ENERGY STAR Computer Monitor Test Methodology

1

• ENERGY STAR Computer Monitor Test Methodology

Craig Hershberg US EPA Hershberg.Craig_at_epa.gov
2
Objectives of Todays Presentation

• Provide an overview of the final test methodology
• Share key points of discussion during the test
  methodology development phase, and their
  resolutions
• Present and gather industry input on two
  remaining issues regarding multiple measurements

3
Test Methodology Goals

• Develop sound procedure for testing computer
  monitors in On Mode incorporate existing
  methodologies for Sleep and Off
• Design a methodology that is comprehensive and
  produces repeatable test results
• Decrease burden on manufacturers by referencing,
  where applicable, existing and widely used
  industry standards for testing
• VESA
• IEC

4
Development Phase

• Extensive input from stakeholders
• Manufacturers, trade associations, European and
  Japanese stakeholders, TCO Development
• Various drafts broadly distributed for comment
  during 9-month development phase
• Final test methodology emailed to stakeholders in
  February 2003 minor changes made since then
• Development Summary captures key additions and
  changes that led to final methodology
• Emailed to stakeholders and available on Web site
• Basis for todays presentation

5

• Non-Energy Factors

6
Removed References to Three Non-Energy Factors

• Three non-energy factors from Draft 1 have been
  removed from Draft 2
• Defective pixels (no effect on power consumption)
• Contrast ratio (VESA Flat Panel Display
  Measurements (FPDM) Standard 2.0 does not define
  parameters for measuring contrast ratio)
• Warranty (market-specific so hard to implement
  and police for global products like computer
  monitors)

7
Retained Two Non-Energy Factors

• Two non-energy factors originally included in
  Draft 1 of the specification have been moved to
  the test methodology
• Refresh rate
• Luminance (brightness)
• Ensure all computer monitors are being tested
  under similar conditions

8
Refresh Rate

• Different refresh rates given for LCDs and CRTs,
  to accommodate differences in technologies
• LCDs Measured at 60 Hz, unless a different
  refresh rate is specifically recommended by the
  manufacturer
• CRTs Measured at 75 Hz (long-time norm in North
  America)
• Discussion over use of 75 Hz or 85 Hz, which is
  consistent with TCO 99 and general practice in
  Europe
• 75 Hz chosen because most appropriate for power
  measurement purposes, whereas 85 Hz better for
  ergonomic purposes (reduces flicker, but
  increases power consumption)

9
Luminance (Brightness)

• Levels chosen that support the implementation of
  a performance-based specification
• Allow for comparisons of power use between
  computer monitors when providing the same visual
  experience for the user
• CRTs Measured at minimum of 100 candelas per
  square meter (cd/m2)
• LCDs Measured at minimum of 175 cd/m2
• Higher than CRTs because 100 candelas per square
  meter is much easier for LCDs to attain than for
  CRTs
• All luminance test patterns borrowed from VESA
  FPDM Standard 2.0

10
125 vs. 175 cd/m2 for LCDs

• 125 cd/m2 Not selected because it would
  unfairly advantage models that only have
  brightness control, or operate in digital mode
• Adjusting brightness to get 125 cd/m2 may lower
  power consumption by reducing power to
  backlights, allowing these models to qualify more
  easily
• Maximum brightness Not chosen because it may
  punish better performing models with a high
  brightness range, to allow for some degradation
  over time

11

• Dark Room Conditions and Color Controls and
  Peripherals

12
Dark Room Conditions

• Suggested by several members of industry for
  light measurements
• References current industry norms through VESA
  FPDM Standard 2.0

13
Color Controls and Peripherals

• All color controls and peripherals should be
  placed at factory default settings
• No external devices should be connected to any
  USB hubs/ports
• Any built-in peripherals should be set to off, or
  placed in their minimum power configuration (as
  adjustable by the user)
• Circuit removal or similar actions not under user
  control may not be undertaken (e.g., removal of
  built-in speakers)

14

• DVI Inputs and Digital Monitors

15
Addressing DVI Inputs and Digital Monitors

• Added a clarifying statement re stable power
  measurements in Off Mode
• Test methodology for all three operating modes
  states that power measurements should be taken
  once the power readings are stable
• Power readings never quite stabilize due to DVI
  input check cycle
• For models with DVI inputs, manufacturers should
  ignore the DVI input check cycle when measuring
  power consumption in Off
• Included information on how to test digital only
  interface monitors
• Monitors with both analog and digital interfaces
  should be tested in analog interface
• See footnote 1

16

• Multiple Measurements

17
Data Variability Analysis

• First round of testing Requested testing of 5
  units at each of 3 different voltage/frequency
  combinations
• 100 Volts/50 Hz, 115 Volts/60 Hz, 230 Volts/50 Hz
• Data analysis Considered the variability of
  manufacturers test data provided at each of the
  voltage/frequency combinations
• Findings 1) Much greater variation by sample
  than by voltage 2) In On Mode, power use was
  often higher at 115 than 230 and 3) More data
  points submitted by manufacturers at 115 than
  other voltages

18
Active Power Variability Among Different Samples

• Up to 30 variability by sample for some models

19
Active Power Variability Among Different Voltages

• Less than 4 variability by voltage for majority
  of models

20
Reduced Number of Measurements per Model

• Second round of testing Requested that
  interested manufacturers only test 3 units at 115
  Volts/60 Hz
• Overall testing reduced from 15 data points per
  model to 3 data points per model

21
Questions for Industry

• Does industry agree that testing at only one
  voltage/frequency combination is sufficient?
• And if so, does it make sense to test at 115
  Volts/60 Hz, as proposed in Draft 2?
• Another option is to test at the
  voltage/frequency combination of the country
  where the monitor will be sold
• Which option would industry prefer, and why?
• Are there any manufacturer concerns with
  requiring three test units per model?
• Do manufacturers have any other alternatives that
  they would like EPA to consider?