LBNL-41332

UC-1600

Bruce Nordman, Mary Ann Piette, Brian Pon, and Kristopher Kinney

Environmental Energy Technologies Division

Lawrence Berkeley National Laboratory

University of California

Berkeley, CA 94720

February 1998

This work was supported by the Assistant Secretary for Energy Efficiency and Renewable Energy, Office of Building Technology, State and Community Programs of the U.S. Department of Energy under contract No. DE-AC03-76SF00098, and the U.S. Environmental Protection Agency, Atmospheric Pollution Prevention Division.

In the U.S., copiers use about 7 TWh/year of electricity, and a similar amount of energy is embodied in the estimated 2.2 million tons/year of paper used in copiers. These cost the economy about $500 million/year for the electricity and $2.2 billion/year for paper. The U.S. EPA launched the Energy Star copier program in 1995 to save money plus reduce energy use and greenhouse gas emissions from copiers. This study evaluated the performance of Energy Star copiers to assess the energy savings they currently achieve and the potential for increasing the savings. The main effect of the program is not to change power used in each operating mode, but to change the amount of time spent in each mode. We defined methods for auditing and energy use monitoring copiers and carried them out on 228 and 11 machines respectively. About 30% of both Energy Star and conventional copiers were left on at night; while most conventional copiers were in a low-power mode, most Energy Star compliant machines were fully off. Extrapolating these findings to all U.S. copiers results in higher electricity use than previous estimates, due to the night and weekend status and longer work days. However, this also implies a greater potential for saving energy with power management from Energy Star copiers. A survey of users found general satisfaction with Energy Star compliant copiers. Enabling of default duplex on two copiers raised their duplexing rate by 15% and 20%.

We would like to thank the following individuals and organizations for their generous assistance in conducting this project: Bernard Aebischer (Swiss Federal Institute of Technology), Barbara Becker (Alameda County), Chris Byrne, Cyane Dandridge, Neil deSnoo and John (City of Berkeley), Jeff Harris, Jon Koomey, Steve Greenberg, Wolfgang Huber, Dave Huelbig (Ricoh Corp.), Satkartar Khalsa, Alan Meier, Delonzo Pope, Roger Picklum (City and County of San Francisco), Alison ten Cate, Ed Vine, and Ryan Wood (Bayview Technologies).

We would particularly like to thank the project sponsors for their support and guidance: from the U.S. Environmental Protection Agency, Andrew Fanara, Linda Latham, and Scott Thigpen, and from the U.S. Department of Energy, Anthony Balducci.

4.1 Audit Results 8

4.2 Electricity Field Monitoring 10

4.4 Paper Use 15

In the U.S., copiers use about 7 TWh/year of electricity, with a similar amount of energy use embodied in the 2.2 million tons/year of paper use. These cost the economy about $500 million/year for electricity and $2.2 billion/year for paper. The U.S. Environmental Protection Agency launched the Energy Star copier program in 1995 to reduce greenhouse gas emissions and energy use, as well as save money. The Energy Star criteria include requirements for copiers to automatically turn off or enter a low-power mode after a time of non-use, and sets specific limits on the power consumption in low-power and off modes. Default duplexing was initially required for some copiers and is now only recommended. This study evaluated the energy performance of Energy Star copiers. We defined methods for copier auditing (configuration and operating patterns) and monitoring (energy use) and carried them out on 228 and 11 machines respectively. We also created a user satisfaction survey, to interview 56 copier users, and studied the effect of making duplexing the default mode on two copiers.

Previous copier energy estimates assumed that copiers are rarely left on at night. However, we found about 30% of both Energy Star and conventional copiers on at full power at night. While most conventional copiers were in a low-power mode, most Energy Star compliant machines were off. Not surprisingly, larger copiers were used for more hours per day than were smaller copiers. From the audits and monitored data, we defined a set of standard operating patterns for different speed categories of copiers. The operating patterns were combined with measures of power use for 130 copiers to estimate average annual energy use for conventional and compliant copiers. Energy savings were considerably below their potential due to power management being disabled on over one third of Energy Star copiers audited.

With approximately one third of the stock of copiers now Energy Star compliant, we estimate the current savings of the Energy Star copier program to be 570 GWh/year. If the entire stock of copiers were compliant, the savings would be 1,700 GWh/year. The potential program savings–all copiers compliant and configured for high power management savings–are 4,200 GWh/year. These figures are higher than previous estimates, due to revised assumptions about the night and weekend operating status and longer work days. The savings are derived primarily from the auto-off feature. Low power modes were envisioned as accomplishing daytime savings, but their primary effect is reducing night and weekend energy on copiers with the auto-off feature disabled.

The user satisfaction survey indicated that most people are satisfied with the performance of their Energy Star copiers. The major complaints cited were about paper jams and duplexing speed. The machines were perceived as more reliable than others, though perhaps this was simply because they were newer. Unfortunately, Energy Star copiers are not faster at duplexing than conventional machines. The difference in speed remains a significant barrier to higher duplexing rates because duplex copying is slower than making simplex copies.

Default duplexing was found disabled on every copier we saw. However, when we enabled it on two machines, the duplexing rate rose by 15% and 20%. Thus, while the initial barrier to default duplexing is considerable (and manufacturers provide little guidance on how to implement it successfully), it remains a significant untapped opportunity for energy and cost savings and reduced greenhouse gas emissions. Survey respondents often cited confusing and hard to use controls as reasons for not using copier features such as duplexing. Copier efficiency could be increased with standardization of terms, symbols, and controls (both among and within brands). Manufacturers should provide information on these topics in operating manuals and on the web.

The greatest opportunity to improve on the current savings from Energy Star copiers is to increase power management enabling rates. Sales and service personnel are probably responsible for most disabling. Anecdotal evidence suggests that service personnel often disable power management before customers ever use the copier, and may completely disable it on an inquiry from the customer rather than explaining the benefits of Energy Star features or altering the configuration for the customers’ needs.

Office equipment is estimated to have consumed about 7% (Koomey et al., 1995) of the electricity used in commercial buildings in 1990. The Energy Star office equipment program was established in 1992 to reduce this demand to save money reduce greenhouse gas emissions. Copiers were added to the program in 1995, joining personal computers, monitors, printers, and fax machines. Copiers are estimated to currently use about 10% of office equipment electricity–about 7 TWh/year–and are expected to drop to about 5.7 TWh/year by 2000 with the effect of the program. The Energy Star program operates by engaging equipment manufacturers in a dialogue to identify energy-saving criteria that are broadly achievable but represent a significant improvement over existing equipment. Compliant equipment may be labeled with the Energy Star logo, and the EPA promotes the purchase of Energy Star products. A separate program for multi-function devices was launched in April 1997.

This study was initiated to evaluate the current energy savings and overall performance of Energy Star copiers, and to provide recommendations for program improvement. Components of the project included a review of the relevant literature, development of procedures for measuring copier performance, night-time auditing of copier status, detailed electricity monitoring, surveying of user satisfaction, and integration and analysis of results.

Table 1-1 shows the requirements for qualification as an Energy Star copier. The criteria specify how the copier is to be configured when shipped to the customer; after installation, it can be changed by the customer or service personnel if desired. Default duplexing was required for high speed machines (>44 copies per minute–cpm) for both Tiers until late 1997; since then it has been optional. As Table 1-1 shows, the main criteria for the program are delay timers that cause the machine to either enter a low-power state, or turn off. A notable feature of the copier program is that EPA expects no extra manufacturing cost for the Energy Star features, so that the customer saves money from the first day of operation. Six months after the program took effect, it was estimated to have captured a notable 33% of the copier market and 7% of the stock (Dataquest, 1996). As copier models have turned over, these percentages have probably risen sharply since then.

Notes: aThere are no low-power requirements for Tier 1 machines (only off-mode power requirements). bAn additional requirement for Tier 2 mid-range copiers is a maximum recovery time from low-power of 30 seconds. c"N.A." means "Not Applicable"–that no requirement exists. d3.85.cpm + 5 = 86 W for 21 cpm, 169 W for 44 cpm, and 390 W for 100 cpm. eCopiers use some energy even when switched off (such as for electronic controls). fThe date in force is the date of the model introduction, not the date of sale.

Several factors indicated a need for this study. As Energy Star copiers were largely new to the market (some compliant models preceded the program), their performance had not been compared to conventional copiers. Previous evaluations of the Energy Star PC and monitor program had found widespread disabling and other implementation problems that defeated power management (Nordman et al., 1996b). Anecdotal evidence of dissatisfaction with and disabling of default duplex indicated that there might be similar problems with the copier program. In addition, an evaluation of actual copiers and usage patterns would provide for close scrutiny of Energy Star program savings estimates. The components of this project included: a review of the existing literature on copier operating patterns, energy use, and imaging rates; the development of methods for field monitoring of copiers (electricity monitoring, audits of night-time status, and duplexing rates); conduct of the field monitoring; and an assessment of user satisfaction with copiers.

With the information developed in the project we: created standard operating patterns for conventional and Energy Star copiers; made better assessments of conventional copiers and current savings from the Energy Star copier program; identified opportunities to improve the program to increase customer satisfaction and energy savings; and made recommendations on program and copier design.

This paper covers the main findings of each part of the project; additional details and results are presented in the appendices (due to the size of the appendices, many are published only electronically). Section 2 reviews previous work in this area, highlighting those findings that influenced the course of the project. Section 3 describes our monitoring and auditing approach, including lessons applicable to other copier research. Section 4 covers the project results, including our audits, electricity monitoring, assessment of ASTM test results, and estimates of annual energy use per copier. Other results include our assessment of user satisfaction, measures of paper use, extrapolations to national estimates of electricity and paper use (and Energy Star copier program savings), and comparisons to similar estimates made by others. Section 5 summarizes key conclusions and research needs.

In this project, we did not attempt to gather statistically representative data for the entire stock of U.S. copiers (a huge task) but rather produced results that are indicative of copier performance and sufficient for program planning and evaluation purposes.

Copier electricity use is complex in its details, particularly within individual imaging events (Acquaviva, 1992). However, for the purposes of this study–evaluating power management and overall energy use–representing that complexity is not required. Power management features include auto-off and low-power ("energy-saver") modes, or both.

To provide a basis for comparing the electricity use of copiers, the American Society for Testing and Materials issued a test procedure in 1987, with a significant revision in 1994 (ASTM, 1994). The ASTM test is the only standard test in widespread use in the U.S. or Europe. The ASTM test was designed to provide performance data suitable for comparing the electricity use of copier models. This is accomplished by recording average power levels for one hour periods as well as a monthly total energy use based on a standard operating pattern. Because it was developed for comparison and not prediction, the operating pattern is not derived from typical use patterns and so the monthly (annual) energy use figures derived from the test should be interpreted with caution. Copiers are assumed to be left on only two nights a month (never for copiers with an auto-off feature) and never on weekends. The ASTM test provides a basis for standard terminology.

Figure 1-1 shows a prototypical weekday loadshape for three copiers–a conventional copier always on, a conventional copier with an average use pattern, and an Energy Star copier fully enabled. Active copying energy is not shown–it only adds about 10% to the electricity use and is not currently a target of power management. Any copier can be manually turned off at night and many conventional copiers have auto-off or low-power features. Energy Star copiers are more likely to have and utilize the features.

Several key factors affect copier energy use, and influence the savings from the use of more efficient copiers. For conventional copiers this includes the fraction of copiers that have power management features, how they are configured (e.g. the fraction enabled), and the number of hours during which the copier is used. For Energy Star copiers, we expect to see shorter delay times, higher enabling rates, and lower ‘off’ power levels. In addition, we expect to see daytime savings on some machines, depending on the distribution of imaging jobs across the day.

The ASTM test defines five copier modes: "copying", actively making copies; "standby", immediately ready to copy, but not doing so; "energy-saver", using less power than when in standby mode, but not off; "plug-in", plugged-in to an outlet, but not turned on; and "warm-up", becoming ready to copy. Many copiers use some energy in plug-in mode, even if they appear to be completely off. Copiers may be switched off manually, or automatically by an auto-off feature; the ASTM definitions do not distinguish between these states, though the copier may draw different levels of power. Most copiers use heat in fusing the toner to the paper, so the warm-up energy and much of the copying energy is used to maintain the fuser roll at the required temperature.

Notes: The arrows indicate electricity savings opportunities for Energy Star copiers. These are average loadshapes for several copiers and days, and so do not correspond to any particular day.

A copier's operating pattern is the distribution of time spent by the copier in each mode plus the imaging rate. We define the three primary modes as "On", "Low-Power", and "Off". Our "on" mode combines the copying, standby, and warm-up ASTM modes. Our "low-power" and "off" modes correspond directly to the ASTM "energy-saver" and "plug-in" modes. A "standard operating pattern" is an estimate of current use patterns from which annual energy use, existing program savings, or potential further savings can be calculated.

The ASTM test assumes that copiers are only rarely left on during nights and weekends, but the Energy Star copier program was developed in part due to the belief that copiers are frequently left on during non-work hours. Thus, a key uncertainty is the level of power use typical during night and weekend hours (which far exceed daytime hours).

Figure 1-2 shows how several various factors combine to determine national energy use by copiers. Machine characteristics include any feature that can change the energy use of the copier; power management is the most apparent of these, but factors such as warm-up time and duplexing speed also affect electricity use. Copier users interact with the machine's controls, with the effect of this seen in the configuration chosen and the operating pattern. The operating pattern is the most variable and uncertain factor. Energy is also consumed in the production of copier paper (discussed briefly in Section 4), toner, and the copier itself.

A Japanese test procedure (MITI, 1994) was created to facilitate an energy efficiency standard for copiers, to be effective in 2000. However, Japan will probably adopt the Energy Star program, making this test superfluous. The standard addresses daytime electricity use only; it says nothing about night or weekend use, perhaps presuming that the machine would be manually turned off during those times. Unlike the ASTM test, the imaging rate is specified by the test procedure.

Various specifications for efficient copier performance have been prescribed over the years. The Energy Star criteria are the most widespread, and are the basis for the Green Seal energy criteria. The Green Seal recommendation adds several non-energy criteria and strengthens the duplexing requirement (Green Seal, 1995). The NUTEK (NUTEK, 1995) criteria are similar to Energy Star. The Swiss "Target Values" for copiers are of the same form, but more stringent in low-power mode for most copier speeds (cited in Dandridge, 1994). An Australian implementation (SEDA, 1997) uses the US EPA criteria, but their savings estimates assume that copiers are left on 70% of nights and weekends.

To develop our assessment procedure for the performance of Energy Star copiers, we first reviewed the available literature on copier electricity and paper use. There have been no previous efforts to measure Energy Star copier performance. We interpret the term ‘literature’ broadly, including academic papers, gray literature, information from manufacturers, and industry market analyses and equipment evaluations.

Given the importance of the ASTM test procedure for copier electricity measurement, wide availability of a database of ASTM test results would allow purchasers to make better purchasing decisions and would aid copier research. The only compilation we found was a database assembled by Dandridge in 1994 (Dandridge, 1997), developed to help design the Energy Star program. The database contains test results for 130 copiers of varying speeds. There are only three Energy Star models are represented, because the database predates the program. We used the data to determine average power levels for each copier speed segment, averaging the models by segment. The results are presented in Section 4.3.

We did not find ASTM test results accompanying field measurements of copier electricity use. We requested ASTM test results from manufacturers, as they often conduct the tests and provide the results to potential customers on request. We were largely unsuccessful in this; one company provided extensive results, but too late to incorporate into this study. Results for four copiers (from two different manufacturers) were obtained, but only under a condition of confidentiality.

Acquaviva (Acquaviva et al., 1994a, 1994b) provides a detailed discussion of how copiers actually use energy. He reviews the effect of job type (as it effects cycle energy) and imaging rates, and assesses copier model efficiency over time.

Dandridge (Dandridge et al., 1994, 1996) studied copier energy use with the design of the Energy Star program in mind, reporting measured copier use patterns from light sensor data and the electricity use of three copiers. The ASTM test was shown to provide a good approximation of measurements made during ordinary usage for the three copiers. Factors such as duplexing and distribution of jobs were assessed to determine their effect on electricity use. The light sensors reliably recorded when imaging occurred, but not the job type.

Several studies assessed office equipment electricity use. Pacific Gas & Electric measured the use of four copiers, compared it to copiers that use less electricity, and estimated savings from the use of timers to turn them off when not used during the day (Martin, 1991a, 1991b, 1992). ASTM tests were not used, but the power used in each mode was combined with actual operating pattern data to estimate annual energy use. This included identifying when each copying job occurred to estimate the benefit of ‘idle’ timers. A Swedish study (Levinsson et al., 1996) included detailed measurements of two copiers, one of which was an Energy Star copier. An Electric Power Research Institute study (Arney et al., 1996) was designed to evaluate baseline office equipment loads and loadshapes. Nine copiers were monitored for periods ranging from two days to several weeks.

Tests have been made of controlling devices by the Florida Solar Energy Center (Floyd, 1997) and Bayview Technologies (Wood, 1997). Although the number of machines tested is small, the devices appear to be successful and accepted. Two British studies (Parsloe et al., 1992a, 1992b) provide limited measurements of copier power.

Several studies involved after-hours assessment of equipment on. A Thai study (Mungwititkaul et al., 1997) aggregated measurements by mode for 19 copiers. Detailed monitoring of four copiers resulted in a distribution of operating modes by time for power-managed and non-power-managed copiers. Power management was disabled in 90% of 19 copiers audited, but security personnel manually turned off machines each night. Night audits of equipment status were collected by Bayview Technologies (Wood, 1997), a manufacturer of controlling devices for office equipment. Our analysis of these data are presented in Section 4.1.

A variety of studies have assessed policy options for reducing copier electricity use. Harris et al. (1994) explored several issues related to copier energy use and efforts to increase their efficiency. The study includes estimates for current U.S. electricity use and the savings potential of copier power management and increased duplexing rates. Koomey et al. (1995) assessed office equipment energy use in commercial buildings for "Business-as-usual", Energy Star, and "Advanced" scenarios; the purpose was to identify the savings from existing EPA programs, and a reasonable potential that could be gained with additional effort. All copiers were grouped into one category (not disaggregated by speed). A Swiss study (BEW, 1992, 1993) assessed electricity used by devices when they are not in active use. Measured power levels were reported for 15 copier models from one manufacturer. We compared these data to those from Dandridge and found 8 models in common. Standby energy ranged from identical values to up to 50% different. Low-power energy ranged from close values to 100% different. There is no clear explanation of these differences. An early assessment (Yanagawa, 1992) reviews methods to reduce copier electricity (including those that reduce imaging and reduce the total number of devices in an office), and the distribution of energy use within a copier. Extrapolations to national (U.S.) estimates of energy use or savings have been made in a variety of policy analyses. Three estimates (Dandridge et al., 1996, Harris et al., 1994, and Koomey et al., 1995) were made before the Energy Star copier program was underway. The EPA makes its own estimates of program performance which are continuously modified as new data are available.

Duplexing increases electricity use by requiring extra energy for counting copies (when that occurs), more time at the copying power level due to the usually slower speed of duplex copying, and potentially by increasing total imaging with increased paper jams. The literature on duplexing rates is discussed in Section 4.4.

The Swedish study included a survey of copier users in two offices. At one, the acceptable warm-up time ranged from 10 seconds to 5 minutes, and at the other, no more than one minute was acceptable.

In assessing how previous efforts inform this project, it is important to note that they had different goals, and nearly all predate the Energy Star copier program. Consequently, data critical to our analysis were not always recorded or reported (examples include the brand and model of each copier measured).

This section presents a general overview of field measurements and techniques, followed by a discussion of the particular methods and approaches used in this study. The types of field measurements reviewed include auditing, monitoring, and user satisfaction surveys. Auditing consists of a one-time visit to ascertain a copier's state or configuration. Monitoring involves use of a datalogger to record electricity or imaging over a period of time, or periodic recording of imaging and paper counters. User satisfaction surveys queried users of copiers about their attitudes and actions regarding power management, duplexing, and overall copier performance. Audits require much less effort than detailed monitoring, facilitating the analysis of larger numbers of machines. The greater numbers of machines visited with audits gives more confidence than can be gained from detailed measurements only.

The existence of common methods makes it more likely that information collected by different researchers will be correct, complete, and amenable to comparison or combination. Experience with analyzing the performance of PCs and monitors further reinforced the need for such guidelines. Finally, researchers need to know the appropriate parameters for field monitoring for variable such as granularity of data sampling, recording, and necessary days of recording.

We identified three major methods to evaluate the performance of energy-efficient copiers–model audits, night-time audits, and detailed monitoring. These illuminate the character and use of features that underlie the Energy Star copier program. We also used user satisfaction surveys and measured duplexing.

Model Audits are used to identify ‘machine characteristics’ (see Figure 1-1), such as a model's power management and duplexing features, and determine how to inspect and change configurations. Examinations of the operating manual and the machine itself are the primary sources of model audit data. It is desirable to conduct model audits of relevant models before other field monitoring activities.

Night Audits reveal copier status during night and weekend hours. Night audits also offer an opportunity to check configurations. Night audits are most complete when the auditor has an inventory of all copiers to aid in locating machines and collecting relevant manuals. For most office environments, audits should take place after 9 p.m. to allow machines to equilibrate to their final state.

Detailed Monitoring can be used to evaluate operating patterns, verify operation of controls, and extrapolate to annual use. In addition to recording the information specified for a night audit, the value of the image counter should be recorded when the monitoring equipment is put on and again when it is taken off, to allow calculating imaging rates and electricity use per image.

Duplexing Measurement allows calculation of baseline measures of duplexing rates and the effect of enabling default duplex. Measurement of duplexing rates requires separate internal counters for total images, paper, and sheets through the duplexing unit (though with any two of these, the third can be calculated). Some machines have easily accessible counters; some have them accessible only to key operators or service technicians (and thus potentially available to researchers); and some do not have such counters at all.

User Satisfaction Surveying aims to reveal the reaction of copier users to the machines generally and to electricity and paper issues specifically. Paper handling (such as default duplex) has a much greater impact on the typical user than does most power management, so user satisfaction concerns were greater when default duplex was part of the Energy Star copier program. In some PCs, power management would lead to equipment malfunction, raising the question of whether this occurred in copiers.

National Estimates of copier energy use and savings are extrapolations based on information about the stock of copiers present in the U.S., average power levels of copiers, and a standard operating pattern, as shown in Figure 1-1. They are used to indicate program potentials and savings.

Details of our techniques can be found on the web, along with a discussion of confounding factors that impede data collection or add uncertainty. These include difficulties in machine identification, ambiguous terms, duplexing issues, audit problems, and on-site problems.

We reviewed the paper and electronic versions of copier characteristics data collected by Buyer's Laboratory, Inc. (BLI, 1997). These provide valuable data on specific models, and previous work (Nordman, 1996a) analysis of several characteristics, including the relative speeds of simplex and duplex copying. In the course of monitoring and auditing, we reviewed the product manuals for about 30 individual copier models and assessed their electricity and paper features.

During our night audits, we attempted to record the following information: copier location, brand, model number, power status (if off, manually, or by auto-off), if any error mode was extant, presence of controlling devices or unusual accessories, plugged-in status, and any relevant signs (to paper or power) posted in the vicinity. When possible, we recorded the power management and duplexing configuration. Weekly timers are used infrequently, but we checked for their use when possible. To allow calculation of duplexing rates, we recorded any available internal counters. We visited 228 copiers at seven sites, most after 9 p.m. The sites included LBNL itself, three city office buildings, a federal office building, the main offices of two large corporations, and a hospital.

We set the monitoring period for this project at two weeks. We recorded power use every 30 seconds using a 15 Amp power transducer and a datalogger capable of storing about 65,000 readings. We did not change the copier's configuration, even if any Energy Star features were not enabled. We monitored eleven copiers at six sites. In reviewing results from detailed measurements, we produced a number of standard graphic representations of the electricity use. A particularly useful one is the ‘calendar’ graph of the daily use (Figure 4-2). We defined a method to derive standard figures of merit for each copier such as average weekday on-time, and annual energy use.

We derived the average weekday and weekend day energy use, excluding any holidays and incomplete days. We then extrapolated to annual energy use based on 250 work days per year and 115 weekend/holiday days (see Table 4-4). We derived an average power level or 'Transition Value' for each copier, above which the copier is assumed to be on for the period. Energy above this level is assigned to copying activity and energy below this level to keeping the copier in a ready-to-copy state. This is intended to correspond to the ASTM "standby" level (so long as low-power and auto-off delay times are less than an hour). A separate transition value identifies the low-power mode level.

To measure the effect of default duplex, we recorded duplexing counters at one week intervals. We recorded this for at least one month with the machine in default simplex, so that the baseline rate was well-determined, and for newly installed copiers, for people to get used to the machine. After enabling default duplex, we recorded the duplexing rate for several months. We measured the effect of default duplex enabling on two copiers.

We surveyed 56 users of 11 Energy Star copiers to determine their awareness of the model's energy saving features, level of satisfaction with them, and suggestions for improvements. We inquired specifically about their awareness of each feature’s presence (auto-off, low-power, and default duplex), what they thought the delay time was that the machine was set to use, and if recovery times from low-power or off were too long. We asked about ease of use of duplexing as well as open-ended questions about the copier. We only included users who reported making at least 10 copies per week. Before surveying users, we audited the copier to find out which features were present and enabled. We did not ask about Energy Star features that were disabled. A total of fifty-six people were surveyed.

To derive estimates for national copier electricity use and savings from Energy Star copiers, we sought out the best available source for each of the main factors. For copier power levels, we used the ASTM test results collected in 1994 (Dandridge, 1997). For the standard operating pattern, we needed to estimate the hours of daytime use and the night-time status. For daytime hours, we relied on monitored data collected for this project and reported by others in the literature. For night status, we used the findings from our night-time audits. We defined an operating pattern for each of four copier speed categories.

This section summarizes the results of our data collection and analysis efforts. We review findings from the late-night audits, detailed electricity monitoring, a comparison with ASTM operating patterns, duplexing estimates and measurements, the user satisfaction survey, and estimates of annual electricity use.

The primary purpose of the late night audits was to determine the percentage of compliant and non-compliant copiers in each major mode (on, low-power, or off) during night and weekend time. From loadshapes in the literature, we determined copier night-time status for 18 copiers (Arney et al., 1996; Levinsson et al., 1997; Martin, 1991; Allen, 1997; and Dandridge, 1997). We found an aggregate status of 31% on and 11% in a low-power mode for a total of 42% at least partially on.

The largest collection of night-time copier audit data we found was from audits conducted by Bayview Technologies, Inc (Wood, 1997). These audits were conducted at about 20 locations in the U.S. from late 1995 through 1997 and were primarily designed to survey the status of personal computers and monitors. Table 4-1 presents the main Bayview results. The Bayview audits were generally conducted between 6 and 7 p.m. on weeknights, early enough in the evening so that copiers would rarely have had enough time to automatically turn off. Thus, they are not late enough to reveal power management operation, they do indicate manual turn-off rates. These figures may therefore exaggerate the percent of copiers on. However, the large fraction of copiers on at night highlights the importance of the auto-off part of the Energy Star program.

Notes: 'On' includes low-power modes. The total set includes many more copiers than the two subsets as the brand and model was not always recorded. The brand and model information are necessary to determine if a copier is Energy Star compliant and its speed.

To further understand the after-hour state of copiers, we conducted a series of night-time audits, assessing a total of 228 copiers. The sites included most of LBNL, three municipal office buildings, a federal office building, two large corporation's offices, and a hospital. Audits were generally conducted after 9 p.m. so that the copiers had enough time to reach their final state for the evening. The primary results are presented in Table 4-2. The Bayview data suggest that the manual turn-off rate is 16% or 18%, while the LBNL audits suggest that it is at most 17%.

Table 4-2. Night-time Copier Status (LBNL audits)

Notes: Many of the Energy Star copiers that were found ‘on’ were verified to have had the auto-off feature disabled. We also found two multi-function devices, both fully on, and three color copiers, two of which were fully on and one in a low-power mode. These raise the total to 228. We also found four copiers on and in an error mode–these seem to prevent auto-off features from taking effect. Three copiers were found broken and off. On 13 of the 22 Energy Star copiers that were found on, we verified that auto-off had been disabled. On 12 compliant copiers at two sites, auto-off was verified disabled–interestingly all were the same model; one non-compliant copier was verified disabled. On 3 compliant copiers (and one non-compliant), auto-off was enabled but the low-power mode was disabled. Delay times were recorded at one site and 8 copiers had 90 minute delays with three having 120 minute delays to auto-off.

There is striking consistency between the Bayview results and the results of our audits for conventional copiers. That Energy Star compliance does not seem to lead to a significant improvement in the fraction of copiers fully on at night is surprising and distressing, but mirrors the problems with disabling that previously were seen to reduce power management savings in PCs and monitors. We expected that power management of copiers would vary significantly with copier speed, so we disaggregated the audits by the Energy Star speed segments as shown in Table 4-3. The results follow no clear pattern; low-speed and high-speed Energy Star copiers were more likely to be on than their conventional counterparts, but in all segments (except very high speed machines) Energy Star copiers were more likely to be off. We added a separate very high speed category (>90 cpm) since such copiers are generally used by a single operator, and thus more likely to be turned off manually.

Table 4-3. Night-time status of LBNL-observed copiers (by speed segment)

Notes: aOver a third of these copiers were of one copier model, and so may not be representative of high speed Energy Star copiers.

Given the small sample sizes (particularly for the small copiers) we interpret the quantitative results cautiously. Higher speed machines are likely to be used by more people (except very high speed machines which typically are used by a single operator) and have longer warm-up times. This could explain the greater disabling of power management of these Energy Star copiers and the lower percentage of conventional copiers off. User complaints about power management may lead to complete disabling of power management rather than lengthening delay times or using only low-power modes. Anecdotal evidence indicates that copier service personnel are partially or completely responsible for much of the disabling of power management. Except for very high speed copiers, Energy Star copiers are more likely to be off than are conventional machines. Low-power modes are also common and frequently enabled in conventional copiers. As people learn that Energy Star copiers turn themselves off, they may abandon manual turn-off altogether. We show the Bayview and LBNL audit results, as well as the ASTM test assumptions, in Figure 4-1.