Flexible, sustainable electrical systems

Current building codes and standards allow designers to build flexibility into the electrical system.
By Raul E. Valdes, PE, LC, LEED AP BD+C, MIES, JBA Consulting Engineers, Las Vegas November 21, 2014

This article has been peer-reviewed.Learning objectives:

  • Know the codes that define electrical distribution system design.
  • Understand flexibility in the compliance path for energy codes.
  • Learn about other design methods for sustainable designs.

Figure 1: Turnberry Place in condominium complex in Las Vegas that consists of four towers, each 38 stories tall, for a total of 736 individual units. Each unit has its own service and is metered individually. All graphics courtesy: JBA Consulting EngineersAlthough the word “sustainable” may invoke different modern-day connotations, the main target remains the same—to maximize resources. Specifically, the electrical systems that engineers design consume resources both during construction and throughout the life of the building. Although a large part of the designs are driven by certain sections of the codes, these codes also contain some flexibility that allows designers to use fewer resources. 

As an example, NFPA 70: National Electrical Code (NEC) Article 220 Part III “Feeder and Service Load Calculations” gives straightforward guidelines on how to calculate feeders and service sizes for commercial and residential projects. For engineers focused on commercial projects, Part IV “Optional Feeder and Service Calculations” is sometimes overlooked. This optional method gives the designer a different approach to calculating both the feeder/service to individual dwelling units and the feeder/service for multifamily dwellings. 

For example, consider a condominium with 40 identical dwelling units. Each dwelling unit is 1500 sq ft and includes a kitchen with an electric range. Each unit also has laundry facilities, with an electric dryer, as well as an electric HVAC split system and other general use appliances. Table 1 summarizes how the calculations will differ when using the Part III or Part IV method. 

Table 1: Load calculation method comparison between Part III and Part IV methods prescribed in NEC 220.

For single dwelling units, the Part III method permits the general lighting load, along with the small appliance and the laundry circuit loads to be taken at the general demand of 35% after the first 3,000 W (NEC 220.52), but only allows for a relatively large demand factor for large appliances like dryers and cooking equipment per NEC Tables 220.54 and 220.55, respectively. Additionally, for other connected appliances, it will allow a demand of 75% only if there are four or more appliances in the apartment unit. HVAC loads are usually taken at the largest coincidental load in the unit, in this case the compressor with the fan coil motor. Using Part III, our load for this unit is 135 amp, which consequently will make us select a 150 amp overcurrent protective device and the corresponding conductor. 

In contrast, the Part IV method allows us to combine the general lighting, the small equipment, the laundry, the connected appliances, dryer, and electric range into one sum and demand the first 10,000 W at 100% and the remaining load at 40%. When combined with the HVAC load, which is calculated similarly to the Part III method, the result is a new calculated demand of 108 amp. Although we could select a 110 amp protection size, most of us will use 125 amp for ease of mind. Either way, the NEC has provided flexible guidelines that allow engineers to save clients in conductor costs, especially when considering the entire 40 units. It’s necessary to clarify that Part IV Article 220.82(A) states that this method of calculation is only applicable to “a connected load served by a single 120/240 V or 208Y/120 V set of 3-wire service or feeder conductors with an ampacity of 100 or greater.” 

Furthermore, in the same code sections, the NEC gives guidelines for the service calculation for multifamily dwellings. Continuing with the same 40-unit multifamily dwelling example, Table 2 illustrates the differences between both calculations. 

Table 2: Load calculation method comparison between Part III and Part IV methods prescribed in NEC 220. Part IV calculation allows for a 28% factor due to the quantity of dwelling units (40). Part III calculation allows for a 26.5% factor applied to the clothes dryers per Table 220.54, and a 22% factor applied to electric ranges per Table 220.44.

When considering the entire 40-unit multifamily dwelling, Part III will allow the application of an additional 25% demand factor to anything over 120 kW of the sum of the general lighting, small equipment, and laundry circuit loads. Additionally, we can apply demand factors to the electric dryer and electric range loads per NEC Tables 220.54 and 220.55. After all factors are taken into account, Part III calculations leave us with almost 1700 amp in calculated demand for which would generally be a 2000 amp service with 6 sets of 500 kc-mil copper (Cu) conductors. 

In contrast, Part IV allows us to apply a demand factor per NEC Table 220.84, in this case 28%, to the entire load. This results in a calculated demand of 1047 amp in load, for which would generally be a 1200 amp service with 3 sets of 600 kc-mil (Cu) conductors. This calculation result gives us the opportunity to save resources on the service entrance equipment.

Remaining within the realm of dwelling units, NEC Article 220.61 allows the reduction in size of the neutral conductor feeding the unit, considering the maximum unbalanced load. Table 3 shows that the neutral conductor could be downsized to 76 amps, with the phase conductors being sized to 108 amp. Note that such reduction is not allowed for 4-wire, wye-connected, 3-phase systems, which is reiterated in by Article 310.15(B)(7)(4). Therefore, depending on the distribution system chosen for a project, a reduction of the neutral conductor may not be allowed.

Table 3: Prescribed method of calculating neutral loads per NEC 220.61 (B).

Further along in the Part IV calculations, the NEC also allows us to use optional methods for schools in Article 220.86 and new restaurants in Article 220.88. Because NEC Table 220.86 applies demand factors to the connected load on a per area basis, this option becomes relevant for schools that demand a high power density (kVA/sq ft), which usually is the case for small schools that lack central plants and opt for remote terminal units (RTUs) and electric heat throughout the building. Tables 4 and 5 show the impact such method can have. In this case, it’s a 20% reduction in the service size. The optional method for new restaurants is similarly prescriptive per NEC Table 220.88.

Table 4: Prescribed method of calculating loads for an example school building per NEC 220 Part III. Notice that NEC 220.12 has an exception allowing for lighting loads to be taken at the values indicated by energy codes (with certain indications). Table 5: Optional method of calculating loads for an example school building per NEC 220.86.

Therefore, we can see how the NEC does allow for flexibility in load calculations for certain types of buildings. This flexibility can help our clients save resources on the front end of the building’s lifecycle.

Probably the most sustainability-driving standards the various energy standards. Whether it’s the International Energy Conservation Code (IECC), ASHRAE Standard 90.1, or California’s Title 24, we observe generally more sustainable requirements in every subsequent release of the standard. But even these standards offer some flexibility to them, and knowledge of this flexibility may prove to be an effective tool when we are on the edge of compliance. 

Figure 2: A general view of the different compliance paths as described in the International Energy Conservation Code (IECC) and ASHRAE Standard 90.1. Refer to each standard for more detailed information. Both IECC and ASHRAE 90.1 offer options in the compliance path. Actually, in Section C401.2, the IECC allows for us to follow either the prescriptive requirements described in the IECC, the mandatory requirements plus the building energy cost method, or the requirements in ASHRAE 90.1. To reiterate, the IECC states that compliance to ASHRAE 90.1 is a valid compliance path for the IECC. Furthermore, ASHRAE 90.1 also has similar options for compliance paths involving complying with prescriptive and mandatory requirements or mandatory requirements and energy cost budgets. Moreover, the most recent IECC now includes a space-by-space method as an option when calculating lighting power densities. Figure 2 illustrates, in general, a decision tree showing the flexibility of compliance paths that the IECC permits. Although the decision tree does not show every option given in the standards, it does let us know that there are different ways we can demonstrate project compliance, especially regarding lighting power densities (LPDs). It is important to note that the maximum LPDs allowable are not necessarily the same between 90.1 and IECC, and one standard may provide better compliance results than the other.

It appears that, by far, the most travelled compliance path is the prescriptive + mandatory path. This road leads to the well-known LPD tables, but even these tables have some flexibility. Primarily, we can elect to perform the calculation by using either the building area method or the space-by-space method. Furthermore, depending on the standard and method chosen, you may be able to apply certain exemptions and allowances such as retail highlighting and decorative appearance lighting. Special exemptions are frequently applied for lighting in casinos, food preparation areas, dwelling units, and many other areas.

Finally, when considering design improvements through sustainable design measures, even the most recent sustainable design trends should be looked into, as they most likely have room for improvement. According to the article “Rethinking Daylighting” (Hans & Stanfield, 2014), a few years ago the general consensus of implementing daylighting into a classroom was to use tall windows with exterior light shelves, along with relatively higher ceilings, and direct/indirect suspended fixtures with daylight sensors. The taller ceilings increased the overall building height, and the suspended fixtures were less effective considering that the indirect portion of the light would have to reflect off the ceiling and then back to illuminate the work plane. 

This trend has evolved significantly. Now, we are starting to see the elimination of the light shelves, the lowering of the ceilings, and the use of LED recessed grid troffers with daylight sensors and controls. The direct light fixture in the lowered ceiling, although not as aesthetically pleasing as the direct/indirect suspended fixture from the higher ceiling, is more effective and, therefore, more sustainable.

In addition to the example above, there are other methods to lower energy and resources purely within the electrical design, without the involvement of other disciplines. First, although already mandated by energy codes and standards, using more efficient (lumens per Watt) lighting fixtures and adequate controls is one of the most significant ways to reduce energy costs. LEDs and high-intensity discharge lamps like ceramic metal halide (CMH) have proven to have the lumen output, efficiency, color temperature, and lamp sizes adequate enough for indoor use. The fluorescent tube also remains an acceptable lamp to use throughout, but it’s being further eclipsed by LEDs as time passes. 

There is still much to be said about controls. Even though the codes and standards mandate them, designers have a lot of latitude as to how complex a system they wish to design. As designers, we should strive for long-term sustainability, and the complexity of the system comes into play when thinking about the long-term implications. Reflecting on the example of the design of daylighting, initially the designers installed complex automatic daylight dimming systems in the classrooms. Often, these systems are not user friendly and require costly re-programming and re-commissioning by the manufacturer, neglecting to a degree the savings provided by the system. 

In response, the new trends suggest to keep the daylight sensors and the vacancy sensor, but install manual dimmers for the teachers, or the students, to control the lights. This obviously leaves room for human misuse, but training teachers and students is possible and even educational. Translating this approach to common areas like offices, closets, bathrooms, and others, the designer should consider stand-alone vacancy sensors tied directly to the fixtures prior to considering central lighting control systems. It may prove easier to train maintenance personnel on stand-alone sensors than to teach them the intricacies of programming central panels in methods unique to each control system manufacturer. 

Ultimately, we need to consider the maintenance/end-user personnel involved in the daily operations, as they may very well bypass the costly systems if they find them too burdensome. Unfortunately, such actions are not uncommon, and could produce higher building energy costs than if simple manual toggle switches were used in the first place. 

In our electrical systems, the 480 V to 120 Y/208 V transformers are also an area of interest when considering operational cost savings. Primarily, the electromagnetic fields produced within a transformer when energized generate heat and produce energy losses, even when not loaded. Throughout the years, developments in transformer construction have decreased these losses. Because our designs traditionally load transformers to about 30% capacity, focusing on reducing the core losses of a transformer will have a more significant impact. NEMA TP-1-2002 and CSA Publication C802.2-00 offer guidelines to meet such efficiency targets, but they do not establish parameters for performance under nonlinear loads, which is not very realistic with the amount of electronics in our systems today. Needless to say, the use of proven low core-loss transformers is a sustainable approach that can have a short payback time.

In the electronic world, we also must consider the effects of harmonics on our systems. Because electronic devices are nonlinear loads, they create both voltage and current distortions. These, in turn, create overheating of circuitry and equipment, which increases the building’s cooling costs and decreases equipment life. The solution in the past was to upsize the system, which meant doubling the size of neutral conductors and using k-rated transformers. This did not solve the problem; it merely alleviated the symptoms while using more copper and larger transformers. 

The more sustainable approach would be to implement a system-wide solution with harmonic mitigating transformers. The harmonic mitigating transformers use different winding techniques to shift the phase of harmful harmonic currents. This results in the vectors subtracting from each other rather than adding to each other as they did through traditional winding techniques. Because harmonic mitigating transformers can vary in phase shift and windings, a system-wide approach should be taken in analyzing which traditional transformers in your single line, if any, should be replaced with which type of harmonic mitigating transformer. This may require assistance from certain trusted manufacturers, but it will certainly save the owner major costs throughout the lifecycle of the building. Because the traditional low loading of transformers inherently mitigates the effects of harmonics in current systems, there is space to debate the sizing of transformers combined with harmonic mitigating techniques to create even more sustainable effects, but this is certainly out of this author’s realm of expertise at this time.  


Raul E. Valdes is associate director of electrical at JBA Consulting Engineers. His expertise is in lighting and power design of commercial buildings, and he sits on the Hospitality Facilities Committee for the Illuminating Engineering Society (IES).