Flux-O-Meter

Making luminous flux measurements more accessible

Andrew Bierman, Martin Overington Funded by EPA Luminous Flux, Lumens, Total Light Output... Who cares?

• Lighting practitioners use illuminance (lux, footcandles) and luminance (cd/m2, footlamberts) • Luminaires are characterized by intensity, intensity distributions, luminance • Flux measurements are usually confined to – National labs – Lamp Manufacturers • The current practice suffices if we don’t care about total system efficacy. Luminous flux is key to the issue of generating and controlling light efficiently.

• Standard photometric reports do not reveal system efficacy • Luminaire efficiency does not fully take into account thermal, positional and ballast effects (system issues) • Based on relative photometry – Numbers scaled to rated lamp light output Luminaire Efficiency Rating (LER)

• Works well for well characterized systems • Insufficient data available for CFL and lower volume/specialty lamps and ballasts • Gets complicated for universal ballasts, different operating positions, atypical luminaire configurations, etc. • Errors from –23% to +86% for CFL luminaires – NLPIP Specifier Reports: Energy-Efficient Ceiling Mounted Residential Luminaires (1999) Must measure flux to determine true system efficacy Traditional Flux Measurements

• Integrating Spheres – Very sensitive to departures from ideal – Substitution method is the practical means – Requires calibrated flux standards – expensive, delicate! Goniometers

• Requires large test distance, large spaces • Sensitive to stray light (need dark room) • Integrate intensity over solid angle to get flux • Only a few facilities have one Alternate Method

Illuminance Sampling • Measure the illuminance at many points around the source (an imaginary sphere) • Integrate over area Integrating illuminance over area

Flux = ò EdA sphere Benefits / Limitations

• No minimum distance • Only gives total flux (no requirements distribution data) • No flux standards • Sensitive to stray light • Not sensitive to (requires black positioning of source (?) surroundings • Not sensitive to source • Rotation axes must be size (?) perpendicular • Not sensitive to distribution (beam vs. blob) (?) Apparatus

• Low-cost approach – Exploits rapid sampling, low-cost, 12-bit computer data acquisition boards – Inexpensive motion control • Analog encoders (rotary variable resistors) • Continuous motion (no feedback motion control) – Cost: < $1000 plus computer, software How it takes measurements

-1

-2 z

-3

-4

-5 4 2 4 0 2 0 -2 -2 -4 -4 y x How it takes measurements

7 Elevation angle Azimuthal angle 6 Illuminance

3 Relative signal 2

0 0 500 1000 1500 2000 2500 3000 3500 Filtered data points Dealing with Modulated Light Sources

Unfiltered Data, CFL Magnetic Ballast 0.65 0.6 7.2° 0.55

0.5

0.45

0.4

0.35

0.3 Illuminance detector signal, V

0.25 Measured illuminance Filtered illuminance 0.2 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1 Time, s Comparison with Integrating Sphere

Lamp Type

60W R- 28W CFL 60W A-19 100W A-19 Lamp Screwbase

Sphere (lm) 811 1576 477 1443

Flux-O-Meter (lm) 828 1610 496 1387

Difference (%) 2.1 2.1 3.9 -4.0 Fixture Comparison with Goniometer

20W Circline (bare lamp) Fixture #1 Fixture #2 Fixture #3

Goniometer (lm) 1232 818 1132 640

Flux-O-Meter (lm) 1272 872 1190 673

Difference (%) 3.2 6.4 5.0 5.0