When the U.S. Department of Energy Issued an Advance Notice of Proposed Rulemaking for higher-efficiency transformers in 2004, Exelon companies Commonwealth Edison Co. (Chicago, Illinois, U.S.) and PECO Energy (Philadelphia, Pennsylvania, U.S.) began closely monitoring these activities.

A typical 37.5-kVA single-phase overhead or pad-mounted transformer in use today at ComEd and PECO has an efficiency of 98.81%. The proposed DOE efficiency standard was 99.01%. While this only represents a 0.2% change in efficiency, it should be noted that between 1950 and 1990, significant technological improvements in materials and manufacturing methods only yielded improved efficiency of about 0.8%. This situation had the potential to have a serious impact on utilities across the country.

DISTRIBUTION TRANSFORMERS

Transformers convert one alternating current voltage to another and are essential components in electrical distribution systems. The primary distribution system (medium voltage) ranges from 4 kV to 35 kV. The secondary (low voltage) used by the customer is usually in the range of 120 V to 480 V. Distribution transformers are very efficient. Most distribution transformers convert at least 95% of input power into usable output power. The law of conservation of energy states that energy cannot be created or destroyed, but only transformed from one form to another. Some of the input energy is converted to heat during this voltage-transformation process. The sidebar on page 60 discusses the components of these losses.

In 1995, a report by the Oak Ridge National Laboratory (Oak Ridge, Tennessee, U.S.) estimated that there were approximately 50 million distribution transformers in the U.S. — 40 million are owned by electric utilities, and 10 million are owned commercial or industrial customers. Annually, utilities purchase about 1 million distribution transformers (about 2.5% of total). The report also stated that approximately 92.5% of the energy generated at power plants is delivered to consumers, 7.5% of the energy is losses in T&D systems, and about 26.6% of the T&D losses are attributed to distribution transformers.

ComEd has approximately 515,000 transformers and annually adds or replaces approximately 14,000 transformers (2.7% of total). PECO has approximately 170,000 transformers and annually adds or replaces approximately 2000 transformers (1.2% of total).

DOE FINAL RULE

Minimum-efficiency levels for liquid-immersed and dry-type distribution transformers are found in Table 1. These new efficiency standards were designed to provide some reduction in energy usage nationally but require an unknown increase in capital investment by the utilities to achieve it.

The following excerpt is from the DOE Distribution Transformers Energy Conservation Standards Final Rule, 10 CFR Part 431:

“By 2038, DOE expects the energy savings from the standards to eliminate the need for approximately six new 400-MW combined-cycle gas turbine power plants. The total energy savings from the standard will result in cumulative greenhouse-gas emission reductions of approximately 238 million tons of carbon dioxide from a variety of generation sources. This is an amount equal to what would be saved by removing 80% of all light vehicles from U.S. roads for one year.”

THE OBJECTIVE FOR COMED AND PECO

The new distribution transformer efficiency levels were anticipated to result in increased capital expenditures due to the need to use more copper and higher grades of core steel. The larger, heavier transformers could impact the cost and resources for replacements because of pole loading, foundation limitations, vault constraints, floor loading or elevator limitations. Transformer installation and replacement costs were estimated to increase approximately 15% to 25%. A utility's ability to recover these increased material and labor costs would typically come through the ratemaking process. The goal of ComEd and PECO was to meet the DOE efficiency requirements with minimal capital budget impact.

In order to manage transformer costs and avoid rate hikes, Engineering and Supply jointly decided to competitively bid the distribution transformer category. The transformer sourcing initiative began in early 2008 and focused on developing a long-term (three to five years) purchasing strategy and developing primary and secondary suppliers. The purpose of this initiative was to capture the efficiencies and economies of scale that come from strategic purchasing while, at the same time, maintaining (or improving) quality and lead times, and building stronger partnerships with key vendors.

PRE-BID HOMEWORK

Engineering and Supply recognized the financial impact and proactively applied their joint expertise to this business challenge in pursuit of innovative lower-cost solutions. To get started, a global supplier scan was conducted to identify low-cost countries for supply of distribution transformers. Both utilities currently conducted business with domestic and Canadian suppliers that were spread coast to coast. Once the global suppliers were identified, a kick-off meeting was conducted with interested vendors. Utility executives announced that a request for proposal (RFP) would be launched and answered vendor questions about the bid process.

ComEd and PECO conducted workshops with selected suppliers to discuss cost drivers and identify ways to reduce costs. The workshops were intended to be mutually beneficial. They were not planned as a forum to negotiate prices but to share ideas and discuss how working together with selected key suppliers could jointly reduce costs and improve productivity. A deeper understanding of cost drivers would eventually allow the Engineering and Supply team to perform fact-based negotiations focusing on costs rather than price. Workshop questions included:

• What modifications could the utilities make to transformer specifications in order to obtain reduced lead times and improve costs?

• Which parts of the utility transformer specifications added the most to costs?

• How would part substitutions impact performance, quality and maintenance requirements?

• What part substitutions could potentially reduce costs?

• Which transformer products are attractive to fill idle production capacity?

• What steps could both the utilities and suppliers take to jointly make improvements in the areas of forecasting, ordering, delivery, billing, payment and reporting that could help reduce cost drives and ultimately the price of transformers?

Manufacturers were asked to submit the size, weight and grade of core steel used in their transformer designs in order to compare the DOE-compliant designs to the non-compliant designs. The supplier workshops provided an opportunity to refine the RFP in order to achieve competitive pricing and extended warranties, and to realize industry-best service levels.

One cost-saving opportunity identified was to reduce the number of dual-ratio primary-voltage transformers purchased. Both ComEd and PECO use dual-ratio transformers that can be operated at either 4 kV or 12 kV. They are intended for installation in areas that will be converted from 4 kV to 12 kV. These dual-ratio transformers are more expensive than straight-ratio transformers because of the voltage selector switch and the extra labor required when winding the primary coil. The new transformer-efficiency standards require dual-ratio transformers meet efficiency standards at both of the primary-voltage settings. Introducing new 4-kV primary-voltage transformers and restricting usage of dual-ratio transformers strictly to 4-kV to 12-kV conversion projects could realize cost savings.

A common conclusion from the workshops was that providing demand visibility and contractual volume commitments could reduce both lead times and costs.

PRICE QUOTES

In order to complete requirements for the competitive bid, a market basket representing 80% of transformer spend was chosen for price quotation. This significantly reduced the number of transformer designs that would need to be evaluated and quoted by both the transformer manufacturers and the engineering teams. Three quotes were requested:

• Lowest total owning cost using current 2004 load/no-load loss factors

• Lowest cost to meet DOE (10 CFR Part 431)

• Lowest total owning cost using 2008 load/no-load loss factors with natural ester-based fluid.

Natural ester-based fluids, also known as vegetable seed oils, have improved biodegradability and fire points when compared to mineral oil. However, these fluids cost more per gallon.

Finally, a scoring matrix was developed to evaluate distribution transformer vendor responses. Suppliers were ranked on 27 key questions on topics, including: quality-control process, supply issues and risks, value adds (vendor-managed inventory, or make and hold), environmental initiatives and DOE compliance, diversity, core steel arrangements and customer base.

TOTAL OWNING COST

Historically, ComEd and PECO have used an economic loss evaluation to determine economic benefit. The value of future losses is added to the transformer purchase prices to obtain a life-cycle total owning/operating cost associated with a transformer. The last evaluation of losses for the distribution systems at ComEd and PECO was conducted in 2004. This evaluation indicated that it was favorable for ComEd to purchase less-expensive but higher-loss transformers.

Corporate investment evaluation updated the core losses and load losses for ComEd and PECO distribution transformers based on the spring 2008 forecasted energy prices. The new evaluation indicated it was now beneficial to purchase higher-efficiency distribution transformers. This change was largely driven by forecasted energy prices.

ComEd and PECO sold their generation plants to become wires-only companies. The utilities now deliver the power that others produce. Therefore, power is purchased on the open market, which is provided to customers at the purchase price. Load losses become more important because there is a high correlation between high prices and high loads.

LESSONS LEARNED

Evaluation of vendor bid responses indicated that there is negligible size and weight impact based on the use of higher-grade (thinner) core steel (typically varies from 7 mils to 14 mils (M2 to M6).

In select cases, manufactures were asked to revise designs to better optimize transformers sizes and weights. No one supplier or holding company was capable of effectively and economically producing all of the transformer designs and types required. The most cost-effective solution did not consolidate suppliers. However, the primary/secondary allocation strategy will reduce supply risks stemming from emergent issues from any single supplier.

Manufacturers indicated that transformer production capacity was not an issue as long as visibility was provided to demand. Designs per DOE specifications need to be completed by mid-2009, but production ramp up for DOE transformers was not an issue for the suppliers. The global transformer sourcing effort will reduce costs and permits adoption of the DOE-compliant designs early.

The dry-type transformers ComEd purchased for use in high-rise buildings were already meeting the DOE efficiency standards.

ComEd and PECO Engineering are currently working with distribution transformer manufacturers to develop a transition plan that will assure compliance by Jan. 1, 2010. Final transformer designs are being submitted to Engineering for approval. In order to meet the new DOE distribution transformer-efficiency standards, ComEd and PECO will begin placing orders for transformers in the third quarter of 2009.

---

Peter Tyschenko (peter.tyschenko@comed.com) is manager of distribution standards at Commonwealth Edison. He received a BSEE degree from the University of Illinois at Chicago in 1990. Joining ComEd after graduation, he has been involved in a field engineering, work management and reliability engineering activities.

Martin Rave (martin.rave@comed.com) is a principal engineer in distribution standards at Commonwealth Edison. He received a BSEE degree from the University of Illinois at Urbana-Champaign in 1991. Joining ComEd after graduation, he has been involved in distribution, substation and nuclear station engineering activities.

Giuseppe Termini (giuseppe.termini@peco-energy.com) is a senior engineer in distribution standards at PECO Energy. He received BSEE and MSEE degrees from Drexel University located in Philadelphia, Pennsylvania in 1982 and 1985, respectively. He joined PECO in 1982 and has been involved in distribution and nuclear engineering activities.

Richard Kahley (richard.kahley@exeloncorp.com) is category manager at Exelon Business Services Co. in Pennsylvania. He holds MBA and BSEE degrees from Drexel University in Philadelphia, Pennsylvania. Since joining PECO/Exelon in 1997, he has been involved in fossil power engineering and supply chain management.

Table 1. Efficiency Levels for Liquid-Immersed Distribution Transformers

Single-phase		Three-phase
Kilovolt- amp	Efficiency (%)	kVA	Efficiency (%)
10	98.62	15	98.36
15	98.76	30	98.62
25	98.91	45	98.76
37.5	99.01	75	98.91
50	99.08	112.5	99.01
75	99.17	150	99.08
100	99.23	225	99.17
167	99.25	300	99.23
250	99.32	500	99.25
333	99.36	750	99.32
500	99.42	1000	99.36
667	99.46	1500	99.42
833	99.49	2000	99.46
		2500	99.49

Table 2. Efficiency Levels for Dry-Type Distribution Transformers

Single-phase		Three-phase
20-kV to 45-kV efficiency (%)	46-kV to 95-kV efficiency (%)	20-kV to 45-kV efficiency (%)	46-kV to 95-kV efficiency (%)
98.1	97.86	97.50	97.18
98.33	98.12	97.90	97.63
98.49	98.30	98.10	97.86
98.60	98.42	98.33	98.12
98.73	98.57	98.49	98.30
98.82	98.67	98.60	98.42
98.96	98.83	98.73	98.57
99.07	98.95	98.82	98.67
99.14	99.03	98.96	98.83
99.22	99.12	99.07	98.95
99.27	99.18	99.14	99.03
99.31	99.23	99.22	99.12
		99.27	99.18
		99.31	99.23
BIL: basic insulation level. All efficiency values are at 50% of nameplate-rated load.

TRANSFORMER LOSSES

The two components that make up losses in distribution transformers are no-load losses (core losses) and load losses (winding losses).

• No-load losses are caused by the “magnetic resistance” or energy lost while magnetizing or energizing the steel core of the transformer. These losses are constant regardless of the transformer load and represent a continuous 24/7 cost for the life of the transformer. The drivers of no-load losses are hysteresis losses and eddy current losses.

  Hysteresis losses are caused by the alternating magnetic field that magnetizes the core. Each time the magnetic field is reversed, a small amount of energy is lost in the form of heat as the molecules in the core realign themselves with the field. Hysteresis losses can be reduced by changing the cross-sections or type of iron used in the core. Increasing the core cross-section will add to the weight and cost.

  Eddy current losses are caused by the alternating magnetic field that induces a voltage and current in the core. Each time the magnetic field is reversed, eddy currents circulate within the core and a small amount of energy is lost in the form of heat. Eddy current losses can be reduced by using thinner laminated core steel or by using a core material with an inherently high electrical resistance, such as an amorphous core steel. Both thinner core steel and amorphous core steel are less available and more costly.

  Core steel is also known as electrical steel, lamination steel, silicon steel or transformer steel. This special steel is produced with unique magnetic properties. Cold-rolled strips of grain-oriented silicon steel called laminations are stacked together to form a core. Differences among core steel grades are mainly due to final gauge (thickness). Core steel typically varies from 7 mils to 14 mils (M2 to M6) and is available in limited supply.

• Load losses are caused by “electrical resistance” or energy lost while passing current through the primary and secondary windings. Winding losses are caused by resistive heating of the conductors that make up the winding and vary with the load on the transformer. Load losses increase by the square of the current from no load to full load. Engineers often refer to these as I2R losses. When current flows through the windings, a small amount of energy is lost in the form of heat. The drivers of load losses are the winding material, copper or aluminum, and the winding cross-sectional area.

Winding losses can be reduced by increasing the diameter of the conductor or by changing the type of conductor used in the windings. Larger conductors will add to the weight and cost. Using copper versus aluminum windings also increases the weight and cost. Based on initial information, the new efficiency levels would have a significant impact on the cost and availability of raw materials used in the production of distribution transformers and will result in increased transformer production costs.

HISTORY OF EFFICIENCY RULEMAKING

The rulemaking process for distribution transformer efficiency can be traced back to the Energy Policy and Conservation Act (EPCA) of 1975, which established an energy-conservation program for major household appliances. The EPCA was later revised to include certain commercial equipment, including distribution transformers.

1975	The EPCA established an energy-conservation program for major household appliances.
1978	The National Energy Conservation Policy Act amended EPCA of 1975 to establish an energy-conservation program for certain industrial equipment.
1992	The Energy Policy Act amended EPCA to add certain commercial equipment, including distribution transformers.
2004	DOE issues Advanced Notice of Proposal Rulemaking for distribution transformer-efficiency standards.
2006	DOE issues Notice of Proposed Rulemaking.
2007	DOE issues Final Rule on Oct. 12, 2007, effective Jan. 1, 2010.