42 Partners - Maintenance Factor - LED luminaires - LLMF Lamp Luminous Flux Factor

CIE 97 2005 lists 4 factors that affect the Maintenance factor of an installation
LLMF (lamp lumen maintenance factor) relates to luminaire losses due to burning hours;
LSF (lamp survival factor) relates to luminaire losses due to failures;
LMF (luminaire maintenance factor) relates to luminaire losses due to dirt accumulation;
RSMF (room surface maintenance factor) relates installation losses due to dirt accumulation.

LLMF data is required for CIE 97 Maintenance Factor calculations
see 42 Partners - Maintenance Factor - LED luminaires
Luminaire manufacturers often quote numbers like: >L70 B20 (50k) according to TM21.
These numbers purport to say that the luminaire will still emit 70% of its initial lumen output at 50,000 hours (5.7 years) and that on average, 20% of the sources would be expected to have failed.

Standards for LED Light Sources
There is a standard methodology to provide performance data relating to losses due to burning hours of LED sources driven by laboratory power supplies:
Test: IESNA LM80 - LED Packages, Arrays and Modules
Prediction: IESNA TM21 - LED Light Sources
-----------------------------------------------------------------
TM21 is a method of predicting the Long Term Lumen Maintenance of LED Light Sources from data collected during testing to IESNA LM80 which is the Approved Method for Measuring the Luminous Flux and Colour Maintenance of LED Packages, Arrays and Modules.
These 2 documents together give a test method and a prediction of the lumen Maintenance of an LED source which would go into a luminaire. LED Manufacturers typically collect over 10,000 hours (60 weeks) of data, then use the TM21 prediction method to extrapolate the life out to 50,000 or 60,000 or even 100,000 hours (5.7 years, 6.8 years or 11.4 years), this data is available from the LED supplier.

Standards for complete LED luminaires
There is a standard methodology to provide performance data relating to losses due to burning hours of complete luminaires:
Test: IESNA LM84 - LED Lamps, Light Engines, and Luminaires
Prediction: IESNA TM28 - LED Lamps and Luminaires
-----------------------------------------------------------------
TM28 is a method of predicting the Long-Term Luminous Flux Maintenance of LED Lamps and Luminaires from data collected during testing to IESNA LM84 which is the Approved Method for Measuring Luminous Flux and Color Maintenance of LED Lamps, Light Engines and Luminaires.
These 2 documents together give a test method and a prediction of the lumen Maintenance of an LED luminaire. LED Luminaire manufacturers should carry out LM84 testing over thousands of hours, then then use the extrapolation method of TM28 to give a prediction of the Lumen Maintenance of the luminaire.

Is LM80 and TM21 data relevant to a luminaire?
The LM80 tests for chips and sources are carried out using very stable constant current sources such as laboratory DC power supplies, in defined temperature conditions.
The LM84 tests for luminaires uses the driver in the luminaire under test and should be performed at the maximum case temperature of the LED used in the luminaire.
The introduction to LM84 says
"At the solid-state lighting (SSL) system level such as LED lamps, light engines, and luminaires, other system components, in addition to the LED light sources, also contribute to luminous flux decay and color change over time."
The authors of LM84 do not believe that testing to LM80 and predicting to TM21 can tell the whole story regarding lumen maintenance of an LED luminaire.

Why is there little or no data available to TM28?
At first glance LM84 testing and TM28 extrapolation requires at least 6,000 hours (36 weeks) of testing, during this period the LED and driver in the luminaire would probably have improved by a generation. Some manufacturers give this as a reason for saying that there is no point in luminaire testing because the luminaire will have been changed before test data is available.
This is not really a valid excuse for no testing because users still need this information. Even though the LED source might have been improved, the data on the previous luminaire coupled with the TM21 data for the newer LED could be used to make a useful prediction. This must surely be better than just lazily supplying TM21 data for the chips.
TM28 includes a methodology that uses just 3,000 hours (18 weeks) of test data plus relevant LM80 data to make a Combined Extrapolation of Lumen depreciation in the shorter test time.
This means that if you start testing in June, meaningful data is available by Mid October, surely this is not too much to ask?

Users of LED luminaires should not be satisfied with TM21 data, relevant TM28 data for luminaires must be made available by luminaire manufacturers.