Comments and Commentaries a.k.a. BLOG

I never liked the term BLOG - but it's here to stay. I have written a lot of material for various discussion groups and some of these articles have proven to be somewhat popular - even argumentative. A lot of people have asked that we resurrect some of these. Since these articles helped form the raison d'être for our company, readers may like to know "where we came from".

Lighting Bylaw - Most Important Limits What is Colour? What's Low-Impact? Products Lag Science Do Plants Like Light? "Needs" Versus "Wants" Liability Sustainability? Full Cut-Off Shielding is Important Seeing Red when out at Night Something Called Glare Why Paths Glow Qualified "Truth" - Reaction Time Qualified "Truth" - Colour Vision Qualified "Truth" - Distinguishing Colour

Qualified "Truth" - LED Spectra Qualified "Truth" - Mesopic Vision Qualified "Truth" - Night Vision How Good is a Luminaire? Bright Patch Under Lamps Photometric Plot Simple Shield for CFL or LED Coach Light Fixtures What Makes a Good House Light? Filtering White Light How Bright a Lamp Should I Use? Require Proof before we Change? Light Across the Water

Lighting Bylaw - Most Important Limits

Most reductions in light pollution and nuisance lighting can be achieved with minimizing glare. (It is hard to argue for an increase in glare.) and, to not allow light trespass. (It is hard to argue that a manager or landowner should impose their illumination on neighbours.) The counter argument is that suitable luminaires are not available, but in fact they have been around for over 1½ decades. Since LEDs are the "new thing", there should be a hard limit for correlated colour temperatures of no >3000K (2700K ± 300) for future lamps. This minimizes low ecological and human health impact. (It is hard to argue against human health.) By the end of the summer 2015 all major producers will have CCT of 3000K or less as an option (emitting <10% of light <500 nm - the same a HPS). And the IDA Dark Sky Friendly designation will be in full effect. (They posted this requirement at the summer's end of 2014 and gave companies 1-year to comply, they must removed this popular marketing attribute from all literature.) Illumination levels should also be reduced, but it is not easy to defend a lower than IESNA recommendation. (However, some cities have decided to illuminate at about 1/2 IESNA levels when they use Full Cut-off luminaires - with no apparent increase in the rates of crime or accidents.) However, if the glare (<10% of the light in the glare zone) and light trespass (beyond property boundaries) criteria are followed, the impact of high illumination will be on the cost of electricity and the cost of large expensive light fixtures, which should be a good deterrent.

What is Colour?

In my studies over the years I find it amazing how different people can be. They differ in more profound ways than just ethnicity. Our differences are more than skin deep. However research only becomes practical when you test the impact of only one variable at a time. This ensures we test a single phenomena, but it may limit the applicability of the results. And, practical constraints that limit the extent of the study may invalidate some results. Here is an example. Our daytime vision uses three types of “cone” cells – one for each of the three basic colours (blue, yellow and red). Each cell of these three receptors has the same sensitivity to light, but that does not mean we see light the same. Our sensitivity to colour is a function of the “number” of each type of receptor. Our sensitivity to blue light is low because they make up only 6% of the total “cone cells”. However, yellow and red are more problematic. The ratio of yellow to red cells can vary between individuals by, at least, 1.5:1 to 1:1.5! A number of published works state that we are 2X more sensitive to red than yellow light. This seems to have been due to earlier studies with a limited number of subjects. Personally, I have trouble seeing with red light. As an amateur astronomer, I was always told to use red light at night, so I did – even though I couldn’t see very much. I now know that I am one of those with relatively few red receptors. A lot of studies are based on undergraduate university students who volunteer to be test subjects to help earn a credit or extra money, and it is VERY difficult to get more than a dozen or so students. (I was one of them.). (This biases the results to young, smart and relatively healthy individuals – not the average human.) So in many (?) cases the research results may not reflect humans as a whole and “common sense” that has developed over generations should carry more weight. (My grandmother always said that the sleep before midnight was the most important. We now know why.) [Modern “Big Data” sets may help with this as long as the test protocols are identical.]

However, we have also learned that just because “a lot of people” are fine with X-lux levels or X-spectrum of light does not mean that it is all right to impose these conditions on others. The wisest course is to minimise the disruption to the natural world – fore which we have evolved – a Hippocratic Oath for social planners. Regarding a definition of a “disturbed” circadian rhythm (CR); a shifted phase can still have a significant impact if it puts the body out-of-step with constant social schedules and the natural environmental cycles. The ramifications can be severe depending on the behaviour or activity and the amount of the CR phase shift. A simple example most of us have all experienced is staying up late before a morning exam, or jet lag.

What's Low-Impact?

I define “low impact” as that which has minimal impact on wildlife and human health. We did a lot of work studying the effects of natural and artificial light at night. Interestingly – even the light of the Moon affects plants and animals, but with the Moon, these effects are cyclic whereas artificial light is constant. High-impact lighting affects almost all life including humans simply because the ecosystem evolved over the eons to tolerate and even exploit the anonymity provided by darkness. Changing the night environment alters the cues to normal behaviour and even the biochemistry of most life forms. Low-impact does not mean no-impact, so care must be exercised when using light in an otherwise natural setting. The attributes of light that affect the ecosystem are the spectrum, brightness, extent and timing. I have published a few works on this. The Lighting Research and Technology Journal has one of them (Lighting Res.Technol.2014;Vol 46:50–66).

We experimented with a “compliant” luminaire and found that it provided remarkably good visibility. We now sell these lights to ecologically sensitive areas that must cater to late-night visitors. Unfortunately, city managers have been told that they “must have” CRI>80 and “at least” IESNA illumination levels. They also think that the glare that results from most LED fixtures is “necessary” for visibility. In a limited number of situations this may be true but there are a lot of “half-truths” being spread around that confuse the uninformed.

Products Lag Science

Humans are well adapted to the environment so problems can result if we change it unless we are careful. We now know enough to do much better, but this requires a change in product promotion, which in some cases may demand "back-tracking" our marketing mantras. The promotion of white light luminaries began with the desire to more effectively display colours and provide daytime "quality" lighting throughout the night. This trend has continued with the marketing of white CF and LED products. However, the industry's marketing machine is slower to react than the research and engineering departments, and this can result in "bad press" and potential loss of business as consumers learn more about the adverse impact of lighting products. White light certainly shows off an automotive retail lot, but this "punch" is becoming subdued with the more ubiquitous use of white light. As more cities convert from an amber high-pressure Sodium lamps to white light we are loosing the colour contrast that is so important for advertising and urban navigation. Within the last decade significant advances in medical knowledge have revealed the adverse affects of short wavelength light at night on human and animal health. However with the industry's focus on white, the current product lines provide no options to avoid these effects. It seems very few industry leaders know of the recent discovery of photochemical image of light on our health. This knowledge results from the discovery of detectors in our eyes that directly affect the nocturnal timing of our body functions. They are primarily affected by short wavelength (<500 nm) light, which makes light look white. In speaking with the industry, LED light sources can be made to reduce the light's impact on our health, but these are not made available by most companies. PRomoting the non-white "amber" light would counter the marketing thrust for white outdoor lighting. The CSbG  EcoLight  is the only readily available alternative to white light luminaires that also provide the other attributes of low-impact illumination.

Do Plants Like Light?

Plants evolved to expect the spectrum of sunlight, though there are more specific colours that can do the job almost as well – but this is getting into the more complicated study of a plant's action spectrum. More important is the timing, or schedule of the light and what type of plant is being grown. Are the plants short-day, long-day or day-neutral? We usually assume plants want what we what. But in fact they don't. Indeed, what we want may not even be good for us? A lot of research has been done on plants and the impact of nocturnal lighting. In some cases artificial lighting is used in green houses to enhance growth. Trees determine the season by the length of the night. If the night is short, then it must be summer. Artificial light at night (ALAN) in autumn “fools” the tree into thinking it is still summer (short dark period) and this can delay or inhibit its preparation for winter - leading to stress or death. The CSbG   EcoLight   has a much lower effect on plants than white light at night. Its spectrum avoids the short wavelengths that partially mediate the plant's response to light.

Of course this has to do primarily with temperate zone plants. Some plants are insensitive to ALAN and are not affected by ALAN. It gets down to the climatic zone and whether the plant is a “long-day plant, a short-day plant, or a day-neutral plant (you can Google these terms). The key point is that all life has evolved with a dark night. In the last century we have begun to illuminate the night and this has fundamentally changed the environment. The vegetation we see in a city is quite different from what we see in the country. We are only now (the last few decades) started to study and discover the adverse side effects of artificial light.

"Needs" Versus "Wants"

Our society permits and even encourages citizens to do some “crazy things” that are well known to be really bad for us (driving cars fast, smoking, drinking, eating a lot, and so on). They are permitted because people choose to do them (it is a democracy). However as professionals, we should help encourage our clients to take a more constructive route. Just because we “want” something, does not mean we “need” it. Having kids teaches this lesson very quickly! And so it is with outdoor lighting. Just because some citizens want to flood light their property all night long - regardless of its affect on their neighbours, does not mean that bylaws should permit it. The same argument applies to commercial lighting and municipal lighting. People point to safety and security to justify the "need" to illuminate large areas but they ignore the adverse side effects, and selectively ignore published research to the contrary. Without an active security system, outdoor lighting does not protect your property; rather it puts it on display. Similarly, a pedestrian standing in a patch of light is not safe, they are on display and blind to what lurks in the shadows. The CSbG  EcoLight   is well shielded to prevent light trespass. This also prevents the glare that can prevent people from looking "beyond" the light into the shadows.

Liability

Liability and litigation are often used to justify current high-impact lighting policies. What surprised me, and others who have looked into it, is the lack of any litigation caused by lighting issues – even when it is attempted in association with a wider claim. It is not surprising this is the case since we are inundated in the press with general “reports” about litigation. The law is also a rather complicated issue since there are different limits: negligence and gross negligence, nonfeasance, misfeasance and malfeasance, Governmental Immunity from Tort, the Anns Principle, the difference between “policy” and “operational” arguments, the weight of AASHTO guidelines, and so on. (I apologize for mixing jurisdictions, but this helps to make my case). Even what we may consider to be common sense liability is complex since there are overriding laws that absolve actions of governments (federal and local) from liability. Our research has failed to find ANY successful lawsuit involving lighting – in Canada or the USA. Even a long burned out streetlight at the scene of a fatal accident involving a pedestrian did not result in the city from being found at fault. One general conclusion is that if a lighting policy is in place, and it is reasonably followed, the city is not liable. This includes illumination at levels significantly below that of the IESNA guidelines.

Sustainability?

This was in response to a posting in a "discussion" forum. Although this seems to be a marketing post, I feel I must point out the shortcomings of most current luminaires. ALL manufacturers have focused on lumens/watt, but the quality of the illumination has suffered because of it. Many of the older luminaires (> 15 years old) have relatively low efficacies - using more electricity to produce light than more modern fixtures. So you will save electricity, and improve sustainability only if you replace very old lamps with LEDs. However, LED luminairs do not do so well against new HPS luminaires. Also, the white light is the most impactful spectrum you can use at night. It has been documented to degrade the ecological balance and has been shown, at one extreme to reduce human health, and at the other extreme there are very strong links to various aliments including hormonal cancers. So users may save a few dollars in electricity but at significant environmental and social cost. This is not “sustainable” in the general definition of the word. I am familiar with a few XXXX Inc. products. The quality of the illumination on two applications was significantly REDUCED when they replaced the existing HPS luminaires. There was a significant increase in the glare and light trespass due to poor shielding. And the increase in illumination level actually increased electrical usage. Although other manufacturers are offering lower CCT LEDs, XXXX Inc. continues to insist on 4000K – claiming that lower temperatures “look yellow”. They definitely do not. You have to go to a CCT in the low 2000K range to begin to see “yellow” at typical road illumination levels. So although “sustainability” is a honourable attribute, marketing “spin” has corrupted its true meaning.

Full Cut-Off Shielding is Important

Roadway luminaires are not in the centre of our vision but they still compromise visibility because of off-axis light scattered in windshields, eyeglasses and especially within our eyes. Even light sources more than 30-degrees from our viewing axis compromise our vision due to these effects and bleaching our rod cells. Luminaires that use the ovoid lens to scatter light create glare in this way. The Full Cut-Off (FCO) luminaires are MUCH better. However most LED fixtures can be worse than the older HID FCO luminaires. On several demonstration projects the glare from the LEDs was, in my opinion – terrible. And, these were not cheep luminaires! Although the manufacturers claimed they were FCO, the amount of light that was emitted in the 80 to 90 degree from nadir glare zone seemed to be much more than the 10% limit. This may have been due to the high perception of glare that is caused by white light, or the data was “massaged”. I understand that luminaires approved for Hawaii have much less than 10% in the glare zone. Better shielding can be achieved with Sharp Cut-off with only 1% of the light in the glare zone, and our luminaire has this shielding. (Virtually no light shines beyond the target area.)

So proper shielding can be incorporated but the metrics used to rate luminaires do not adequately take good shielding into account. Manufacturers with good shielding for their luminaires cannot compete in the typical urban market because of “hardwired” requirements for higher-glare fixtures. The best some of us can do is to demonstrate our products and let municipal engineers judge if their city would benefit from less glare.

Seeing Red when out at Night

Astronomers have always used red light when they needed to see better at night without undermining their night vision. This worked well with red-filtered incandescent lamps, but the new red LEDs work very differently, and this age-old advice is no longer helpful. Incandescent bulbs emit a broad range of wavelengths beginning in the blue and increasing across the visual spectrum through the red and into the infrared. Filtering out what little blue light there was still a broad enough spectrum to be seen by most of the daylight colour receptor cells in the retina – the L-cones (red sensitive) and M-cones (yellow sensitive). These account for about 94% of all the cone-cells in the retina and provide good visual acuity for reading. Some people have more L-cones than red-cones while other people have the opposite mix. Red LEDs emit a narrow band of red light that can be seen by only the L-cones. This reduces our visual acuity to less than ½ that for filtered incandescent light. Amber LEDs can be seen by both the L- and M-cones and so preserves our visual acuity. Our sensitive night vision cells also detect this light, and it will affect our night vision, but this is also a good thing. Much less light is needed to read and recognize shapes with amber light than with red LEDs. You may read more about this in a paper I published in Sky and Telescope Magazine for June 2016.

Something Called Glare

Glare from a light fixture is inevitable but understanding where it comes from can be helpful in reducing its impact. The key point is the need for good optics. Without reasonable optics, the light distribution is lambertian (it is bright perpendicular to the fixture’s window, which then decreases closer to the horizon), and the 1/r2 decrease in the illumination on the ground with distance. Optics can at least partially compensate for these two terms. In my experiments, I identify two regions in the illumination pattern, which are somewhat defined by the location of the observer. One involves the luminance (brightness) of the luminaire and the other the luminance of the ground. With the observer in the main illuminated area (target area), the lamp of the luminaire is visible and produces a lot of glare. This light is illuminating the ground. As you walk away from the luminaire, eventually the sharp cut-off of [my] luminaire hides the lamps and reduces the glare to nearly zero. However, the light that shines “below” my eyes continues out to illuminate 25% to 50% farther from the luminaire. The ratio of the "glare distance" to "illumination distance" is related to the height of the observer's eyes, and the mounting height of the luminaire. The glare will usually be less for a walking pedestrian than a sitting motorist. Either way, the shielding of the luminaire should put this glare as far from the “centre of our field of view” as practical. Another contribution to glare is the apparent size of the light source. Spreading out the light in the fixture and thereby reducing the apparent intensity of the source will reduce the visual impact of the glare. This can be done with well designed diffusers. Instead of seeing a single bright spot or array of bright spots - as produced by typical LED luminaires, the observer would see a glowing patch. So this suggests a different perspective. If you are in the target area, then you will see the lamp and experience glare. The key point in glare control is to ensure that the lamp is only visible from the target area and not many “pole heights” down the road or second story bedroom windows, where the contribution to illumination is minimal and the glare approaches the centre of our vision.

Why Paths Glow

We are told, or it has been implied, that we need artificial light to see at night. Indeed we see people with flashlights walking along park pathways. Surely they won’t be using flashlights if they didn’t need them. Well most pathways require us to carry flashlights, or they are designed to require flashlights. Or the city runs electric power to the park and installs artificial lighting. This is based on “Best Practice” that is followed by most cities. But let’s consider what’s going on. Perhaps there is a “better way”. A “Best Practice” for one situation, based on a limited point of view, may then necessitate a requirement that may have been avoided if better judgement was applied in the first place. Communities like pedestrian and bicycle pathways. The “idea” encourages people to get out and walk around. They may be lined with vegetation for aesthetic reasons and the surface paved with asphalt to produce a smooth surface. Asphalt is both cheap (roughly $40/m2) and easy to maintain. This seems to work well during the day when there is plenty of light, but what about at night. Most urban paths are not planned for night use. Consider the asphalt surface. It only reflects 5-10% of the light that illuminates it. It has roughly the same brightness as the grass beside the path. So, without a flashlight, it’s easy to step off the path and into difficulty. Perhaps this is why people use flashlights. But let’s consider another option. In the figure above we see the reflectivity of various surfaces and plants. Green surfaces reflect a bit better at wavelengths around 0.5 microns, so “green things” reflect more green light than the other colours. However we can’t differentiate colour at night, so the grass and the asphalt will look about same – and we run off the path into the bushes.

However if we coated the path with crushed stone (limestone) then the path surface will reflect about 10X more light (moonlight on rural paths and sky glow on urban paths) than the surrounding vegetation. The path will appear to “glow’ and no artificial light needs to be installed. This will save a LOT of tax payer’s money! Even if the grass is not healthy, the stone covered path will still be 2x brighter than the brown grass. The crushed stone surface will appear brighter than even a concrete surface. When presented this way, it’s common sense NOT to pave paths. The crushed stone will let rain water drain through the surface so it won’t pool, and it’s cheap (<1/2 the cost of asphalt) and easy to maintain. Asphalt is a cheap by-product of oil production, but using it on pathways leads to more expensive-in-the-longer-term infrastructure.

Qualified "Truth" - Reaction Time

Truth: White light supports faster reaction times than our more sensitive, but colour blind, night vision. However …. the typical reaction time during the day is about 1/6-second, and that for our rod vision can range from ¼-seconds to perhaps 2/3 seconds. But it is well known that the reaction time for typical motorists is about 2-seconds. Why? Because the visual clutter along a roadway is a distraction, and delays our reaction time reaction times. If the outdoor lighting illuminates more than the roadway, the visual sensory overload can reduce our effective reaction time and therefore the safety of a street. This means that the brightness and colour of roadway lighting has only a small effect on driver reaction times, and by implication, roadway safety.

Qualified "Truth" - Colour Vision

Truth. A relatively new tool in the quiver of lighting design is “small object visibility”, or how well you can see a small object in your line of sight. More light improves the chances of seeing small objects. To “see” something, you must be able to resolve it. During the daytime we use our three sets of cones, which are sensitive to blue, yellow and red light (S-, M- and L-cone cells respectively). Further, broadband light (white) allows us to use all our full-set of cones, which maximises our resolution. Narrowing the range of colour reduces the number of cones and reduces our ability to resolve detail. However, …. although our perception of “white light” requires the use of all our cones, there are very few blue sensitive cones in our eyes (about 6% of the total). They are widely distributed across our retina – providing poor resolution in the blue. Therefore in so far as resolution of detail is concerned, we can ignore them. Also, our lenses do not focus blue light very well, so the blue image is blurry.

This leaves the yellow and red sensitive cone cells, which our lenses focus well. Light that is detected by only these cells looks “amber”, but provides similar resolution because they compose about 94% of our colour vision. The luminaire “energy”, which is used to create the blue light, is under-utilized. Further, the blue light undermines our night vision that would otherwise help us see into the periphery.

Qualified "Truth" - Distinguishing Colour

Truth: Only broadband (white) illumination will allow us to see subtle differences in colour. The Correlated Colour Index (CRI) characterizes the illumination we require to differentiate colours. Sunlight and halogen lamps have a CRI=1, and high-pressure sodium lamps have approximately CRI=0.2. Typical white-light LEDs provide a CRI=0.8 or better. However, …. the CRI characterization of lamps is done at over 1000-lux (IESNA Handbook, Chapter 4). We depend on bright illumination to “see” colours because we need to use our three sets of daytime cone cells. But at the average brightness of a residential street (<10 lux, which is still about 100X the illumination of the full Moon) our cone vision does not respond very well to light. Fortunately our vision is complimented by our much more sensitive night vision (rod cells). Below roughly 10-lux, our cone-vision is failing and the equivalent CRI falls from 0.9 to very roughly 0.3.

This is only approximate because of the colour sensitivity of this effect (Chin, et.al., 2004) (Figure Chroma vs Illuminance.jpg). At these light levels colours become “desaturated” and much more difficult to identify. At typical urban illumination levels at night, we are unable to take advantage of a lamp with a CRI=0.8. Indeed, my own observations suggest that our eyes respond to 1-3-lux white light as though it had a CRI of <0.4. Therefore high-CRI lamps provide little real benefit at typical roadway illumination levels. The electricity that is absorbed to produce the high-energy blue light is wasted.

Qualified "Truth" - LED Spectra

Truth: White light appears brighter than light of longer wavelength, so less light can be used for the same apparent brightness. This reduces energy use. However, ….there is one design and two biochemical mechanisms that undermine any benefits. The spectrum of LEDs is composed of a large broadband emission between 500->650 nm, which appears amber. This amber is combined with a strong narrowband emission of blue light around 460 nm that will make the colour look white. Cool white light contains a brighter blue spectral component than “warmer” light. A white LED has a very different spectrum than the smooth spectrum of a hot incandescent lamp. Instruments that measure light of most luminaires is sensitive to our daytime vision – an anomalously strong blue emission, will not be properly represented in the measurement. However designers base illumination on the assumption that the spectrum resembles a smooth solar or incandescent bulb spectrum. So white illumination will be under rated. There are very few blue-sensitive cone cells in our retina (only 6% compared to 30-60% for the red and yellow sensitive cells, respectively), so our brain applies greater “weight” to blue signals from these cells to give us the “impression” of white light.

Experiments indicate that blue light has roughly 10X the “glare impact” as amber light, which is detected by our other cone cells. If white-light LEDs are used, our luxmeters will read a reasonable illumination level but our brain will perceive a much brighter surface.

Qualified "Truth" - Mesopic Vision

Truth: By using both our daytime cone vision and sensitive night rod vision we can see "beyond" the illuminated area. The illumination level that allows both our cones and rods to be used is called Mesopic vision. The illumination level must be bright enough for our day vision, but not so bright that it bleaches the rod cells of our night vision. However, ..... the mesopic vision occurs for illumination levels roughly around 1-3 lux, and it depends on the colour or spectral content of the light. Our cone cells adapt to the bright light much faster than our rod cells adapt to faint light. The speed of this adaptation to the new illumination level also depends on the brightness of the pre-exposure. This hysteresis puts our sensitive night vision at a disadvantage. The natural adaptation to light has evolved to accommodate the fading, and brightening of twilight. The illumination during evening twilight fades to about 1/2 every 5-minutes. Our rod cells adapt at about this speed. On the other hand when walking out of a lit home into the night, or walking from a brightly lit road on to an unlit road produces a transition that is too fast for our rod cells to follow. This leaves us temporarily blind.

Our cones begin dark-adapted when we walk from the dark outdoors into a lit room - though they are still not sensitive enough to see outdoors. The bright room rapidly bleaches the rods, and the cones respond quickly by partially bleaching to compensate for the brighter surroundings. So, our eyes rapidly adjust to the light. When the illumination is in the mesopic range, the rod cells will be bleached faster than they can recover from the exposure. Therefore the rod cells will be rendered progressively blind, leaving only the cone cells to support our vision. Therefore the true mesopic range does not last long. For example if our eyes are dark adapted and a light is turned on corresponding to the mesopic vision, we will initially see into the dimmer areas, but this ability will fade over less than a minute.

Qualified "Truth" - Night Vision

Truth: Our daytime cone-cells give us the ability to differentiate the colours of the day. But our sensitive rod-cells are colour blind because they don't let us distinguish colour at night. However, ......although the illumination by daylight provides a good colour balance, the sky at twilight is dominated by blue light of the sky - since the "yellow" Sun has set. The predominance of blue exists even when it is cloudy, because the top-side of the clouds are illuminated by the blue sky. We don't notice the blue cast of light because our brain has the ability to correct the colour, or "white balance" the scene.

Our night vision evolved to help us see during the time of fading twilight after the cone-cells are no longer effective - due to the lack of blue sensitive cone-cells. Our rod cells are sensitive to the blue, green and yellow colours. The rod cells are therefore more sensitive to the blue twilight than the cones. They are also more sensitive to, and easily blinded by, the blue-light components emitted by white LEDs.

How Good is a Luminaire? (added Oct.2017)

We can determine how well a luminaire will perform by inspecting its photometric chart. But these can appear confusing because two sets of formation are combined onto a single sheet of paper. Let’s explore what is plotted on a photometric chart. This chart is for a GE EVOLVE Roadway luminaire (ERL1_02C3227-120V, 32W and a CCT of 2700K). The polar coordinates are the directions the light is emitted. Downward is to nadir (under the light). The circular lines are the luminance scale (cd/m2). You can see that the peak luminance is away from nadir and is about 800 cd/m2. This luminaire is Full Cut-off because there is no light shinning directly above the horizon and into the sky, AND <10% of the emitted light shines within the zone of 10-degrees below the horizon. The extended lobe to the right and left represent the light that is “thrown” far from nadir. This prevents an overly-bright patch under the light. Blue Line The blue line represents the luminance emitted from the luminaire as seen from various places ALONG the road. The farther you are from the nadir; the “brighter” the light will appear - because more light is need to illuminate the road far from the light. More light is needed in periphery because the illumination falls off with distance from the light. This process is described in the Blog post “Bright Patch Under Lamps”. Red Line The red line represents the distribution of light across the road. The light that shines behind the luminaire is backlight. It is virtually impossible to prevent some backlight due to the light scatter by imperfect optics. The lobe of light partially levels-out the illumination. But if the lobe is too high, it causes light trespass into homes across the street. Amber Line If you want a uniform light distribution, you need to throw a LOT of light into the periphery. However you should expect to see a fall-off in the light beyond the distance equivalent to the mounting height. Good luminaires do their best to extend this uniformity, but there is not much they can do without multiple sets of optics as is done by some LED luminaires. They group LEDs and use lenses that preferentially focus light into the periphery.

Bright Patch Under Lamps (added Oct.2017)

Why do we see such a bright patch under a light? The trivial reason is because that is where the light shines. But there is some simple (i.e. common sense) physics involved. The first is the fact that the farther you are from a light - the dimmer it looks. This is because the illumination spreads out as it covers a larger area. If you double the distance (x2), you increase the illuminated area by 4x (r2). So the illumination is diluted to 1/4 (1/r2 - the inverse square law).

Also, the angle that the light hits the ground becomes shallower with distance. Consider sunlight. As it rises in the east, the ground is not illuminated as much as it is at noon when it's high in the sky. At sunrise the light is hitting the ground at a glazing angle (low angle). The brightness is actually a function of the angular distance between the zenith and the Sun - a cosine function to be precise. So, it's called the Cosine Law. Another techy-term for the variation in illumination with this angle is the "Lambertian" distribution. The Cosine Law is compounded because the window of the luminaire is seen from the side to also be foreshortened - giving a cosine2 effect. These phenomena work together to cause a rapid fall-off in illumination from nadir. Without clever optics, a simple lightbulb can only illuminate an distance of about 2X its mounting height from nadir (below the lamp). Beyond that, the surface is significantly dimmer than the "glare" at nadir. This is farther than a flat-glass luminaire, but you pay the piper with considerable glare. However with good optics you can "flatten" out this distribution - but only to perhaps 2-3X the mounting height from nadir.

Photometric Plot (added Oct.2017)

Another way to present the characteristics of a luminaire is with a photometric plot that shows the distribution of light that shines on the ground. This is an illuminance chart (lux). This differs from the chart above (How Good is a Luminaire?) in that the one above shows how much light shines in various "directions". The difference seems subtle - but it is important. This figure shows a plan view - looking down towards the ground from above the luminaire. The street-side is to the right and the house-side is to the left. This is a "Type III" light distribution because it preferentially shines the light over the street and down the road. For this plot, the luminaire was set at 6m above the ground and the grid lines are 6m apart, or one mounting height. The isophots (lines of equal brightness) form concentric circles around the nadir that is directly below the luminaire.

Close spacing of the isophots indicate a rapid fall off in illumination with distance. Fixtures with very simple optics have circular isophots indicative of Lambertian distributions. The luminaire used in our example here "throws" the light down the road to increase the uniformity and reduce the number of lights. The small isophot in the middle indicates a bright patch at nadir, which can be quite distracting. Indeed, it can be so bright that it is the second source of glare after the lamp. However in this case, the maximum brightness within the inner circle is 12-lux, or only 20% higher than the isophot surrounding it. For some fixtures it can 2-3X the inner isophot, which is quite bad. There is some light that shines behind the light on the house-side. Unless there is a sidewalk behind the light that needs to be illuminated, this contributes to light trespass and is wasted energy. If the light shines too far over the street, onto neighbouring properties it will also be light trespass. The values of the isophots are also important. The difference in brightness between the inner and outer isophots should not be more than a factor of 10. The outer isophots that are less than 1/10 the inner isophot provide no useful light, because their dim contribution is overwhelmed by the inner brighter area. Sometimes these are shown to give the impression of a wide area of illumination. However, this extended pattern is actually indicative of glare at a distance. This is because, for the light to reach that far, the lamp must be visible and the lamp produces the glare and light trespass. There should be a rapid fall-off in the illumination pattern at the edge of the area that is intended to be lit. There is a good reason for a "graceful" fall-off in illumination along the road. This allows two adjacent patterns to overlap that improves the uniformity of the illumination. However if the patterns extend farther than the adjacent pole, the contribution will be too dim to be helpful. And, the fact that the lamp shines that far means that it is visible at a distance, and thus produces glare down the road.

Simple Shield for CFL or LED (added Dec.2017)

Incandescent bulbs are rapidly being replaced with Compact Florescent Light (CFL) bulbs. This is good for saving energy, but there is a significant cost to the ecology. A CFL emits considerable blue light into the night environment - up-setting the ecology.	There is an easy "fix" for this that takes advantage of the low-power consumption of CFLs. It will limit the contamination to the immediate area around the lamp.
This pattern can be traced onto cardboard, bent round, glued or taped and painted in about an hour - a good craft for kids. The low temperature of a 13W CFL prevents any fire hazard. An improvement is to replace the white-light CFL with the amber "Bug Light" version, which is readily available on most hardware stores during the spring and summer months.

Coach Light Fixtures (added Dec.2017)

Coach Lights are essentially architectural fixtures often marking the location of a laneway. They provide very poor ground illumination and considerable glare along the road. If marking your laneway is your goal, a low wattage CFL bulb (13 watts) can be used. But we usually want the entrance road to be illuminated too. So how can we modify the existing fixture to minimize glare yet still shine sufficient light down onto the ground? Coach Fixtures are usually selected on what they look like in daylight. So any modifications should not be too visible during the day. This model has coloured windows to reduce the glare from pure white light. Although red light is better for astronomers, neighbours may not like it. Amber is a good compromise. Here is one way it can be done using a thin piece of aluminum and two extension light sockets. Typical Coach Light             Metal Shield in Top of Fixture   Socket Extension for Bulb       The metal shield is cut to fit behind each window in the top of the fixture. I suggest mounting it inside so that the modification is not apparent. Aluminum foil will work but it may have to be replaced every few years as it gets weathered. These shields (only one is shown in the picture) prevent light from shining directly into the eyes of motorists and pedestrians. The windows in the lower half of the fixture allow the light to shine down onto the laneway. The inexpensive coach lights have the light bulb near the bottom of the fixture. By raising the bulb into the upper half of the fixture, the metal shields can be made to work much better. I suggest using two socket extensions from the electrical department of a hardware store. Light shouldn't shine to the side into the bushes, so shield those lower windows as well. "The animals will love it if you do" (Paul Simon). Now the light from the fixture will minimize glare along the road, it will illuminate the driveway and, the bulb will be easier to replace. The light that would otherwise shine into the sky is reflected down where it is needed. Without the glare, it can use a lower wattage bulb. The wattage may depend on local lighting conditions but in the country a 23-watt CFL bulb per fixture is more than adequate, so why not try 13-watt. If the coach light holds several light bulbs, why not just use one bulb and save on electricity and the cost of extra bulbs.

Virtually every house has a light. Since they tend to be left on all night long, if you are going to use one, it shouldn't affect your neighbours. 1) It should produce a wide uniform illuminated area, but not so wide as to shine off your property. 2) It should not produce glare at the curb, which will make it harder for motorists to navigate your neighbourhood. It would also be annoying to pedestrians who are out for an evening stroll. 3) It should not be so bright that it "blasts" your night vision - preventing you, your neighbours and the police from "seeing" into the shadows. 4) It shouldn't be too "white", which contains blue light components, because even the light that reflects off the ground can distrub your neighbours. If you want like your house to "standout" amid your neighbours, consider this. Your house will be the only one lit by, what looks like, "candlelight". This example is the CSbG-EcoLight.

>

Filtering White Light (added June.2019)

As mentioned in the notes titled “Do Plants Like Light?” and “Qualified "Truth" - LED Spectra“, we comment on some aspects of colour as it relates to the ecology and our vision. It becomes clear that not only are the blue spectral components not helpful but they can also adversely impact the ecological integrity of the environment and our health. Filtering out the blue light can reduce the impact of an otherwise high-impact light. For this to be practical, the filter should be readily available and inexpensive. I bought some sheets of the Roscolux Deep Straw #15 from BH Photo-Video. The filter is fairly transparent from yellow to red light, but transmits relatively little blue and green light. The second image shows a comparison between a number of LEDs with Correlated Colour Temperatures ranging from 2100K to 5000K (according to the product packaging) and one incandescent lamp. We see that a single filter film significantly reduces the apparent effective colour temperature. Only the 5000K LEDs seem to require two filter layers. The 5000K light gets rather dim because much of its energy goes into the blue light, which we are filtering out. The exposures have been adjusted to account for the different bulb wattages. And, the exposures were kept short to help show the colour changes with each filter layer. There are limitations to filtering. Filters “remove light” so the efficacy of the luminaire is lower. And, the filter material is degraded by weather, so it should be sandwiched between protective layers of transparent glass or plastic.

How Bright a Lamp Should I Use? (added July 2019)

Require Proof before we Change? (added March 2020)

Light Across the Water (added July 2021)