This two-part article discusses in detail a simple, obvious, and consumer-friendly way for the government to make these savings a reality. The question is whether the political will exists to mandate a level of efficiency that will likely never arise voluntarily. Part I of this article discusses power supplies and power supply efficiency, total loads and avoidable losses, potential savings, and potential criticism of these analyses. Part II discusses power supply efficiency standards, technical feasibility, commercial feasibility, cost and cost-effectiveness, political feasibility, and concludes with an economist’s perspective on mandatory efficiency standards.

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.

Introduction

Household electronic devices require a power supply to convert 120 volt alternating current from the wall outlet to lower voltage direct current. External power supplies are the ubiquitous black cubes attaching PCs, printers, cordless phones, battery chargers, video game consoles, and so on to wall outlets. Internal power supplies are inside televisions, cable & satellite set-top boxes, microwave ovens, tabletop audio equipment, home security systems, and so on. The average household has at least 15-20 power supplies. The most important ones -- in televisions, set-top boxes, video game consoles, audio equipment, PCs, and printers -- are plugged in at all times.

The key phrase in that last sentence is “plugged in at all times.” In almost every case, a plugged in power supply is not “off,” rather it is in standby mode. Standby is what permits the use of remote controls on audio/video equipment, refreshes cable & satellite connections, and enables printers and televisions to warm-up in seconds rather than minutes. Studies done in the late 1990s and early 2000s at Lawrence Berkeley National Laboratories (LBNL) and other institutions showed that the average U.S. household draws 60-100 watts in standby.(1) While 60 watts per household is equivalent to a single incandescent light bulb, 60 watts of continuous load for an average of 20 standby hours per day aggregated over all the households in the U.S. exceeds 50 billion kWh/yr, or more than 3% of all household electric usage. This raises another obvious question: if more than 3% of all household electricity is used when electronic devices are “off,” how efficient are these devices when they are “on”?

Power Supplies and Power Supply Efficiency

A large percentage of the external power supplies in use and on the market and a smaller but still significant percentage of the internal ones are inexpensive but inefficient copper- and iron-based linear power supplies (LPS). Essentially all of these LPS could be replaced by high-efficiency switch-mode power supplies (SMPS) using advanced integrated circuit technology. High-efficiency SMPS can cut standby loads to less than one watt for most small- to medium-wattage devices (versus a rough average of 2-3 watts today) and to three watts or less for most high-wattage household devices versus a range of 5-50 watts today. Australia, which has the strictest mandatory standards in the world, estimates that a 1-watt standby limit would ultimately eliminate about 60% of standby loads versus business-as-usual (BAU).(2) Figure 1, a 2004 update of a 1999 LBNL figure, suggests similar potential in the U.S.(3,4) It demonstrates that standby loads for most devices vary by a factor of five or more. This variation implies enormous avoidable inefficiency since the functional differences among the devices in each category bear no relationship to the standby loads. The data also demonstrate that the most efficient power supplies in almost every category already achieve standby loads of less than one watt among smaller devices and three watts among larger devices.

Sixty percent savings in standby is only part of the story. The same SMPS substantially reduce active loads compared to LPS. The Environmental Protection Agency (EPA) has developed voluntary power supply efficiency standards as part of its Energy Star program. Figure 2 compares a scatter plot of single-voltage external power supplies up to 10 watts (typical of most small household devices) to the Energy Star standards.(5,6) The figure shows that 60-70 percent of the power supplies failed to meet the standards, many by several tens of percentage points. Overall, the average shortfall is around 10-15 percentage points versus the standards and around 15-25 percentage points versus the most efficient units. This demonstrates a very high level of inefficiency among products on the market. Perhaps more importantly, the figure shows that a commercially significant number of power supplies can meet the standards, proving that the standards are technically reasonable and achievable.

The results are similar for higher wattage power supplies. EPA’s tests of more than 600 power supplies up to about 160 watts showed that 36% met or exceeded the standards for active loads, 39% met or exceeded the standby standards, and 23% met both standards.(7) For the power supplies above 10 watts, the average shortfall is in the range of 5-10 percentage points versus the standards and 10-15 percentage points versus the most efficient units.

Measuring active mode efficiency in power supplies operating at multiple voltages (especially those for PCs) and most internal power supplies is more difficult but the results are similar. For example, Intel recently tested 188 power supplies and found that 95% met its required level of efficiency, 47% met its recommended level of efficiency, and 5% met its highest level of efficiency.(8,9) High-efficiency internal PC power supplies exceed 80% efficiency but the average on the market is only 65-70% efficient.(10)

Total Loads and Avoidable Losses

Household surveys from around the developed world show standby loads in the range of 40-80 watts.(11) Measurements from 10 California houses published in 2000 found standby loads ranging from 14 to 169 watts, with an average of 67 watts, or squarely in the middle of the world surveys.(12) In terms of usage, other surveys show that for the key loads (PCs, televisions, set-top boxes, etc.) a good estimate is 4 hours/day active and 20 hours/day standby.(13) If we assume standby in the U.S. averages 67 watts per household, 110 million households, and 20 standby hours/day, total household standby usage is about 54 billion kWh/yr. (110 million households * 67 watts/household * 20 hours/day * 365 days = 54 billion kWh/yr.).

In 1998, LBNL published a bottom-up estimate of about 43 billion kWh/yr. for what it termed standby loads for miscellaneous household devices, essentially all of which would be broadly defined today as consumer electronics.(14) LBNL also estimated that demand in this category would grow at a rate of 2.7% per year. Multiplying 43 billion kWh/yr. by 2.7%/yr. for the period from 1998 through 2006 produces an estimate of about 55 billion kWh/yr. today. This calculation almost certainly understates the actual demand, however, since growth forecasts made in the 1990s preceded the emergence of such devices as personal video recorders (e.g., Tivo) and high-definition tuners and preceded the boom in demand for big-screen flat panel televisions and digital set-top boxes.

In 2002, the Natural Resources Defense Council (NRDC) estimated power supply usage at 207 billion kWh/yr. and standby (plus the related “sleep” category) at 27%, or 56 billion kWh/yr.(15) If we assume that 15-20% of NRDC’s 56 billion kWh/year was commercial usage rather than household usage but that overall standby loads have increased 15% or so since the data reported in 2002 were collected, then we are back to standby loads of 50 billion kWh/yr or more.

These estimates are higher than estimates based on Energy Information Administration (EIA) reports. Using EIA household usage estimates from 2004 (1.3 trillion kWh),(16) the most recent EIA estimates for the home electronics percentage (7.2% in 2001),(17) and extrapolating to 2006, we would get total usage around 1.35 trillion kWh/yr., home electronics usage around 97 billion kWh/yr., and standby usage around 26 billion kWh/yr. This standby load implies a household average of only about 30 watts, less than half the measured average in the developed world.

These EIA-implied estimates understate actual loads for at least three reasons. First, they don’t include small appliances, power tools, security systems and many other miscellaneous devices. Second, they don’t fully reflect the rapid increase in information technology (IT) usage. A recent and detailed analysis of household IT usage by TIAX calculates loads 40% greater than EIA’s implied estimates, equating to total consumption in this category alone of 42 billion kWh/yr. and rising fast.(18) Third, EIA has not yet captured the transition in home entertainment devices, as CRT televisions have given way to much bigger flat panel televisions, small video game devices have given way to high-powered consoles, and analog and over-the-air television receivers have given way to high-powered digital set-top boxes, personal video recorders, high-definition tuners, etc.

Taking these factors into account, the present article conservatively estimates household standby loads at 50 billion kWh/yr. and active loads at 125 billion kWh/yr. The 175 billion kWh/yr. total implies household electronics usage at 13% of total household usage and rising fast.

Although not addressed in this article, the same type of analysis implies very large loads and avoidable losses for commercial office electronic devices (PCs, printers, copiers, fax machines, modems, etc.), many of which are identical to or higher-power versions of household devices.

Potential Savings

The literature on power supply efficiency suggests potential savings of 50-75% in standby, with most figures above 60%.(19,20) These savings would be achieved by taking the typical 2-3 watt loads for low-wattage devices to Energy Star standards of 0.5 – 0.75 watts and taking the 5-50 watt loads for higher-wattage devices to 3 watts or less.

Estimating savings in active mode is more complicated. The BAU mix in today’s market is not all inefficient linear power supplies. In the higher-wattage categories and among devices with design requirements relating to heat, weight, size, or portability, SMPS already dominate. At the lower-end, LPS occupy a very substantial share of the market. Overall, this article is estimating usage-weighted average efficiency of 65% for the BAU mix on the market today (which includes LPS and SMPS). This article estimates that a complete conversion to Energy Star levels (including levels for which the standards are under development, notably PCs and internal power supplies) would increase usage weighted-average efficiency by 10 percentage points. Figure 3 combines the estimated standby and active savings from complete adoption of high-efficiency SMPS versus the BAU devices now on the market. Ultimately, the savings reach 47 billion kWh/yr. and more than $4 billion per year from a 60% reduction in standby loads and a 10 percentage point increase in average active mode efficiency. To put these figures in a power generation context, 47 billion kWh per year exceeds the delivered output of 7,000 MW of baseload coal-fired or nuclear generating capacity operating at an 80% capacity factor and subject to average transmission and distribution losses.

Potential Criticism of Estimated Savings and Response

Critics of the analysis presented here will note that power supply efficiency is increasing as the market move towards SMPS on its own and therefore the present article overstates the potential savings of high-efficiency SMPS. This argument has some merit for certain information technology devices but very little for home entertainment devices. In the past decade or so there has been an exceptionally large increase in loads attributable to the changing composition of entertainment devices (e.g., big-screen flat panel televisions vs. smaller CRT televisions) and the quantity of installed devices. Efficiency gains have been swamped by the shift towards more devices per household and more wattage per device.

Even in the PC market, which is the major market segment moving most rapidly towards efficient power supplies, the absolute impact of new technologies is staggering. The Intel 386 processor used in the mid- to late-1980s operated at less than 2 watts. Today’s top-end Intel desktop processor operates up to 130 watts.(21) Intel estimates that savings from efficient PC power supplies alone could save 16 billion kWh/yr.(22) High-end video cards for PC gamers draw 50-75 watts, with cards in the 150 watt range on the way.(23) The Xbox 360 draws 231 watts, or almost 50% more than the original 160-watt Xbox and well beyond the typical desktop PC of 5-10 years ago.

In the entertainment segment of the market, inefficiency abounds. Standard CRT televisions of the 1990s drew 75-150 watts. Big-screen televisions today may draw three times as much. In fact, the difference in usage between the most efficient big-screen flat panel televisions today and the least efficient comparable models exceeds the total usage of the largest CRT televisions of only 10 years ago (i.e., more than 150 watts in active mode and 30-40 watts in standby mode).(24) A high-end home entertainment center comprising a cable or satellite digital set-top box, big-screen HDTV, and a video game console (e.g., Xbox 360) can draw more than 600 watts in active mode and more than 50 watts in standby mode. For comparison, 600 watts is in the range of radiant electric room heaters. In addition to the huge absolute loads of the newest top-end televisions, there is the load from displacement rather than replacement. Very often a new 300-watt flat panel television enters the house and the old 100-watt CRT television doesn’t leave but rather moves to another room and continues to draw standby power.

Part II

Part II discusses power supply efficiency standards, technical feasibility, commercial feasibility, cost and cost-effectiveness, political feasibility, and concludes with an economist’s perspective on the issues.

Part II will be published tomorrow on EnergyPulse.

Endnotes:

1 http://standby.lbl.gov/CEC_Workshop/Docs/CA_Energy_Use.pdf

2 http://www.energyrating.gov.au/library/pubs/2005-projectimpacts.pdf (Page 23)

3 http://www.kemco.or.kr/up_load/pds/meier%201%20watt-v2.ppt

4 http://standby.lbl.gov/Data/SummaryChart.html

5 http://energyefficiency.jrc.cec.eu.int/pdf/Workshop_Nov.2004/PS% 20meeting/PS%20CoC%20meeting%20101104%20Fanara%20External.pdf

6 Ibid (page 7)

7 Ibid (page 6)

8 http://www.efficientpowersupplies.org/pages/IPS_workshop_Jan06/ Internal_Power_Supply_Workshop_Jan2006_Summary.pdf (Page 8)

9 http://www.extremetech.com/article2/0,1558,1539846,00.asp

10 Efficient Power Supplies.org, Supra, page 9

11 http://www.kemco.or.kr/up_load/pds/meier%201%20watt-v2.ppt (Slide #5: Silicon Valley 40-200, California 67, USA 60, Australia 80, Japan 50, Europe 40, Urban China 35, Korea 57)

12 http://rael.berkeley.edu/rossmeiernaples.pdf

13 http://www.psnh.com/Household/ReduceBill/Applianceusage.asp (120 hours active per month = 4 hrs/day)

14 http://enduse.lbl.gov/info/LBNL-40295.pdf (page ii)

15 http://www.nrdc.org/air/energy/appliance/app2.pdf (page i)

16 http://www.eia.doe.gov/cneaf/electricity/epa/epa_sprdshts.html (1990 - 2004 Retail Sales of Electricity by State by Sector by Provider (EIA-861), lines 1263, 1387, 1510, and 1633, respectively.

17 http://www.eia.doe.gov/emeu/reps/enduse/er01_us.html

18 http://www.tiaxllc.com/reports/resid_info_tech_energy_consump_2006.pdf

19 http://rael.berkeley.edu/rossmeiernaples.pdf

20 http://www.standardsasap.org/candidate.pdf

21 http://www.silentpcreview.com/article308-page1.html

22 http://www.extremetech.com/article2/0,1558,1539846,00.asp

23 http://www.silentpcreview.com/article227-page2.html

24 http://reviews.cnet.com/4520-6475_7-6400401-3.html?tag=txt

To join in on the conversation or to subscribe or visit this site go to:  http://www.energypulse.net

Copyright 2005 CyberTech, Inc.