IESNA TM-21 ADD B-2015 Projecting Long Term Lumen Maintenance of LED Light Sources.pdf

1、 Addendum B for TM-21-1: Projecting Long Term Lumen aintenance of LED Light Sources If you, as a user of IESs TM-21-11, believe you have located an error not covered by the folowing revisions, you should e-mail your information to Pat McGillicuddy, pmcgillicuddyies.org or send a letter to Pat McGill

2、icuddy, Manager of Technology, IES,120 Wall St. 17thFloor, New York, NY 10005. Additions will be posted to this list online as they become available. This errata list was last updated October 5, 2015. This Addendum replaces specified sections in document IES TM-21-11 based on the post- publishing pr

3、actice and users inputs. This Addendum supersedes and nullifies the statements made in Addendum A for TM-21-11: Projecting Long Term Lumen Maintenance of LED Packages. 4.3 Luminous Flux Data Collection and Selection Additional measurements after the initial 1000 hours at intervals smaller than 1000

4、hours (including every 1000 hour points) are encouraged. Additional measurements beyond 6000 hours are encouraged and wil provide the basis for more accurate lumen maintenance predictions. Data collected within a 48 hour window of each “1000 hour measurement point”, e.g., from 952 hours to 1048 hour

5、s, from 1952 hours to 2048 hours, etc., are aceptable for use in product lumen maintenance projections. The 48 hour window is also applicable to other intervals smaller than 1000 hours. The exact time designation shall be used in the projection calculation. 5.2.3 Data Used for Curve-fit For data set

6、s of test duration, D, from 6000 hours up to 10000 hours, the data used for the curve- fits shall be the last 5000 hours of data. Data before the 1000 hour reading shall not be used for the curve fit. Data collected within a 48 hour window of each “1000 hour measurement point”, e.g., from 952 hours

7、to 1048 hours, from 1952 hours to 2048 hours, etc. shal be acceptable as each “1000 hour reading” and shall be used for luminous flux maintenance projection with the actual measured elapsed time value. The 48 hours window is also applicable to other intervals smaler than 1000 hours. The actual measu

8、red elapsed time values shall be used in the projection calculation. The data collection points used in the projection calculation shall be equaly dispersed in time (to within 48 hours). No two consecutive data collection intervals after the initial 1000 hours shal differ by more than 96 hours in le

9、ngth. For example, data may be used in the projection calculation which are collected every 1000 hours ( 48 hours), every 500 hours ( 48 hours) etc., but not every 1000 hours and ocasionaly at 500 hours, since this wil give excessive statistical weight to certain data points. For data sets of test d

10、uration greater than 10000 hours, the data for the last 50% of the total test duration shal be used for curve-fit. In other words, al data points betwen D/2 and D shal be used. For example, if the test duration is 13000 hours, use al data points betwen 6500 hours and 13000 hours. If there is no data

11、 point at D/2, then the next lower time point shal be included in the data fiting. For example, for 0 of 13000 hours of data taken every 1000 hours, use the data points betwen 6000 and 13000 hours. 6.0 Temperature Data Interpolation When in-situ DUT case temperature, Ts,i, is different from the temp

12、eratures used for LM-80-08 tests, the following procedures should be used to predict lumen maintenance life of the DUTs coresponding to the in-situ case temperature with the same operational condition (e.g., drive curent). 6.5 Limit for Extrapolation Extrapolation of any Lpvalue above the operating

13、temperature used in the LM-80-08 test shal not be performed. For example, if the highest LM-80-08 test temperature is 85C, an Lpat 100C shal not be extrapolated. If the actual operating temperature is below the lowest temperature that is used in the LM-80-08 test, the reported Lpvalue shal use the l

14、owest temperature in LM-80-08 test. For example, if the lowest LM-80-08 test temperature is 55C, an Lpat 45C shal use Lpat 55C. For interpolation betwen case temperatures with test durations of unequal lengths, in instances where the higher temperature has shorter testing duration, projections shal

15、not extend beyond 6 times the test duration of the higher temperature for 20 or more samples (5.5 times for 10 to 19 samples). In instances where the lower temperature has shorter testing duration, projections shal not extend beyond 6 times the linearly interpolated testing duration for 20 or more s

16、amples (5.5 times for 10 to 19 samples). 7.0 Report The report of lumen maintenance life projection shal include the following information shown in Table 1. The reported L70values shal be rounded to 3 significant digits. and B values shal be rounded to 4 significant digits. Ts,i, (0C) Ts,i, (K) iRep

17、orted L70(D k) Table 1 - Recommended information to be included in the report at each LM-80-08 test condition. Description of LED light source tested (manufacturer, model, catalog number, etc.) Sample size Number of failures DUT drive current used in the test mA Test duration hours Test duration use

18、d for projection hour to hour hour Tested case temperature oC B Reported L70(Dk) hours When interpolation is used, the additional following information shal be presented, as shown in Table 2. Table 2 - Recommended information to be included in the report for interpolation (refer to Section 6 for def

19、initions). Ts,1, (0C) Ts,1, (K) 1B1Ts,2, (0C) Ts,2, (K) 2B2Ea/kBA B0The Calculated and Projected Lp(Dk) are not to be reported in Table 1 or Table 2. These values are set for the purpose of calibrating the calculation tools for the users. Due to the statistical uncertainty stated in Annex D, the cal

20、culated result in hours beyond reported projection limit set in Section 5.2.5 does not have any practical or meaningful value. These Tables within Annex E - Data Test Set for Validation of Calculation Examples are to be replaced with the versions presented in this Addendum B. Table E3 - Least square

21、 curve-fit for 6000 hours LM-80-08 test data at temperature point Ts,1 = 55 oC. Point # Time h ln (Average) xy x y x2 1 100 -0.03252 -32.5 100 -0.0325 1.000E+06 2 200 -0.02850 -57.0 200 -0.0285 4.000E+06 3 300 -0.0387 -101.6 300 -0.039 9.000E+06 4 400 -0.04093 -163.7 400 -0.0409 1.600E+07 5 500 -0.0

22、4573 -28.7 500 -0.0457 2.500E+07 6 600 -0.0468 -281.3 600 -0.0469 3.600E+07 Sums -0.284 -864.8 2100 -0.284 9.100E+07 Slope -3.730E-06 Intercept -2.502E-02 13.730E-06 B19.753E-01 Reported L70(6k) 36,00 Table E4 - Least square curve-fit for 6000 hours LM-80-08 test data at temperature point Ts,2 = 85

23、oC. Point # Time h ln (Average) xy x y x2 1 100 -0.03718 -37.2 100 -0.0372 1.000E+06 2 200 -0.03459 -69.2 200 -0.0346 4.000E+06 3 300 -0.04625 -138.8 300 -0.0463 9.000E+06 4 400 -0.0591 -236.4 400 -0.0591 1.600E+07 5 500 -0.06571 -328.6 500 -0.0657 2.500E+07 6 600 -0.06785 -407.1 600 -0.0679 3.600E+

24、07 Sums -0.31069 -1217.2 2100 -0.3107 9.100E+07 Slope -7.416E-06 Intercept -2.582E-02 17.416E-06 B19.745E-01 Reported L70(6k) 36,00 Ts,i, (0C) 70 Ts,i, (K) 343.15 i5.339E-06 Reported L70(6k) 36,000 Table E5 - Parameters of interpolation using 6000 hours of LM-80-08 data for in-situ case temperature

25、Ts,i = 70 oC. Table E9 - Least square curve-fit for 10000 hours LM-80-08 test data at temperature point Ts,1 = 55 oC. Point # Time h ln (Average) xy x y x2 1 500 -0.04573 -28.7 500 -0.0457 2.500E+07 2 600 -0.0468 -281.3 600 -0.0469 3.600E+07 3 700 -0.04604 -32.3 700 -0.0460 4.900E+07 4 800 -0.05087

26、407.0 800 -0.0509 6.400E+07 5 900 -0.0547 -492.9 900 -0.0548 8.100E+07 6 1000 -0.05182 -518.2 1000 -0.0518 1.000E+08 Sums -0.2961 -250.3 4500 -0.2961 3.550E+08 Slope -1.684E-06 Intercept -3.672E-02 11.684E-06 B19.639E-01 Reported L70(10k) 60,00 Ts,1, (0C) 55 Ts,1, (K) 328.15 13.730E-06 B10.9753 Ts,

27、2, (0C) 85 Ts,2, (K) 358.15 27.416E-06 B20.9745 Ea/kB2692 A 1.365E-02 B09.749E-01 Ts,i, (0C) 70 Ts,i, (K) 343.15 i2.413E-06 Reported L70(10k) 60,000 Table E10 - Least square curve-fit for 10000 hours LM-80-08 test data at temperature point Ts,1= 85 oC. Point # Time h ln (Average) xy x y x2 1 500 -0.

28、06571 -328.6 500 -0.0657 2.500E+07 2 600 -0.06785 -407.1 600 -0.0679 3.600E+07 3 700 -0.07021 -491.5 700 -0.0702 4.900E+07 4 800 -0.0753 -620.2 800 -0.075 6.400E+07 5 900 -0.08251 -742.6 900 -0.0825 8.100E+07 6 1000 -0.07893 -789.3 1000 -0.0789 1.000E+08 Sums -0.4274 -379.3 4500 -0.427 3.550E+08 Slo

29、pe -3.354E-06 Intercept -4.863E-02 13.354E-06 B19.525E-01 Reported L70(10k) 60,00 Table E11 - Parameters of interpolation using 10000 hours of LM-80-08 data for in-situ case temperature Ts,i = 70 oC. Ts,1, (0C) 55 Ts,1, (K) 328.15 11.684E-06 B10.9639 Ts,2, (0C) 85 Ts,2, (K) 358.15 23.354E-06 B20.9525 Ea/kB2699 A 6.283E-03 B09.582E-01