TM-21 Adjust Results (X-Times Rule)
?Some indications from limited 6,000 hr data are not believable –limits on predictions are needed
?Analysis was completed on 40 sets of data to determine model fit uncertainty and relationship to prediction limits
–A 0.40% relative combined uncertainty of the measurement system was assumed–Prediction confidence of 90% applied
–Multiplier (i.e., 6 times) is tested to see if it falls within confidence band of projections at a given sample size
Data Set CCNormalized to 1 at 0 hours1.21.11.00.90.80.7Extrapolation 1,000 to 6,000 hours0.6Hours?Results:
?For sample size of 20, max projection = 6 times test duration
?For sample size of 10 -19, max projection = 5.5 times test duration
20| Solid-State Lighting Program
www.ssl.energy.gov
TM-21 Interpolation of Results
?The format of LM-80 testing (and followed by TM-21) is that testing LED modules at multiple temperatures provides data for matching in-situ performance (i.e., when installed in a luminaire)
?In-situ temperature must be within the LM-80 test temperatures (55°C, 85°C, and user chosen)?Use Arrhenius equation to interpolate between test temperatures
–Arrhenius accounts for the temperature effect on decay rate constants –important for temperature dependant LEDs!–Multi-step process
?Extrapolation outside of LM-80 test temperatures is not recommended or supported
21| Solid-State Lighting Program
www.ssl.energy.gov
TM-21 Interpolation Example
?Parameters of interpolation using 10,000 hours of LM-80 data for in-situ case temperature Ts,i= 70 oC
Ts,1, (0C)Ts,1, (K)α1B1Ts,2, (0C)Ts,2, (K)α2B2Ea/kBAB055328.151.684E-060.963985358.153.354E-060.952526996.283E-039.582E-01Lumen Maintenance [%]90%Ts,i, (0C)Ts,i, (K)αiProjected L70(Dk)Reported L70(Dk)100p343.152.413E-06130,131>60,00080U oC85 oC70 oC55 oC data85 oC data700,00020,00030,000Time [h]40,00050,00060,00022| Solid-State Lighting Program
www.ssl.energy.gov
TM-21 Result Reporting Notation
?Apply consistent “life” notation to results.
General form: Lp(Dk)
p= the percentage of initial lumen output that is maintainedD= the total duration of the test in hours divided by 1000 and rounded to a nearest integer
?Examples:
L70(6k) = 34000 hoursL70(10k) = 51000 hoursL70(6k) > 36000 hoursL70(4k) = 4400 hours
for 6000 hours test datafor 10000 hours test data
for values with the 6 times rule appliedfor value reached experimentally
23| Solid-State Lighting Program
www.ssl.energy.gov
TM-21 Reporting/Documentation
?Report test and
calculation details using consistent and complete format
?When interpolation used, calculation results must also be shown
Ts,1, (0C)Ts,1, (K)α1B1Ts,2, (0C)Ts,2, (K)α2B2Ea/kBAB055328.151.684E-060.963985358.153.354E-060.952526996.283E-039.582E-01www.ssl.energy.gov
?Recommended reporting informationDescription of LED light source tested (manufacturer, model, catalog number, etc.)Sample sizeDUT drive current used in the testTest durationTest duration used for projection Tested case temperaturemAhourshour to houroCTs,i, (0C)Ts,i, (K)αiProjected L70(Dk)Reported L70(Dk)70343.152.413E-06130,131>60,000αBCalculated L70(Dk)Reported L70(Dk)hourshours24| Solid-State Lighting Program
DOE解读IES LM79,LM80和TM21的差异
