resulting in larger forced-hours downtime/yr, λr). This implies that while the transformer is quite reliable, means must be accounted for to deal with the long outage experienced to replace it where high overall system availability is required. A secondary selective system using double-ended substations allows additional protection for transformer failures or maintenance (see Figure 3).
However, not all buildings on campus may necessitate this additional level of reliability or costs associated with the double-ended substation concept. It then becomes a programmatic decision to select which buildings on campus are classified as “vital” in operability and the costs associated with the additional redundant components can be used to weigh its benefits, such as for computerized data centers and research buildings.
An alternative to the double-ended substation concept implemented at the recently completed Wisconsin Institutes for Discovery building at the University of Wisconsin is the sparing transformer system (see Figure 4). Having established that transformers are highly reliable and rarely fail, the outage time reduced by the double-ended substation may be replicated by installing a spare transformer, which is essentially interconnected initially with a main circuit breaker and several interlocked tie circuit breakers. A true double-ended redundant system implies that transformers are sized at 50% of their ratings. An advantage is that in this sparing scheme, each transformer carries only its own load (single-ended), and transformer kVA ratings can effectively match building load and would not need to rely on fan ratings for the transfer events. This sparing system also has the advantage of a reduced footprint compared to double-ended substations.
At a Midwestern university, a primary selective system composed of a main utility line, with a secondary reserve line, is distributed to buildings throughout the campus, creating a loop system (see Figure 5). The sectionalizing switches are used to create the main outer loop that is open at one location in campus to allow for the utility service to come from two directions. The sectionalizing switches are used to isolate any feeder faults at this level. These switches are then used to create a second inner loop, which interconnects sections of campus in terms of geography via pad-mounted transformers with integral oil-immersed sectionalizing loop switches and draw-out current-limiting fuses. This system allows for any section of the loop system to be isolated, and also for the removal of any local transformer from the loop without disruption to other buildings. Figure 5 shows radial secondary systems at each building (single-ended), but increased reliability can be added by introducing double-ended equipment at the “vital” facilities on campus. This system is very common and adequate for a large university campus.
Reasons to provide backup power systems at university buildings stem from various code requirements for emergency and standby systems, as well as programmatic requirements for optional user-specified sensitive equipment. Generator systems, storage battery systems, and UPS systems are typical choices for universities to fulfill these special power requirements. The need for a generator system may arise from the total allocated capacity of the emergency, standby, and optional standby loads. Depending on the building type, the emergency loads are typically the smaller load, where the standby loads constitute the larger balance and may tip the scale toward the generator selection. Also, standby loads include a large component of motor loads, better handled by a generator system versus battery storage systems. When a generator is then selected, its capacity selection becomes a controversial topic in terms of determining what other optional standby loads will need to be served. Generators are mostly suited for larger buildings where the standby and optional loads add up, such as high-rise buildings and buildings with high-tech requirements, such as data centers. Smaller buildings with minimum emergency requirements can rely on battery systems (centralized or unitized) for the backup source.
UPS systems are frequently employed in data centers and information technology (IT) equipment, and universities may need to determine which UPS system configuration is the most appropriate to install. For example, for the distributed technology equipment located in a typical floor of a facility, a choice can be made between individual rack-mounted UPS systems versus a centralized UPS system distributed to every IT closet. A centralized UPS system, with a wrap-around maintenance bypass component, may reduce maintenance calls, and may prove cost effective in a larger facility. Also, some researchers may request uninterruptible power for sensitive experiments, which similarly can be dealt with by a point-of-use UPS or a central system.
Another typical approach is to handle the optional standby loads as a separate system that will seldom be needed because, depending on the campus infrastructure, most outage times are brief in duration, but may need to be planned for as catastrophic events. In disaster planning, creating a connection point for a portable generating system may prove prudent, investing only in the initial infrastructure to have the ability to provide service to optional systems, such as refrigerators housing sensitive research samples. For these portable units, universities would typically have a rental agreement with a generator supplier and upon the warning of an approaching storm, for example, they could install the portable unit as a safety precaution. It is also common to provide this type of portable system provision in a central university building (i.e., Student Union), which can become a campus shelter for disaster planning and have power for normal business operations during the event (food preparation, student services). With this initial pre-investment, provisions also can be made to install a future permanent generator at this location.
Electrical metering requests are becoming ubiquitous at university campuses, stemming from the need to monitor energy usage at individual buildings for billing purposes, LEED requirements, and additional sub-metering requests from individual university departments for energy studies or sub-billing. For buildings pursuing LEED certification, the process associates three credit points within the Energy and Atmosphere category, under Measurement and Verification. Electrical meters are required to monitor total building energy consumption, as well as submetering of various process categories, such as lighting loads, equipment, and plug loads.
A measurement and verification plan—even if not pursuing LEED—is a valuable tool in managing energy savings in a facility. These meters can provide information to continuously monitor the energy use of the facility and improve on the performance over the life of the building. Furthermore, they become an important tool in planning for future growth on campus. A university implementing a metering strategy will have average demand load data that can be analyzed per building type. These real data can be used to predict the energy usage of future installations and assist in planning for future expansion, and can be an invaluable tool for right-sizing equipment. Benchmark values derived from these data (Watts/sq ft) greatly assist the designers of future buildings.
A research building in a higher education campus will undergo a series of physical renovations during the life of the structure as the building has a substantially longer expected design life than many of its individual building components. The life expectancy of most electrical distribution equipment ranges from 35-40 years, indicating that several renovations will be required over the course of the 50- or 100-year building’s existence—and that building should be designed to facilitate the equipment replacements and upgrades accordingly (see Figure 6).
Incorporating this basis of design during the planning of the facility is paramount. To accomplish such a goal requires that electrical spaces are created with sufficient means for replacement of individual components with limited disruption to other operational equipment. Providing equipment removal paths as part of the design documents meets this purpose. For example, ensuring that a unit substation has enough room for its transformer to be removed, transporting via a clear path to an accessible exterior door sounds logical. But what if the unit substation is located at the lowest level of the facility, which happens to be below grade? Does the elevator have enough lifting capacity to handle the weight of a 2,500 kVA transformer core that weighs 20,000 lb? Such concerns play a major role in determining the replacement plan, not just for electrical elements but for all dynamic systems that will require reconfiguring during the life of the building (casework, laboratories, HVAC, and piping). The solution cannot be independent of other systems, as many opportunities arise by sharing equipment removal access among other trades.
Changeability also implies the ability for the facility to handle programmatic changes with prudent pre-investment. How can a university attract a researcher requesting high power into a building not planned for such high power research initially? Changeability features should be intuitive and transparent, meaning the facility owners must be aware of these elements. Prudent pre-investment may be allocating space for installation of future equipment, in lieu of procuring the equipment during the initial construction phase. For example, the main electrical service entrance room may include physical floor space for future substations, if a large electrical load is anticipated in the future, perhaps with monetary (grant) funding. Spare medium-voltage switches may be procured initially to serve this future equipment as a means of reducing facility downtime during the future renovation.
The 300,000-sq-ft Wisconsin Institutes for Discovery interdisciplinary research building on the University of Wisconsin campus in Madison was envisioned as a new direction in research facility design, flexible and sustainable (a 100-year building) while functionally demanding (see Figure 7). Among the various changeability features implemented at Discovery was to provide enough spare circuit breakers at the electrical panelboards initially to satisfy future space modifications. Adding an electrical circuit is ranked at the top of the service request when remodeling a space—at least a yearly event. A common design practice is to include 25% spare breakers in the initial design budget, 5% for changes that may occur during construction, and 20% for future modifications. Another implemented pre-investment was to provide stubbed-out empty conduits from the panelboard to the nearest accessible ceiling space. The quantity of spare conduits should provide capacity to handle the conductors needed to match the amount of spare breakers installed.
Other changeability features within a specific research lab or open classroom included the use of ceiling-mounted electrical devices. The concept of open spaces with ceiling-mounted utilities facilitates the future changeability of the space itself. Proper planning includes the standardization of receptacle types to be compatible with movable casework, as well as providing alternative voltages (i.e., 120 V, 208 V single- or 3-phase) at this space level. Another viable alternative is providing plug-in track busway where, for example, 120 V and 208 V, single- and 3-phase power can be derived with standard plug-in units via built-in drop cords.
As defined by IEEE, maintainability is "the ease with which a software system or component can be modified to correct faults, improve performance or other attributes, or adapt to a changed environment.” Maintenance is vital for the continuous and safe operation of higher education campuses. A well-planned electrical system will provide an up-front approach to maintainability by considering existing facility guidelines in conjunction with evolving codes and standards. Electrical working clearances (per National Electrical Code, Article 110) must be designed properly and maintained, taking into consideration that they are minimum code recommendations, and that they do not take into account the replacement means of the equipment. Standardization of components and in-stock spare parts can reduce downtime of operations considerably. For example, it is common for a draw-out switchgear design across a college campus to standardize circuit breaker frame sizes and maintain spares in stock that can be replaced among various facilities.
Equipment access remains a critical factor in maintaining a facility. During the design stage, consultants can facilitate this process with the BIM that has become a standard for sophisticated projects. BIM can visually present electrical systems in 3D software, and can include the equipment code clearances and access space as part of the illustrations. This tool proves invaluable in providing an avenue to ensure that the maintainability of the systems can be achieved. For example, cable tray access can be modeled—not just with the physical dimensions of the tray itself, but with the recommended continuous access space above of 12 in. and 24 in. on at least one side.
One must not forget about the experience that the building’s facilities and service personnel bring to the equation. Site standards, preferred vendors (offering familiarity and standardization), safety protocols (preferred arc flash categories and infrared scanning), lessons learned, and operational requirements are parts of a successful maintainability goal that these groups of individuals must ultimately execute. Managing these requirements can be quite challenging, as the initial capital investment must balance the pre-investment of future maintenance costs.
Providing today’s higher education clients with a reliable, flexible power infrastructure based on prudent first cost investments, managed with energy monitoring and optimized maintainability, and capable of supporting 100-year buildings is the unique challenge for the design professional delivering an electrical system. The extent to which the client’s expectations are met and all these elements of the campus are realized will be fundamental to the future success of the institution.
Cordero is a senior electrical engineer with Affiliated Engineers Inc. in Madison, Wis. He specializes in complex higher education, health care, and research facilities. His most recent school projects are the Wisconsin Institutes for Discovery at the University of Wisconsin (2012 Lab of the Year) and the Eckhardt Research Center at the University of Chicago.