This article was published in the November/December 2013 issue of LEDs Magazine.
Visit the Table of Contents and view the e-zine version in your browser. You can download a PDF of the magazine from within the browser e-zine.
Researchers around the world have raised concerns about the safety of LED lighting on a regular basis, with the focus ranging from the impact of blue light to issues such as disability glare. In a multipart series of articles that concluded in February 2012, we discussed the photobiological safety of solid-state lighting (SSL) and other non-laser sources of optical radiation. This article builds on the prior work and discusses new work being done in standards bodies and research labs centric to the issue of blue light.
The photobiological safety of lamps and luminaires intended for general lighting service (GLS) applications is currently evaluated by implementing the GLS classification criterion of the IEC/EN62471 specification — namely by reporting (but not necessarily measuring) at a distance at which the source produces an illuminance of 500 lx, not less than 200 mm.
This current situation has provided little satisfaction in the lighting industry on a number of points:
- Disagreement over which lamps should be considered in the GLS category (does this include, for example, spotlights or desk lamps?)
- Questions regarding the value of an evaluation at 500 lx, which may not represent a realistic exposure scenario
- Lack of information provided by evaluation at 500 lx, since for the majority of sources an exempt risk group classification is obtained
- Issues with the implementation of a method, provided in IEC/TR 62471-2, to permit the transfer of LED manufacturers' data to finished products. Such an evaluation, based on worst-case conditions (not a realistic representation of the use of the LED), often results in RG2 classification, requiring the use of warning labels, and presents the problem of how this classification should be transferred to finished products.
The photobiological safety panel of IEC sub-committee SC34A: Lamps (part of IEC technical committee 34: Lamps and related equipment) has looked at this question, the result of which includes the publication of IEC/TR 62778 Edition 1: "Application of IEC 62471 for the assessment of blue light hazard to light sources and luminaires," and the amendment of various lamp and luminaire standards, many of which have already been published and updated under the Low Voltage Directive. This approach is in line with the idea of IEC 62471 as a horizontal standard vs. a vertical, product-specific standard.
Safety concerns of lighting products
IEC 62471 gives consideration to two ultraviolet (UV), two retinal, and two infrared (IR) hazards to the skin and eye over the spectral range 200–3000 nm. The optical radiation emitted by GLS products, broadly encompassing incandescent, fluorescent, discharge, and LED sources, need not cover the entire spectral range nor be of a level to present cause for concern. A consideration of photobiological safety depends therefore on lamp type and is treated by technology-specific standards.
|FIG. 1. Retinal image size and irradiance vary with exposure time.|
While the actinic UV hazard has been considered in certain standards, and in guidance on the provision of luminaire protective shields for lamps emitting a high level of UV radiation, the retinal blue light hazard has not been hitherto addressed; it is on this aspect that IEC/TR 62778 has focused. The IR hazard will, where required, be dealt with by marking with labels warning of the presence of IR radiation.
Scope of IEC/TR 62778
IEC/TR 62778 has been written to provide guidance in the assessment of the retinal blue light hazard of all lighting products, emitting principally in the visible region, 380–780 nm, and on the transfer of data from LED/lamp to finished products (taking care that the operating conditions are comparable in both cases).
The relationship between correlated color temperature (CCT) and blue light hazard is also discussed, namely that the more a source emits light in the blue region, the higher the CCT and the greater the blue light hazard posed. Guidance is provided on the use of source luminance, illuminance, and CCT to determine if a source falls below RG1 threshold values, determined by calculation, including a safety factor of two.
Retinal blue light hazard
Evaluating the retinal blue light hazard effectively requires taking account of the irradiance of the retinal image of the source viewed. For momentary viewing, the retinal image subtends the same angle as does the source (Fig. 1). With increasing exposure time, the retinal image is spread over an increasingly large area of the retina due to eye movement (saccades) and task-determined movement, resulting in a corresponding reduction in retinal irradiance. A time-dependent function of the angular subtense of the retinal image, for exposures from 0.25 sec (aversion response time) to 10,000 sec is defined, ranging from 1.7 mrad (taken as the smallest image formed on the retina) to 100 mrad.
In order to determine the irradiance produced on the retina resulting from viewing a source, one must perform a measurement of radiance, which includes a term accounting for the solid angle of emission of the source and a term accounting for the area of the source measured.
The solid angle term relates to the collection of light by the eye through the pupil — which, from the perspective of measurement, can be simply viewed as an averaging aperture. The area term, determined by the field of view (FOV) of the measuring instrument, is fundamental since it relates directly to the area of the retinal image considered. In evaluating blue light hazard, one should consider the exposure time-dependent area of the retina exposed rather than the area of the source. Note that while the FOV is defined as the solid angle viewed by the measuring instrument, it is commonly represented by a planar acceptance angle.
The classification structure of IEC 62471 is based on exposure time before a hazard being posed. One can use this time-basis by measuring the source radiance in the corresponding FOV, to test each risk group in turn to determine if the risk-group emission limit be exceeded or not, starting with exempt and proceeding through the risk groups only where emission limits have been exceeded (refer to table).
In evaluating the blue light hazard, the measured quantity is more accurately termed "physiological" radiance as opposed to the "true" radiance typically encountered in spectroradiometry, which by definition samples only the emitting area of the source (Fig. 2). Where the physiological radiance is measured in an FOV greater than the angle subtended by the source, the result is an average of the true source radiance and the dark background. The law of conservation of radiance, which states that radiance cannot be increased by an optical system, applies only to the measurement of true radiance.
The retinal blue light hazard is determined by measuring the spectral radiance of the source over the range 300–700 nm (applying the blue light hazard weighting function) and over the appropriate FOV for the risk group considered. Extending this measurement to 780 nm allows the determination of luminance (cd/m2). It will be seen that this information may be required in the IEC/TR 62778 analysis.
Blue light irradiance limits
An indirect manner of determining source radiance is to place an aperture directly at the source to define the FOV, and perform a measurement of irradiance at the evaluation distance. Radiance may then be computed from the product of the irradiance and the solid angle subtended by the aperture in steradians (sr). One can in this manner derive from the table irradiance-based emission limits using the solid angle corresponding to the risk-group related FOV.
It is on this basis that simplified irradiance emission limits for blue light small sources are determined in IEC 62471, from the product of the emission limits in the table and the solid angle corresponding to 11 mrad (9.50332.10-5 sr). It is against these limits that sources having an angular subtense less than 11 mrad should be evaluated.
|FIG. 2. The hatched area shows example FOVs of the measurement of "true" radiance (left) and "physiological" radiance (right).|
The angular subtense of a source at a given distance is computed from the ratio of the 50% emission size of the source and the distance (Fig. 3). The reported source subtense is taken as the geometrical average of the maximum and minimum source subtenses, having applied limits (if less than 1.7 mrad, set to 1.7 mrad; if greater than 100 mrad, set to 100 mrad) prior to taking the average.
IEC/TR 62778 approach
Based upon the assumption that light sources classified as exempt or RG1 for blue light hazard are safe, and require no labeling or user information according to IEC/ TR 62471-2, IEC/TR 62778 applies a test to determine if, at a distance of 200 mm (taken by IEC 62471 as the near point of the human eye, the worst case for retinal exposure), the source in question is above or below the blue-light RG1 emission limit. This analysis requires a measurement of spectral radiance, or spectral irradiance where blue light small source conditions apply.
Where a white light source has a true luminance (in an FOV under-filling the source) less than 104 cd/m2, one may directly apply RG0 unlimited classification. This guidance is based on the expectation that, at this level (considered as one visually comfortable to view), emission limits will not be exceeded. In practice, this threshold luminance is particularly low, exceeded by many, even low-power, white LEDs.
|FIG. 3. Determine the 50% emission dimensions of a source to compute angular subtense.|
In the case of sources emitting too much light or heat for the measurement equipment to withstand at 200 mm, another measurement distance may be employed. No clear guidance is provided in terms of suitable measurement distances or FOVs. Where one desires to transfer the analysis from a lamp/LED to a finished product, then the FOV should be chosen to measure the same area of the source that would have been measured at 200 mm and 11 mrad.
Source subtense ≥11 mrad
For a source subtense greater than 11 mrad, a measurement of spectral radiance at 200 mm and in an 11-mrad FOV (a circular area of 2.2-mm diameter at the source) should be performed, and the determined blue light radiance compared with the RG1 emission limit of 10,000 W/m2sr.
Where the result is below this emission limit, the classification "RG1 unlimited" can be applied to component lamps/LEDs or finished products. The RG1 unlimited classification of component lamps/LEDs can be directly transferred to all finished products using this source. One presumes that true radiance was measured, which cannot be increased due to the conservation of radiance.
|FIG. 4. The IEC/TR 62778 procedure for determining threshold distance.|
If the RG1 limit is exceeded, the boundary between RG1 and RG2 should be determined. In the case of component LEDs/lamps, the threshold illuminance, Ethr, at which this boundary occurs, should be included in the datasheet for transfer to the final product. For finished products, the threshold illuminance, Ethr, can be converted to the threshold distance, dthr, at which Ethr is obtained.
Where the measured blue light radiance is below the RG0 emission limit of 100 W/m2sr, the light source is classified "RG0 unlimited." For correct exempt risk group analysis, the 100-mrad FOV should be used; it may be included as an additional measurement should this be required by an application.
Source subtense <11 mrad
For source subtense less than 11 mrad, one should consider the blue light small source case through a measurement of spectral irradiance at 200 mm. The determined blue light irradiance should then be compared with the RG1 emission limit of 1 W/m2.
Where the result is below this emission limit, the classification "RG1 unlimited" can be applied to finished products only. One cannot measure true radiance so the conservation of radiance cannot be used to transfer data from component lamps/LEDs to finished products. In this case, the threshold illuminance, Ethr, should be reported in the datasheet to allow transfer to the final product.
If the RG1 limit is exceeded, the boundary between RG1 and RG2 should be determined for finished products through computation of the threshold illuminance, Ethr, and subsequent determination of the threshold distance, dthr, at which Ethr is obtained.
Computation of Ethr
The threshold illuminance, Ethr, at which the boundary between RG1 and RG2 is to be found, may be computed by considering the irradiance-based emission limit for this RG1 measurement in an 11-mrad FOV. From the blue-light-hazard RG1 emission limit radiance (10,000 W/m2sr) and the solid angle corresponding to an 11-mrad FOV (9.50332.10-5 sr), one obtains the blue light irradiance emission limit of 1 W/m2, which value is already provided for sources having a subtense less than 11 mrad.
Now, the ratio of the blue light hazard quantity to the corresponding photometric quantity is a constant, defined by the spectral distribution and not the absolute measured quantity. This parameter is defined in IEC/TR 62778 as KB,v.
The ratio of blue light radiance to luminance (or for small sources, blue light irradiance to illuminance) will be equal to the ratio of the blue light irradiance emission limit (1 W/m2) to the illuminance at which this threshold is obtained, Ethr, in lux. It follows that Ethr may be computed from the ratio of luminance to blue light radiance in an 11-mrad FOV for sources subtending greater than or equal to 11 mrad, and the ratio of illuminance to blue light irradiance for sources subtending less than 11 mrad.
The threshold illuminance, Ethr, representing the boundary between RG1 and RG2 can be converted to a threshold distance, dthr. It is valid to determine this parameter for finished products only. The region between dthr and the source is classified blue light hazard RG2; elsewhere RG1 unlimited is the correct classification.
The method recommended by IEC/TR 62778 to determine dthr is to use goniophotometric data of the test product if available (Fig. 4). Knowledge of the maximum luminous intensity, Imax (cd), and the inverse square law allows one to calculate dthr from Ethr, where dthr = √(Imax/Ethr). This procedure ignores the fact that for many sources, the inverse square law will not be applicable, particularly for directional sources, as are many GLS sources.
No guidance is provided in those cases where goniophotometric data is not available. It is reasonable to use an illuminance meter to find dthr. This method does not rely upon questionable use of the inverse square law; one might suggest that it would be more reliable.
In both cases, if the source subtends less than 11mrad at the determined location of dthr, one can state that this value of dthr is incorrect, since in its determination, Ethr was not sought in the limited 11-mrad FOV. It follows that an overly-conservative threshold distance will be determined. A more appropriate method of evaluating arrays is under consideration for implementation in future revisions of the technical report.
The introduction of IEC/TR 62778 has simplified evaluation of the retinal blue light hazard of GLS products, although further guidance on the determination of a more realistic value of dthr may be important. Given the fast pace at which SC34A has worked to publish IEC/TR 62778 and amend lamp and luminaire standards, it should not be long before we start seeing the reporting of Ethr and dthr in product literature.