TM-30: Introduction and Latest DevelopmentsMichael Royer, PhD | Pacific Northwest National Laboratory

Credit(s) earned on completion of this course will be reported to AIA CES for AIA members. Certificates of Completion for both AIA members and non-AIA members are available upon request.

This course is registered with AIA CES for continuing professional education. As such, it does not include

content that may be deemed or construed to be an approval or endorsement by the AIA of any material of construction or any method or manner ofhandling, using, distributing, or dealing in any material or product.

___________________________________________

Questions related to specific materials, methods, and services will be addressed at the conclusion of this presentation.

3

TM-30 was developed, through a consensus process, to provide a comprehensive set of objective information—going beyond simple average values—that can be used collectively to make informed decisions about subjective perceptions, such as preference or naturalness, given a context. New research has shown the value of this robust system in capturing human judgements of lighting quality. Nonetheless, TM-30 is a tool, not an answer. Its limitations must be understand, and it must be combined with other color information, such as chromaticity, luminance, and distribution of light, when choosing a source.

Abstract:

4

Summarize the process that led to the recommendation by the IES Color Metrics Task Group

Describe the components of the color rendition evaluation system and how they can be interpreted to aid design and specification

Comprehend the conceptual framework for the underlying calculations

Recognize the limitations of the this system, and more generally recognize the limitations of all measures for color rendition

Understand how the objective information in TM-30 can be used to aid in subjective design decisions (i.e., matching the right source to an application).

Learning Objectives:

5

6

Ra (CRI) = 78R9 = -11

Ra (CRI) = 68R9 = -37

7

Ra (CRI) = 78R9 = -11

Ra (CRI) = 68R9 = -37

8

Fidelity Index (Rf)Gamut Index (Rg)

High Level Average Values

Core Calculation EngineModern Color ScienceNew Color Samples

Color Vector GraphicColor Distortion Graphic

Graphical Representations

Skin Fidelity (Rf,skin)

Fidelity by Hue (Rf,h#)

Chroma Shift by Hue (Rcs,h#)

Fidelity by Sample (Rf,CES#)

Detailed Values

Color Fidelity

Fidelity Index (Rf)

TM-30 Method for Evaluating Color Rendition

The accurate rendition of color so that they appear as they would under familiar (reference) illuminants

(0-100)

Perfect Fidelity

Increase Saturation

DecreaseSaturation

Positive Hue Shift

Negative Hue Shift

CRI = 80 CRI = 80

Constant Fidelity (CRI)

(Also possible to change lightness, not shown)

Color Fidelity

Fidelity Index (Rf)

TM-30 Method for Evaluating Color Rendition

The accurate rendition of color so that they appear as they would under familiar (reference) illuminants

(0-100)

Color Gamut

The average level of saturation relative to familiar (reference) illuminants.

Gamut Index (Rg)~60-140 when Rf > 60

50 60 70 80 90 10060

70

80

90

100

110

120

130

140

Fidelity Index, Rf

Gam

ut In

dex,

Rg

Combinations not

possible for white light.

Reduced Fidelity

Incr

ease

d Sa

tura

tion

Decr

ease

d Sa

tura

tion Reference

IlluminantTwo-

Axis

Syst

em • Evaluate tradeoffs between fidelity and saturation.

• Cohesive system from the same calculation engine.

• But average values don’t tell the whole story…

Color Fidelity

Fidelity Index (Rf)

The accurate rendition of color so that they appear as they would under familiar (reference) illuminants

Color Gamut

The average level of saturation relative to familiar (reference) illuminants.

Gamut Index (Rg)

(0-100)~60-140 when Rf > 60

Gamut Shape

Changes over different hues

Color Vector Graphic

TM-30 Method for Evaluating Color Rendition

Hue Bin FidelityHue Bin Chroma Shift

Rf = 75 | Rg = 100 | CCT = 3500 K Rf = 75 | Rg = 100 | CCT = 3500 K

Decreased Saturation

IncreasedSaturation

Hue Shift

15

0

20

40

60

80

100

76 7264

7485 82

75 72 7568 72 71

83 87 84 81

Fide

lity

Inde

x by

Hue

, Rcs

,hj

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16-40%-30%-20%-10%

0%10%20%30%40%

12% 11%5%

-3% -6%

3%10%

16% 14%10%

4%

-1% -4% -5%

2%9%

Chro

ma

Chan

ge b

y Hu

e, R

cs,h

j

0

20

40

60

80

100

74 7363

7280 80 79

85 8374 70 72

83 78 74 73

Fide

lity

Inde

x by

Hue

, Rcs

,hj

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16-40%-30%-20%-10%

0%10%20%30%40%

-14%-11%-3%

5%11% 11%

4%

-2%-8%

-13%

-3%

7% 9% 11%2%

-5%

Chro

ma

Chan

ge b

y Hu

e, R

cs,h

j

16

Sum

mar

y of

Upg

rade

sCRI Calculation Engine (1974) TM-30 Calculation Engine (2015)

CIE 1964 U*V*W* CAM02-UCS (CIECAM02)

Von Kries CAT CIE CAT02

8 color samples 99 color samplesMedium chroma/lightnessSpectral sensitivity varies

Uniform color space coverageSpectral sensitivity neutralVariety of real objectsMunsell samples only

Technical Improvement

Ref Illuminant Step Function Ref Illuminant Continuous(Uses same reference sources, but blended between 4500 K and 5500 K)

No lower limit for scores 0 to 100 scale (fidelity)

Nice to Have

Fidelity Only Fidelity, Gamut, Graphical, Hues Philosophical Change

70 75 80 85 90 95 10070

75

80

85

90

95

100

CIE Ra (CRI)

TM-3

0 R

f

70 75 80 85 90 95 10070

75

80

85

90

95

100R² = 0

CIE Ra (CRI)

TM-3

0 R

f

~16 point spread in Rf scores at Ra = ~80

70 75 80 85 90 95 10070

75

80

85

90

95

100 FilamentDaylight ModelsNarrowband Fluo-rescentBroadband FluorescentHIDHybrid LED

CIE Ra (CRI)

TM-3

0 R

f

20

49 point spread (error) in fidelity score at CRI of 80.

50 60 70 80 90 10040

50

60

70

80

90

100

CIE Ra

TM-3

0 Rf

5,000 Real and Modelled* SPDs*All modelled SPDs composed of combinations of Gaussian primaries; chromaticity on Planckian locus between 2700 K and 7000 K

For more information:Smet KAG, David A, Whitehead L. 2015. Why color space uniformity and sample set spectral uniformity are essential for color rendering measures. Leukos 12(1–2):39–50.

21

0 20 40 60 80 10060

70

80

90

100

110

120

130

140

CIE Ra

TM-3

0 Rg

0 20 40 60 80 10060

70

80

90

100

110

120

130

140

TM-30 Rf

TM-3

0 Rg

Is the error systematic?

D227

22

D222

23

D223

24

D224

25

D225

26

Illuminance: ~20 fcCCT: 3500 KLighting Conditions: 26Objects: Generic Consumer, balanced huesApplication: UndefinedParticipants (28): 18-65, 16 females 12 malesRating Questions: Normal-Shifted, Saturated-Dull, Like-Dislike

60 65 70 75 80 85 90 95 10070

80

90

100

110

120

130

TM-30 Fidelity Index, Rf

TM-3

0 Ga

mut

Inde

x, R

g

60 65 70 75 80 85 90 95 10070

80

90

100

110

120

130

TM-30 Fidelity Index, Rf

TM-3

0 Ga

mut

Inde

x, R

g

60 65 70 75 80 85 90 95 10070

80

90

100

110

120

130

TM-30 Fidelity Index, Rf

TM-3

0 Ga

mut

Inde

x, R

g

60 65 70 75 80 85 90 95 10070

80

90

100

110

120

130

TM-30 Fidelity Index, Rf

TM-3

0 Ga

mut

Inde

x, R

g

60 65 70 75 80 85 90 95 10070

80

90

100

110

120

130

TM-30 Fidelity Index, Rf

TM-3

0 Ga

mut

Inde

x, R

g

60 65 70 75 80 85 90 95 10070

80

90

100

110

120

130

TM-30 Fidelity Index, Rf

TM-3

0 Ga

mut

Inde

x, R

g

60 65 70 75 80 85 90 95 10070

80

90

100

110

120

130

TM-30 Fidelity Index, Rf

TM-3

0 Ga

mut

Inde

x, R

g

36

17 22 26 15 16 23 6 24 18 9 8 7 25 14 10 12 5 20 4 21 13 11 3 19 1 20

102030405060708090

100

Setting ID in Rank Order

Fide

lity

Inde

x Rf

Most Liked Least Liked

Fidelity and Preference?

R2 = 0.06

37

1 2 3

(These aren’t necessarily the most preferred sources possible, just the most preferred sources from this experiment).

60 70 80 90 10070

80

90

100

110

120

130

IES TM-30 Rf

IES

TM-3

0 Rg

Model r2 = 0.68

p =

0.00

0

p = 0.042

Dislike

Like

5.5

5.0

4.5

4.0

3.5

Fidelity + Gamut and Preference?

Same Fidelity, Same Gamut, Significantly Different Rating.

-30% -20% -10% 0% 10% 20% 30%1

2

3

4

5

6

7

8

f(x) = 85.457325247 x³ + 12.74591052 x² − 9.620714244 x + 4.1387295218R² = 0.813215625190091

Hue Bin 16 Chroma Shift (Rg,h16)

Mea

n Pr

efer

ence

Rati

ng

Dislike

Like

Red Chroma Shift and Preference?

Best Fit Model for Preference: Like-Dislike = 7.396 - 0.0408(Rf) + 103.4(Rcs,h163) - 9.949(Rcs,h16)

2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 72.02.53.03.54.04.55.05.56.06.57.0

R² = 0.935547960432472

TM-30 Model Predicted Rating

Parti

cipa

nt R

ating

(Pre

fere

nce)

42

Normalness = Fidelity + Red Fidelity/Saturation

Saturation = Red Saturation

Preference = Fidelity + Red Saturation

Rf > 80 Rf,h1 > 80 0% < Rcs,h1 < 8%

Maximize Rcs,h16, Rcs,h1

Rf > 74 0% < Rcs,h16 < 15% 0% < Rcs,h1 < 15%

(Rg > 100)

Context =

50 60 70 80 90 10060

70

80

90

100

110

120

130

140Phosphor LED

Color Mixed LED

Hybrid LED

Standard Halogen

Filtered Halogen

Triphosphor Fluorescent, 7XX

Triphosphor Fluorescent, 8XX

Triphosphor Fluorescent, 9XX

Metal Halide

Fidelity Index, Rf

Gam

ut In

dex,

Rg

Experimental Preferred Zone*

-50%

-40%

-30%

-20%

-10%

0%

10%

20%

30%

40%

50%

Rcs,h

16

Experimental Preferred Zone*

Same Fidelity, Same Gamut, Significantly Different Rating.LER = 343 LER = 311

Why so few red-enhancing sources?

CIE R a

IES

TM-3

0 R f

Why so few red-enhancing sources?M

odel

r2 = 0

.06

Common Commercially Available Sources (Developed for Ra):

Ra 74, LER 348 Ra 85, LER 343 Ra 83, LER 309

Ra 80, LER 272

“Enhanced” Sources:

Ra 77, LER 136 Ra 87, LER 295

49

Context…

50

51

52

53

54

55

56

57

58

59

Und

erst

andi

ng T

he To

ol1. A metric value doesn’t tell you how the product will perform in any

given environment.

2. The “accuracy”/applicability of the metric depends on if the sample set is similar to the actual space.

3. An average color rendering metric shouldn’t be used to predict how a source will render reds, or skin tones, or any specific set of objects.

4. TM-30 offers substantially more information, which is essential for evaluating color rendering characteristics.

5. The best source is depends on the context (objects, type of space/application, illuminance, occupants, etc.)

60

Reso

urce

sIES Technical Memorandum (TM) 30-15 (Includes Excel Calculators):IES Method for Evaluating Light Source Color Renditionhttp://bit.ly/1IWZxVu

Optics Express journal article that provides overview of the IES method:Development of the IES method for evaluating the color rendition of light sourceshttp://bit.ly/1J32ftZ

Application webinar co-sponsored by US Department of Energy and Illuminating Engineering Society:Understanding and Applying TM-30-15: IES Method for Evaluating Light Source Color Renditionhttp://1.usa.gov/1YEkbBZ

Technical webinar co-sponsored by US Department of Energy and Illuminating Engineering Society:A Technical Discussion of TM-30-15: Why and How it Advances Color Rendition Metricshttp://1.usa.gov/1Mn15LG

LEUKOS journal article supporting TM-30’s technical foundations:Smet KAG, David A, Whitehead L. 2015. Why Color Space and Spectral Uniformity Are Essential for Color Rendering Measures. LEUKOS. 12(1,2):39-50.http://dx.doi.org/10.1080/15502724.2015.1091356

61

Reso

urce

sLEUKOS editorial discussing next steps:Royer MP. 2015. IES TM-30-15 Is Approved—Now What? Moving Forward with New Color Rendition Measures. LEUKOS. 12(1,2):3-5.http://dx.doi.org/10.1080/15502724.2015.1092752

Lighting Research and Technology, Open Letter:Correspondence: In support of the IES method of evaluating light source colour rendition (More than 30 authors) http://dx.doi.org/10.1177/1477153515617392

DOE Fact Sheet on TM-30http://energy.gov/eere/ssl/downloads/evaluating-color-rendition-using-ies-tm-30-15

DOE TM-30 FAQs Page:http://energy.gov/eere/ssl/tm-30-frequently-asked-questions

This concludes The American Institute of Architects

Continuing Education Systems Course