|
||||||||||||
Power Supply Efficiency Standards
Voluntary power supply efficiency standards, like Energy Star (supported by EPA) or 80 Plus (supported by a consortium of electric utilities), exist in the U.S., Europe, Japan, China, and other countries and regions. The real issue is mandatory standards. In 2001, Executive Order 13221 established mandatory standards for standby power usage in Federal agencies.(1) More recently, Australia and California (with Arizona, New York, Oregon, Rhode Island, and Washington piggy-backing on the California approach) established mandatory standards covering certain power supplies at the Energy Star levels. Figures 1, 2, and 3 show the California standards and their effective dates.(2) Part I of this article compared measured power supply loads to these standards (and related standards under development for other power supply categories) and showed that savings of about 60% in standby and about 10 percentage points in active mode were realistic if all the products on the market met the standards.(3)
About 25-35% of single-voltage external power supplies on the market meet the standards in active mode, standby mode, or both. Overall, the issue of technical feasibility is rapidly shifting towards a limited set of specialized constraints. Comments by the Association of Home Appliance Manufacturers and the Consumer Electronics Association show that the main objections to mandatory standards relate to very low-wattage or low-voltage devices, infrequently used or charged devices, appliance battery chargers, compatibility across manufacturers, noise for very low standby loads, sourcing, and compliance schedules.(4,5,6) These objections focus on schedules, product cycles, and business operations practices, not technical feasibility for mainstream household electronics devices.
Internal power supplies and PC power supplies (internal or external) pose additional technical challenges, including measuring efficiency for power supplies integrated into the circuit board or across multiple voltage buses and multiple operating modes. Formal Energy Star PC and internal power supply standards are scheduled for 2007. Industry opposition is likely to be the same as for mainstream external power supplies – a lot of concern over measurement, schedules, product cycles, and business practices and far fewer questions about broad technical feasibility.
Details on technical feasibility and standards development can be found at many sites including Energy Star,(7,8,9) 80 Plus,(10) NRDC,(11) the Consumer Electronics Association,(12) and Efficient Power Supplies.org.(13,14)
Commercial Feasibility
Manufacturers voluntarily use SMPS for two main reasons. The first reason is that heat, size, weight, or portability materially influence purchase decisions. Some PCs and cell phones and most servers fall into this category. Apple Computer and Hewlett-Packard are among the most-well known major American manufacturers promoting energy efficiency in their products(15,16) and are perhaps the only two major manufacturers supporting California’s mandatory standards. The second reason is that the rising cost of copper- and iron-intensive linear power supplies makes them more expensive than SMPS in some cases.
On the other hand, when consumers don’t care and the inefficient products are selling, inertia takes over. For example, among big-screen flat panel televisions, active mode loads often differ by 50-100% or more across otherwise equivalent models. Standby loads sometimes differ by more than a factor of ten.(17) In the category of 61-62” rear projection televisions, a highly efficient JVC model draws 3.4 watts in standby while a comparably rated Mitsubishi model draws 39.8 watts.(18) Among LCD televisions, a 20” model from Sharp and a 27” model from Envision operate at about the same power in standby but in active mode the Sharp model uses 57% less electricity per sq. inch of viewing area than the Envision model. Given the functional similarity of most big-screen televisions of the same type and size, especially in standby mode, the obvious conclusion is that energy efficiency is not a universal design requirement.
Among retailers, the opposition to mandatory standards includes complaints over the loss of choice, the loss of products, noise, pricing, sourcing, compatibility, product cycles, and regulatory lead times. While these issues have merit on the margin, the primary (though unstated) issue is that consumers don’t use efficiency as a purchase decision factor. It is no coincidence that almost all the power supplies carried by one of the most vocal retail opponents of mandatory standards (Radio Shack) are inexpensive, inefficient, and non-compliant LPS.(19)
Despite a high degree of manufacturer and retailer inertia, there is a growing market for SMPS among certain manufacturers and in certain product categories. The industry beneficiaries of this demand include Power Integrations, the world’s leading supplier of high-efficiency SMPS integrated circuit chips.(20) Other major chip suppliers include Fairchild Semiconductor, On Semiconductor, STMicroelectronics, and Infineon Technologies. The leading suppliers of the complete SMPS include Delta Electronics and Sea Sonic.(21,22)
Cost and Cost-Effectiveness
The integrated circuit chip at the core of a high-efficiency SMPS typically costs around 50 cents. The remaining components, circuitry, packaging, etc. bring the wholesale cost of the complete power supply up to the range of a few dollars for small power supplies to $10 or more for large ones. For very high-volume purchasers of very low-power SMPS (e.g., a large cellphone OEM buying trickle chargers) the SMPS might be a dollar apiece or less. Relative to copper- and iron-based linear power supplies, the incremental costs for low-wattage power supplies are typically less than $1.00 and may be less than zero. For higher-wattage power supplies the incremental costs may be $5 or more. In some cases, especially where the market is small, the incremental costs can be tens of dollars.
A concern of some PC manufacturers, such as Dell, is that as the efficiency of BAU power supplies increases beyond the 70% range in active mode, the incremental advantage of going to 80% or more is not cost-effective.(23) Intel, for example, requires 70% efficiency at full load, 72% at typical load, and 65% at light load but recommends 77%, 80%, and 75%, respectively.(24) These efficiency standards are higher than those of only a few years ago and close to what is expected from the forthcoming Energy Star standards.
While there is disagreement over cost-effectiveness in some cases, the broad issue of cost-effectiveness is in little doubt. For example, Lawrence Berkeley National Laboratory (LBNL) concluded in 2002 that there was no evidence of any systematic difference in costs to consumers for power supplies meeting the standby standards.(25) To the extent there is uncertainty over cost-effectiveness, the primary issue is not the incremental cost of the devices but the level of BAU efficiency. A study by TIAX for the Consumer Electronics Association argued that analyses of the retail market in 2005 showed substantially lower standby loads than studies done only 3-5 years earlier and that the net cost-effective savings from further improvements in efficiency were uncertain.(26) On the other hand, another recent study by TIAX, this time for the Department of Energy, estimated that annual household information technology usage would increase from 42 billion kWh today to 53 billion kWh in 2010 in a business-as-usual environment but decrease to 31 billion kWh in 2010 in a high-efficiency environment (27) In other words, the opportunities for efficiency are enormous, only the market incentives and government actions are missing.
Additional details on costs and cost-effectiveness can be found at the summary level in the LBNL memo noted above(28) and in reports by Ecos Consulting / NRDC,(29) and the California Energy Commission.(30)
Political Feasibility
In the U.S., California has always been the national leader in mandatory appliance efficiency standards, of which power supplies are only one category.(31,32) Mandatory federal standards, which cover appliances like refrigerators and air conditioners but not power supplies, have always lagged the California standards in breadth, depth, and timeliness. The current situation is no exception. Because the California standards take effect starting in 2007, and because there are various phase-in timetables regarding existing and new products, it will be a couple of years before the full impact of the standards can be measured. If the standards are successful, it is likely that either most other states will follow, making the California standards de facto national standards, or the U.S. will adopt some version of the standards to simplify the regulations for manufacturers. This assumes, of course, that Congress doesn’t enact mandatory inefficiency standards instead.
In late 2005, the House Energy and Commerce Committee approved an amendment by Representative Mary Bono (R-CA) that would set meaningless efficiency standards for digital television adapters (DTAs) and, more importantly, prevent the individual States from imposing standards of their own. (DTAs will be necessary in July 2009 for all households using over-the-air analog reception because broadcasters will be required to switch to digital transmission.) The argument raised by the manufacturers and accepted by the committee on a party-line vote was that the California standards would price low-income households out of the DTA market. The committee rejected the argument that a DTA manufactured to California standards would cost a household an estimated $6 more at the time of purchase and save $42 in electricity over its estimated five-year operating life, paying for itself in less than one year.(33,34) While the Bono amendment only pertains to DTAs, it is always possible that Congress will override California and extend other Federal inefficiency standards to the entire country.
One Economist’s Perspective on Mandatory Standards
There are many general reasons why industry and Congressional objections to mandatory power supply standards at the proposed levels are incorrect and why California is acting in the public’s best interest.
First, it has long been recognized that voluntary standards are of little value. In fact, this is the reason for mandatory appliance efficiency standards in the first place. Consumers are myopic with respect to the time-value of money, ignore life-cycle energy costs, and focus almost exclusively on a few functional or appearance-related variables, all to their long-run financial detriment. Manufacturers respond to this myopia as expected. The notion that the market will deliver privately-efficient results (let alone publicly-efficient results) collapses in the face of implied discount rates for energy efficiency of 50-100% or more. These exceptionally high discount rates may make economic sense if the products have materially different non-energy characteristics, i.e., consumers may weight the non-energy characteristics more highly than the energy characteristics. This is the case with CAFE standards for vehicles. It is impossible to make this case where the energy-efficient products have no material non-energy deficiencies, as is the case with SMPS (unless reduced heat, size, and weight are considered deficiencies). There is a lengthy literature on the inability of consumers to make efficient choices, the implications of split incentives, the net costs to consumers, and the net benefits of government action so this issue need not be belabored further.(35,36,37)
Second, many studies purporting to find mandates cost-ineffective do not properly measure the benefits. These studies include benefits only against standards that are in place (e.g., standby) but give no credit for savings for which no standards yet exist (e.g., active mode). While manufacturers can claim, correctly, that they cannot actually make power supplies to standards that do not yet exist, a benefit-cost analysis must include estimates of savings in all modes. Since the same SMPS technology that delivers the standby savings delivers the active mode savings, both categories should be evaluated.
Third, the U.S. government spends billions of dollars annually researching, developing, and subsidizing electric power generating technologies (nuclear, clean coal, solar & wind, etc.) that, by definition, do not meet commercial investment requirements. At the same time, the U.S. government ignores (or opposes) legislative or regulatory actions that deliver returns of 50-100% or more on electrical efficiency. This makes absolutely no economic sense.
Fourth, economists are concerned with competition, not competitors. The issue isn’t whether compliance costs for the least-efficient competitors will make their products uncompetitive but whether any manufacturers can make compliant and competitive products. This is where politics takes center stage. For all their oratory about favoring free competition, most politicians gain very little from competition but a great deal from competitors – especially those who contribute to their campaigns or employ workers in their districts. There is a large literature on what economists call rent-seeking behavior as corporate and political self-interest trumps individual and public welfare. Periodically, however, even these activities become so blatant and so objectionable to taxpayers and voters that politicians are forced to act in a way that benefits the public. With respect to power supply efficiency standards, California recognized this after its electricity debacle five years ago. Five other states have agreed and many other states are watching closely. The U.S. Congress is, at best, missing in action or, in the case of the House Energy and Commerce Committee, favoring mandatory inefficiency to the detriment of the public.
Finally, governments must be concerned with external costs and benefits. High efficiency power supplies have few external costs but large external benefits, including reduced CO2 emissions, reduced demand for power plants, reduced transmission congestion, and reduced demand for fossil-fuel transportation infrastructure. For example, the EPA estimates U.S. average CO2 emissions at about 1.4 lbs/kWh.(38) At this rate, saving 47 billion kWh/yr. implies reducing 35 million tons of CO2 per year, or equivalent to more than 0.5% of total U.S. CO2 emissions. It is also equivalent to the power output of 7,000 MW of baseload generating capacity operating at 80% capacity factor and subject to typical transmission and distribution losses.
Manufacturers and, to a much lesser extent, retailers do have one very valid objection to mandatory standards. Manufacturers cannot continually redesign their products to meet new standards. Mandates must be certain and stable at commercially reasonable levels over a commercially reasonable period of time. Legislators and regulators cannot say, for example, that standards will be set at the equivalent of “best available control technology” or that the standards will be reviewed for upgrading every year or two. With legislative and regulatory certainty and with reasonable lead times, phase-in periods, exemptions for non-mainstream products, etc., then the establishment of mandatory power supply efficiency standards is a winning outcome for all parties. Indeed, compared to the inertial movement in a voluntary system, it is the only winning outcome for all parties.
Conclusion
U.S. households could annually save at least 47 billion kWh and $4 billion with efficient power supplies. There is a trend towards more efficient power supplies in some products, especially PCs, but there is a much larger trend towards very high-wattage consumer devices, especially big-screen, flat panel televisions, digital set-top boxes, video game consoles, and graphics-intensive PCs. The net effect of these two trends is that total consumer electronics loads and total avoidable inefficiency are rising rapidly.
High-efficiency switch-mode power supplies (SMPS) can dramatically increase the efficiency of these household electronics devices. Incremental costs are modest and sometimes absolutely lower than costs for inefficient linear power supplies (LPS). However, industry inertia and consumer myopia far outweigh all other factors and guarantee that the slow voluntary trend towards efficiency lags the rapid trend towards more devices per household and much more wattage per device. A single top-end home entertainment center (big-screen, flat panel television; video game console, and digital set-top box) now draws more watts in active mode and about as many watts in standby mode as all the electronics devices combined for the average household of only a few years ago. Under current voluntary efficiency standards there is little prospect of this changing anytime soon.
If avoidable energy inefficiency is to be minimized, it will be the result of mandatory efficiency standards enacted by California and thus far adopted in whole or in part by five other states. Unfortunately, there is no indication that the U.S. Congress has any interest in accelerating this process and, in one glaring counter-example, demonstrated its preference for retarding it. Assuming California’s actions are not broadly overturned by Congress, there is a chance that high-efficiency power supplies in some categories will become de facto national standards over time simply because major manufacturers will not find it practical to redesign their products only for the states with mandates. Beneficiaries will be consumers, taxpayers, the electric power system, and the environment. In the meantime, industry inertia, consumer myopia, and Congressional inaction are costing U.S. consumers billions of dollars per year in unnecessary electricity expense.
Finally, in considering the significance of power supply efficiency, note that the potential to save 47 billion kWh per year, $4 billion per year, 7,000 MW of baseload generation, and 0.5% of U.S. CO2 emissions is attributable to easily achievable efficiency gains in a grand total of one technology segment (electronic devices) in a grand total of one market segment (residential). Although not discussed in this article, there is a significant literature on the potential savings from easily achievable efficiency gains in other major technology segments (e.g., lighting, appliances, motors, and buildings) across all three major market segments (residential, commercial, industrial). In total, the savings dwarf those discussed in this article. What is missing, however, is the same: market-driven demands for efficiency and government-directed mandates for efficiency. Without a dramatic change in these factors, all the technological improvements in the world won’t make a difference in the long-run.
Endnotes:
1 www.ofee.gov/eo/eo13221.pdf
2 http://www.energy.ca.gov/2006publications/CEC-400-2006-002/CEC- 400-2006-002.PDF, Tables U-1, U-2, U-3. The standards starting July 1, 2006 on Table U-1 (Figure 1 in this article) have been delayed until January 1, 2007.
3 Energy Star standards for active mode efficiency for devices with internal power supplies and PCs are in development and expected in 2007. In terms of active mode usage, internal power supplies represent a much larger total load, notably coming from home entertainment and information technology devices.
4 http://www.energy.ca.gov/appliances/documents/2006-02- 03_AHAM.PDF
5 http://www.energy.ca.gov/appliances/documents/2006-02- 10_CEA_POSITION_SUMMARY.PDF
6 http://www.energy.ca.gov/appliances/documents/2006-02- 10_EPS_TECHNICAL_PRESENTATION.PDF
7 http://www.energystar.gov/index.cfm?fuseaction=find_a_product.showProductCategory&pcw_code=HEF
8 http://energyefficiency.jrc.cec.eu.int/pdf/Workshop_Nov.2004/PS% 20meeting/PS% 20CoC%20meeting%20101104%20Fanara%20External.pdf
9 http://www.energystar.gov/ia/partners/prod_development/revisions/ downloads/computer/Industry_Power_Supply_Slides.pdf
10 http://www.80plus.org/index.htm
11 http://www.nrdc.org/air/energy/appliance/app2.pdf
12 http://www.energy.ca.gov/appliances/documents/2006-02-10_EPS_TECHNICAL_PRESENTATION.PDF
13 http://www.efficientpowersupplies.org/
14 http://www.efficientpowersupplies.org/ips_workshop.html
15 http://www.apple.com/environment/design/energy/success.html
16 http://www.hp.com/hpinfo/globalcitizenship/environment/ productdesign/energyefficiency.html
17 http://reviews.cnet.com/4520-6475_7-6400401-3.html?tag=txt
18 Ibid.
19 http://www.energy.ca.gov/appliances/documents/2006-02-10-RADIOSHACK.PDF
20 http://www.powerint.com/
21 http://www.deltaww.com/company/about_profile.asp
22 http://www.globalsources.com/gsol/I/Switching-power-manufacturers/b/2000000003844/3000000154250/12199.htm
23 http://www.energystar.gov/ia/partners/prod_development/revisions/ downloads/computer/Industry_Power_Supply_Slides.pdf
24 http://www.silentpcreview.com/article28-page2.html
25 http://www.dlis.dla.mil/green/attributes/evaluationbenefits.asp
26 http://www.energy.ca.gov/appliances/documents/2006-02-10_TIAX.PDF
27 http://www.tiaxllc.com/reports/resid_info_tech_energy_consump_2006.pdf
28 http://www.dlis.dla.mil/green/attributes/evaluationbenefits.asp
29 http://www.aceee.org/conf/mt02/HorowitzWS44.pdf
30 http://www.energy.ca.gov/appliances/documents/index.html
31 http://www.powerint.com/PDFFiles/CEC-400-2005-012.pdf
32 http://www.energy.ca.gov/appliances/documents/index.html
33 http://www.aceee.org/press/0511dta.htm
34 http://www.aceee.org/buildings/resappl_type/dtatbl.pdf (Cost estimates rebased to one DTA for this article)
35 http://www.iea.org/textbase/papers/2005/efficiency_policies.pdf
36 http://enduse.lbl.gov/Projects/EfficiencyGap.html
37 http://www.aceee.org/buildings/policy_legis/stnds_info/cato.pdf
38 http://www.epa.gov/appdstar/pdf/brochure.pdf (page 8). The marginal rate is estimated at 1.2 lbs/kWh.
