Can we trust relative photometry? Absolutely.
First of all, I agree with Mr. Peifer that relative photometry reports can have significant differences compared to actual application performance. However, his description of relative photometry is inaccurate and misleading. In relative photometry, the light output of the light source is measured separately from the luminaire, but that same light source is then installed in the luminaire and the light output is measured again. The ratio between the light source being in the luminaire and being out of the luminaire is the efficiency of the luminaire. It is a measured quantity, not an assigned value. These measured efficiencies are reliable and are usually accurate and well within 3 percent. Since the light source(s) and any associated ballast(s) are measured within the luminaire, the effect due to heating from the ballast(s) is incorporated in the results of the relative photometry report. However, it is true that this process factors out the effect of the ballast factor as the readings are typically normalized to the manufacturer's initial lumen rating of the lamps in a relative photometry report.
To properly compare relative photometry to absolute photometry, a seasoned professional should know that the ballast factor must be applied. A proper laboratory set-up is required to measure the ballast factor accurately.
However, a rough estimate can be made based on the presented data. In Tables 1 and 2 of his article, Mr. Peifer compared the relative and absolute measurements of a luminaire. If we take the wattage consumptions listed and use a rough estimate of the wattage consumption of the ballast (6 watts), a rough ballast factor can be calculated as follows: overall wattage consumption minus ballast consumption divided by rated lamp wattage consumption, or in this case (82-6)/(3 x 32) = 0.79. If the lumen outputs are compared, the ratio between absolute and relative output is 4309/5686 = 0.76. Hence, the majority of the discrepancy between the relative and absolute results can be explained by applying the ballast factor.
When used correctly, relative photometry can present more useful data than absolute photometry. A problem can occur with using absolute photometry as the measurements are of one sample, and if it is used to evaluate the performance of a full layout of luminaires, there is a danger of over-generalization. The sample may have not been representative of the average luminaire. Ballasts and lamps vary in performance. Relative photometry, by its nature, normalizes the measurements to reflect more accurately the average performance without the influence of lamp and ballast variation. Ballast factors published by manufacturers are representative of the average performance of their product, so applying them to relative photometry reports should produce a fairly accurate prediction of the average performance.
There is a specific reason that LED luminaires are tested using absolute photometry instead of relative photometry. For relative photometry, the light source must be separable from the luminaire. For most LED luminaires, this cannot be done so only absolute photometry can be performed.
Due to misinterpretations by many experts of the use of relative and absolute photometry, there is a growing misconception that absolute photometry is better than relative photometry. Properly applied relative photometry can provide better information than absolute photometry if the luminaire can be tested by both methods.
K. Frank Lin, Ph.D. Lighting Sciences Canada Ltd. Waterloo, Ontario, Canada
Another great column from Don Peifer in the December issue, with lots of really helpful information about absolute vs. relative photometry. There's just one point he made about DOE's "no commercial use" policy for CALiPER testing results that I'd like to clarify, because others in addition to Don may have misunderstood it. The sole purpose of that policy is to prevent manufacturers from using CALiPER test results to either promote their own products or to bash those of their competitors. That's all we mean by "no commercial use." At the same time, one of the primary purposes of the CALiPER test program is to disseminate independent, unbiased product performance information, so sharing CALiPER results is not only permitted, it's actively encouraged--provided, of course, that everyone is informed about the "no commercial use" policy. Many lighting designers, not to mention utilities and energy efficiency programs, are using the CALiPER results to make informed decisions. And they do so with DOE's wholehearted blessing, because that's precisely the sort of use we intended.
James Brodrick U.S. Department of Energy Washington, D.C.
The list of 2009 IES Illumination Merit Award recipients published in the November issue was incomplete. The complete list now appears online at www.ies.org. IES regrets the omissions.