C&I Medium-Voltage Distribution Transformers Research Agenda July 13, 2000

ABSTRACT CEE and Energy Star introduced initiatives to help promote the market for high-efficiency commercial and industrial transformers in 1997 and 1998, respectively. Both initiatives adopted the National Electrical Manufacturers' Association (NEMAs) voluntary standard, TP-1, to define energy-efficient, low-voltage products. Thanks to a unique collaboration of efforts among utilities, state and federal governments, equipment manufacturers, regional energy-efficiency collaboratives and trade associations, there has been remarkable progress in promoting this market in a relatively short period of time.

Given the success of its efforts to promote low-voltage products, CEEs C&I distribution transformers working group is now considering stepping-up its efforts to promote medium-voltage products. While a significant portion of these products are bought and operated by utilities, the working group is particularly interested in those medium-voltage products bought by non-utilities. The working group is interested in exploring the C&I medium-voltage market as a possible opportunity for CEE members to promote to their customers through public-benefit programming. The working groups findings on medium-voltage products will likely be of interest to NEMA and to ENERGY STAR as well. At the moment, however, the working group lacks a clear understanding of the medium-voltage marketboth in terms of energy savings potential and market structure. This paper summarizes the working groups current understanding of the market and key questions for further research.

INTRODUCTION Dry-type distribution transformers, purchased by commercial and industrial customers to convert utility electric line voltages to voltages suitable for building equipment, offer a largely untapped opportunity for energy savings in buildings. To address these losses and encourage the purchase of more efficient transformers, NEMA published an efficiency standard (TP 1-1996). Transformers that meet TP 1 can cut a facilitys total electric bill and pay back in about three years, but, in the past, been difficult to find or were highly priced as special order items.

Several national and state activities, each relying on TP 1 as its platform, have emerged to address the lack of information in the marketplace. These initiatives include the Consortium for Energy Efficiencys (CEEs) Commercial and Industrial (C&I) Transformer Initiative and an ENERGYSTAR TM labeling program. Contributions to these efforts by manufacturers, electrical contractors, facilities managers, and others have led to a greater understanding of the distribution transformer market, particularly for low-voltage products. Recent market developments include:

* Minnesota revised its state building code, mandating TP-1 transformers, on July 20, 1999. * Massachusetts adopted TP-1 standard as the minimum efficiency guideline for all new construction projects as of January 1, 2000. * Wisconsin adopted TP-1 for all new state buildings March 2000. * Participation in Energy Star C&I Transformers has increased 5-fold in less than two years, including Cutler-Hammer, Square D, and Honeywell. * Participation in CEEs Initiative has climbed to 13 organizations, including NYSERDA, National Grid USA, and Pacific Gas and Electric.

So far, the Working Group has focused primarily on opportunities to promote dry-type, low-voltage transformers since C&I customers most commonly buy these products. Currently, ENERGY STAR does not label energy-efficient medium-voltage products. It is unclear what portion of the most medium-voltage market already meet the efficiency specifications set by NEMA TP 1. Therefore the working group is interested in working with NEMA and ENERGY STAR to better understand this market.

The first task at hand to understand the market for medium-voltage distribution transformersboth in terms of energy savings potential and market structure. The remainder of this paper summarizes the working groups current understanding of the market and key questions for further research.

MEDIUM-VOLTAGE TRANSFORMERS

What we know

Function. Medium-voltage transformers step line voltage down from utility line voltages to lower voltages, depending on the application. Medium-voltage dry-type transformers will convert higher voltages to 480-volt 3-phase power to service equipment such as large motors. Smaller low-voltage transformers, will in turn take this power and further reduce it to 208/120-volt service. Table 1 below shows the characteristics and typical applications for different distribution transformer technologies.

Table 1 - Typical Applications for Distribution Transformer Technologies Low-voltage Medium-voltage

Liquid- immersed

Very limited application Utility Market Generally outdoors Often built to purchasers specifications

Dry- Type C&I Market Typically indoors Commodity products C&I Market Often indoors, although may be located just outside of facility Often special order

Losses. Collectively distribution transformers account for an estimated 140 billion kWh in electricity losses annually, of which a significant portion, 22 percent, is attributable to medium-voltage transformers (see Figure 1 below). Compared to overall electricity sales of about 3,300 billion kWh per year, transformer losses represent about 1 percent of annual sales.

Figure 1 - 1995 Estimated Losses by Type of Transformers (Millions of kWh) Source: Barnes et. al. 1997.

Sales. The markets for both utility and C&I distribution transformers are driven by continuous replacement of retiring transformers as well as new construction and major renovation activity. Total annual medium- and low-voltage transformer sales were collected as part of an Oak Ridge National Laboratory. In 1995, according to this study, total dry-type transformer sales were estimated by capacity to be 20,660 MVA (megaVolt-Ampere) (Barnes et. al. 1997). Discussions with major manufacturers suggest that this sales level has increased in recent years due to the strong economy and high level of new construction. Approximately 77 percent of distribution transformer sales by capacity is devoted to the utility market, while 22 percent of sales is devoted to the non-utility, or commercial and industrial market. Of these non-utility sales, about 60% of the dry-type market is low-voltage equipment, with the other 40% medium voltage (see Figure 2). Given that medium-voltage transformers represent less than a 10% of total transformer sales by capacity, the fact that their losses contribute nearly 25% of the total losses is significant.

Figure 2 - 1995 Distribution Transformer Sales by Type (in MVA) Source: Barnes et. al. 1997

Additional insights on the medium-voltage distribution transformer market include the following:

* Utilities and commercial and industrial users purchase more than one million new distribution transformers annually. * Commercial and industrial customers purchase virtually all dry-type transformers. * Medium-voltage dry-type transformers are generally special-order items, because of their high first cost and the need to specify the exact model and characteristics desired. Some purchasers, possibly as many as 60% of C&I users, consider the cost of operation in their purchasing decisions. (Barnes et. al. 1996) * There are a relatively large number of dry-type transformer manufacturers, though three to four manufacturers (including GE, Square D, and Cutler-Hammer) control 65 to 80% of the market. Many of the smaller manufacturers have specialty lines of equipment (addressing harmonic loads or other specific needs), or have larger market shares in a particular geographic region of the country. * Liquid-immersed transformers tend to be medium-voltage and dry-type tend to be low-voltage.

Efficiency. NEMA standard TP-1 includes medium-voltage products. Similar to its standards for low-voltage products, TP-1 provides different efficiency levels for liquid-immersed and dry-type medium-voltage products. Because liquid-immersed cores are a more efficient technology, TP-1 efficiency levels for liquid immersed are higher than those for dry-type medium-voltage products. There is an important difference. Unlike low-voltage products, most of the market for medium-voltage meets TP-1. It is also important to keep in mind that utilities dominate the market for the more efficient liquid-immersed, medium-voltage transformers. See Figure 3 below.

Figure 3. NEMA TP-1: Medium-Voltage, Three Phase Requirements

Amorphous Metal Transformers. A rather recent development in the market has been the entrance of amorphous core technology into the C&I transformer market. Amorphous core technology can achieve energy savings that are significantly greater than NEMA TP-1. Replacement of the nations existing 20 to 40 million distribution transformers could result in an estimated energy savings of 6 to 14 billion kWh annually (Kennedy, 5) If all new distribution transformers were amorphous metal transformers, this could result in energy savings of nearly 600 million kWh per year. If the estimated 40 million distribution transformers now in service in the US were replaced with amorphous metal units, over 40 billion kWh of lost energy could be saved (Kennedy, 136).

Traditionally utilities have been the key market for amorphous core transformers. In the US, distributed generation applications, especially for fuel cells and solar generation, are also a high-growth application for amorphous core transformers. (See attached chart). Over the last two or three years, there has been growing interest in transformer efficiency from C&I customers. In addition to marketing amorphous core distribution transformers to the C&I market through Honeywell, Allied Signal is talking with other transformer manufacturers to see if they would like to license the technology for their own product lines. (Chris Preuss, Allied Signal)

What We Dont Know

The information above leaves the Working Group with a lot of unanswered questions. Specifically, the Working Group would like to know whether a market transformation opportunity exists in the medium-voltage, C&I distribution transformer market. The first step requires developing the research questions that are needed to make this determination. Below are a list of questions and issues that constitute the working groups research agenda for medium-voltage distribution transformers for the C&I Market.

1. Are there two separate markets for medium voltage dry and medium voltage liquid? (need industry data other than from ORNL). If they are separate markets, why?

2. What are the codes that influence medium voltage dry type transformer purchases, and what other factors are driving the market?

3. What is the opportunity in the dry type market?

4. Should the ENERGY STAR MOU be revisited if C&I users are exhibiting an increasing trend toward purchases of LI transformers?

5. What is the range in efficiencies available from manufacturers for dry type (best/worse case)? (need a confirmation outside of ORNL data)

6. What is the breakdown of medium voltage dry type (and less importantly liquid immersed) purchases by utility and non-utility consumers?

7. How highly are medium-voltage dry type products loaded?

8. Does the typical load vary for dry vs. liquid?

9. Are dry type medium-voltage commodity or custom products?

10. What is the typical purchasing behavior of C&I customers like?

11. To what extent is lifecycle costing used for larger transformer purchases?

12. What is the difference between dry-type and liquid immersed medium voltage transformer efficiency? Does this difference exist and how big is it?

13. The savings potential estimated by ORNL (97 Supplement, Table 4.8, p.4-14) shows relatively small relative savings from non-utility liquid medium voltage (215 million kWh) vs Dry medium voltage (1041 million kWh); is this due to higher sales volume for dry, or efficiency gap?

14. How many units and total kVa of medium voltage, liquid are sold to non-utility purchasers? What are typical/representative non-utility sizes?

15. Could we get data like ORNL 97 Supplement Table 5-4 for non-utility medium voltage liquid and dry?

16. Is there any data available on typical/average efficiency levels for non-utility medium voltage liquid?

17. What are typical efficiencies (core and winding losses) for dry type and what is the range available from manufacturers (best case/worst case)? We need this less, since ORNL collected this (see Table 5.8) although confirmation and current data would be helpful. This table provides manufacturer survey data on load losses and load losses for non evaluated transformers (e.g., the worst efficiency case) and transformers designed to meet TP-1. Hence with accurate loadings -- which to date we don't have, we could calculate efficiencies.

18. Who are the top 5 manufacturers of non-utility medium voltage liquid, or is this just designer/customer preference?

19. Does the development of new technologies hitting the market, such as amorphous core transformers offer an opportunity for Energy Star to target efficiencies higher than TP-1?

20. What is the opportunity in the DT market -- starting with real measured loadings (in comparison with the F0% used in NEMA TP1. Here we REALLY need the data.

21. On savings potential, both sales volume and efficiency gap contribute to the gap in savings potential between DT and LI MV transformers sold to non utility customers. All of the data to support this statement is in the ORNL report.

NEXT STEPS After the issues above have been addressed to the satisfaction of the working group, the next step will be to determine whether or not the resulting savings and market structure lend themselves to a targeted initiative. If the opportunity looks promising, the working group will need to consider avenues for action, including:

* Gathering information that confirms (or denies) the ORNL data which basically indicates that utilities purchase liquid-immersed transformers and the C&I sector purchases largely dry-type transformers. * Determining appropriate efficiency specifications for medium voltage products * Brief ENERGY STAR on the medium-voltage C&I market * Coordinate Working Group findings with NEMA as it revises TP-1. * Request CEEs Board to revise its C&I Transformers Initiative, if appropriate.

Figure 1. Estimated Losses by type of transformer (expanded version of Table 4.8, 97 ORNL Supplement

Non-utility Utility Total Non-eval Eval Total Non-eval Eval Total Liquid, MV, 1 phase 44 3 47 258 1040 1298 1345 Liquid, MV, 3 phase 177 38 215 61 326 387 602 Dry, LV, 1 phase 88 0 88 0 0 0 88 Dry, LV, 3 phase 1041 0 1041 0 0 0 1041 Dry, MV, 1 phase 25 0 25 0 0 0 25 Dry, MV, 3 phase 519 0 519 0 0 0 519 Total 1894 41 1935 319 1366 1685 3620 Utility Liquid, Medium Voltage 1685 Non-Utility Liquid, Medium Voltage 262 Non-Utility Dry, Medium Voltage 544 Non-Utility Dry, Low Voltage 1129

Figure 2. Estimated 1995 sales by type of transformer (expanded version of Table 5.4, 97 ORNL Supplement

Liquid, MV, 1 phase 45586 Liquid, MV, 3 phase 22565 Dry, LV, 1 phase 950 Dry, LV, 3 phase 11383 Dry, MV, 1 phase 208 Dry, MV, 3 phase 8118 Total 88810 Liquid, Medium Voltage (about 90% utility use) 68151 Dry, Medium Voltage (virtually all non-utility) 8326 Dry, Low Voltage (virtually all non-utility) 12333

Figure 3: Medium Voltage, 3 phase Requirements in TP1

KVA Liquid Dry % eff MVA sold % eff MVA sold 15 98.0 45 96.8 12 30 98.3 91 97.3 26 45 98.5 182 97.6 119 75 98.7 455 97.9 141 112.5 98.8 273 98.1 82 150 98.9 1182 98.2 116 225 99.0 910 98.4 126 300 99.0 2638 98.5 184 500 99.1 3320 98.7 274 750 99.2 2694 98.8 1050 1000 99.2 2155 98.9 1240 1500 99.3 3098 99.0 2431 2000 99.4 1212 99.0 969 2500 99.4 4310 99.1 1348 Total 22565 8118

-- Mahri Lowinger (mlowinger@ceeformt.org), August 02, 2000