Tuesday, December 13, 2016

Imaginary Colors

I was watching QI (a BBC program of semi-serious questions and semi-serious answers, QI = Quite Interesting) and one of the questions which came up a few weeks ago was about imaginary colors. They messed up the graphics quite badly like this:


So colors between deep blue and purple-red are not supposed to exist. QI did not explain why. I've seen in some textbooks that these colors are described as mysterious or anomolous. In the textbooks  this diagram is used...



...which at least gives a bit more "explanation" of why they should not be visible. According to some people, since they are on that strange lower edge, and not on the "spectral edge" they are therefore non-existent colors.

(I found the reasoning to have the same weight as those who say "science says bumble bees should not be able to fly, but they can, so science is wrong!" It is clear to anyone with half a brain that bumble bees are not shaped like aircraft, are lighter and less dense. The science which explains bumble bee flight is going to be different to the science which explains the flight of huge passenger jets.) 

Back to colors though. There's two things wrong with this "science says we can't see mysterious colors but we can!" "reasoning":
  1. We can see colors which are not on the spectral edge, white is a good example. In other words we can see colors which are inside the shape shown above. Light yellow is another good example. So how near to the mysterious edge does a color need to be for it to become become mysterious? Clearly an arbitrary distance.
  2. We see colors because our eyes/brain (during daylight) mixes singals from three sensors (red green and blue). Our eyes don't even "know" about that mysterious edge. The diagram above is useful but it is not what goes on in the brain.



Tuesday, November 22, 2016

Explanation of the CIE88 2004 Tunnel Lighting Standard

The standard is officially called "Guide for the Lighting of Road Tunnels and Underpasses" and this article is about the 2004 version, which is currently in use. It replaces the 1990 version. 

This article is a user friendly summary of the standard and ignores some details, for example daylight screens and emergency lighting.

I'll just be looking mainly at the requirements for lighting tunnels in the daytime, a large part of which concerns the effect of the luminance of the areas surrounding the tunnel entrance.

For an explanation of what luminance means, this book will help you:

Buy Candelas Lumens and Lux as a paperback

You may have noticed that as you approach a long tunnel in bright sunlight the entrance often seems like a black hole: 
 


In these cases, if there was an obstacle (like a stopped car or a drunk pedestrian) just a few meters into the tunnel you would not know it until it you heard the thump!

(Note that you need to know the difference between luminous intensity, illumination and luminance to understand this article.)

The problem with tunnel lighting design is that tunnels have many variables which affect their safety. Some of the most important are:
  • The age of the driver.
  • The dryness or wetness of the road.
  • The speed of the automobile.
  • The external light conditions.
  • Atmospheric conditions (fog being an extreme example)
  • Traffic density.
  • Maintenance of the tunnel lighting.
  • The inclination of the tunnel.

It is neccessary that the driver sees into the tunnel entrance from the outside so he or she is justifiably confident about entering the tunnel. Of course the other concern is that once inside the tunnel they can understand the geometry of the tunnels easily. i.e they can see the walls and know that they are in the correct lane!

Generally lighting a tunnel at night is easier than lighting a tunnel during the day. If it is dark outside, then the change from driving outside a tunnel and driving inside a tunnel is smaller.

During daylight hours the problem is that while the eye can adapt from light (outside the tunnel) to dark (inside the tunnel), it cannot adapt very quickly. A slow moving car gives the eye more time to adapt than a fast moving car. That's one reason not to exceed the speed limit when entering a tunnel!


The obstacle

The CIE88 2004 standard has taken the position that it is impossible to define a level of visibility for all sorts of objects you might find inside a tunnel. Some tunnels contain only cars, some tunnels contain pedestrians and cyclists. Some cyclists don't switch on their lights! So the standard takes a "standard object" which must be visible at the entrance to the tunnel. This standard object is is a 0.2m x 0.2m square having a reflectance of 0.2: 


The object or obstacle is placed at the entrance to the tunnel, standing on the road surface, and the "stopping distance" is such that when the driver sees the object he must be able to stop in time without crashing into it.

Long and short tunnels


There are officially defined long and short tunnels, and in general short tunnels are those where the driver can easily see the exit of the tunnel from the entrance. This means that some "short in length" tunnels are treated as long ones when the short tunnel bends and the driver cannot see the exit from the entrance.

So there are three cases
  1. Geometrically long tunnels.
  2. Optically long tunnels (they may be geometrically short, but they're bent).
  3. Short tunnels (short and straight)
The sort of lighting level which needs to be applied is determined by the decision graph shown below. You start at the top answering the questions as you go downwards until you come to one of the three answers:




(I often wonder what we are expected to do if the answer to question 4 is 0.3, i.e. medium wall reflectance.)

The threshold zone lighting level is the lighting level at the start of the tunnel.


Explanation of terms used in CIE88 2004 standard.

Design speed: The speed for which the tunnel is laid out. It is often the same as the maximum speed allowed just outside the tunnel.

Reference point: A point in the center of the approaching lanes 1.5m above the road surface and outside the tunnel at the stopping distance away.

Stopping distance: The distance neccessary to safely stop the vehicle moving at the design speed. It is composed of the distance taken for the reaction (of the driver) time and the distance taken for the braking time.

Vertical Luminance, Ev: Vertical luminance is simply the luminance of a vertical plane, the normal to the plane is horizontal.

Contrast Revealing Coefficient, qc : The ratio between the luminance of the road surface and the vertical luminance at a given position: qc = Lr/Ev , illustrated here:



Symmetric Lighting: When the luminaire throws light equally backwards against the traffic flow and forwards with the traffic flow. In the example below the photometric solid has "wings" which are symmetrical along the traffic flow.



Counter-beam Lighting (CBL): When the luminaire throws light "backwards" into the flow of the traffic. 

Pro-beam Lighting: When the luminaire throws light along the flow of the traffic. 

The luminance curve for tunnels

A very important graph shows how the the luminance should change as the car moves into, through and out of the tunnel:



In general for long tunnels the interior zone is much longer than shown above. I've contracted the interior zone so the interesting entrance and exit luminances are show clearer. Remember that the above graph is a graph of luminace against distance.
Let's look at a more detailed version of the first half:



Lseq, Lth, Ltr and Lin are all luminances, hence the "L". They are explained in more detail below. Luminance is roughly the apparent brightness, what the eye percieves, not to be confused with illumination or luminous intensity.

Note that the Access Zone and the Stopping Distance (SD) are the same. The access zone is the section of road before the tunnel entrance, starting outside the tunnel, at the stopping distance from the tunnel entrance. So Lseq is the luminance in that section of "open" road. Notice that luminance falls once we get near the tunnel, because the tunnel mouth will start to dominate the visual field. This is shown graphically here:




Consider the three images, as the tunnel gets closer the "average brightness" percieved by the eye goes down. However well lit, in the daytime, the tunnel always has a luminance lower than the external environment.

Once the car is inside the tunnel it is in the "Threshold Zone", called this because the car is on the threshold between external road and the tunnel proper. As shown above the length of the threshold zone should be at least the stopping distance (SD).




Right after the Threshold Zone is the Transition Zone, where the luminance will fall to a (more or less) fixed value which most of the tunnel will have.

The Interior Zone has the fixed luminance value which will last until the car gets to the Exit Zone.  

The Exit Zone is often where external daylight illuminates the last part of the tunnel, and where the driver sees the external, brighter, landscape dominate his or her visual field.
The values Lth etc are generally taken to be minimum, and tunnel lighting should be at these minimums of above them.

Percieved contrast.

The percieved contrast is defined like this:



So it is the relationship between the luminance of the object and the luminance of the road. Obviously we'd like them to be different, if they are the same the contrast is 0! Remember that the object is a 0.2m square with reflectance (rho) of 0.2.

Now Lop and Lrp (used in the equation above) are defined as a sum of other luminances passing through mediums of varying transparency (transmittance). The windshield for example will reduce the luminance of the object because it does not transmit all the light which hits it. And the atmosphere too is not completely transparent. 

Lop and Lrp are calculated like this:



For example the atmosphere between you and the obstacle has a luminance (very small usually, unless you are in brightly lit fog) and this lumiinance is attenuated by tws, the transmittance of the windscreen.

Lseq is important. It is called the Equivalent Veiling Luminance. When light enters your eyeball it bounces around and gives a veil of light over the ordinary clean image. Lseq is considered to come from all the objects around a 2° cone of vision. The driver should be concentrating on that 2° cone, but the veiling luminance will reduce the contrast of what he sees.



It is not explicitly stated in the standard but I assume that the 2° cone of vision goes to the high resolution part of the retina. Other parts of the retina are medium or low resolution.

Compare perceived contrast with intrinsic contrast. The latter is the contrast when the you are very close to the object, in other words when there is no atmospheric or glare effects. Percieved contrast is different from intrinsic contrast because you are far from the object and light from other sources enters your eye, and the atmosphere between you and the object also reduced the contrast.

Lighting in the threshold zone.

You must be able to see other road uses in the dark threshold zone while you are driving outside the tunnel and are at the stopping distance away from the tunnel entrance. Obviously we are trying to avoid the "black hole" effect. Mathematically the percieved contrast should be at, or higher than, a given minimum.

Lth is the luminance in the first part of the tunnel, and is the horizontal section of the threshold zone (after the tunnel entrance in the graphs above). Lth is calculated like this:



Cm is the minimum percieved contrast required percieved contrast required. Rho is the reflectance of the obstacle (often set at 0.2) and qc is the contrast revealing coefficient. Generally all these numbers are given to us, except for Lseq...
So Lseq is the luminance created inside the 2° cone by light outside of the 2° cone. So this surrounding light veils what you are looking at, reducing the contrast.
How is Lseq calculated? You can either actually go to the tunnel and measure it with appropriate instruments, or use a graphical method explained below.
A polar grid is superimposed on the view of the tunnel entrance and its surroundings. Here is the grid :



You can understand it better if you see it over a photo:




The 2° cone is shown by the inner circle with the X in the middle. Inside your eye light from the other sectors invade that inner disk (on the retina of your eye) and reduces visibility there. The grid helps us get an idea of the luminance surrounding the cone.

Each area has been calculated to have the same influence on the 2° cone as all the others. Larger areas at the edge of vision have the same effect as smaller areas near the center of vision, given the same luminance in both areas.
The image above is actually a screenshot from a program which will sum the areas in the correct portions for you, giving you a value for Lseq. The program is LITESTAR 4D Tunnel Plus from OxyTech. Here is a fuller screenshot:



(For different standards there are slightly different radial grids, click here for a comparison of UNI11095 2003, CIE88 2004 and UNI11095 2011)

The standard requires that each quadrilateral is assigned a percentage of Sky, Road, Rocks, Building, Snow, Vegetation and Tunnel mouth. And each type of area is assigned a luminance.

Different areas occupy different amounts of the "quadrilaterals". Here is an example:



For example Sky is 8 kcd/squ-m in the example shown below. The luminances change with season and hour of course.



Lseq is a weighted sum of all the quadrilateral areas. The weights are the percentages of area type (Sky, Vegetation etc.) present in the area.

Back to that horrid looking formula:



We calculate Lseq with the grid, and all the other values in the equation are known to us. The threshold zone's constant luminance of Lth should last half the stopping distance, and then fall linearly to 40% of Lth. At which point we move into the transition zone...

The transition zone length and luminance.

The transition zone is the last zone before we hit the internal lighting zone of the tunnel. Here is a closeup of the move from threshold zone to transition zone. 


In the transition zone a new formula takes over, as shown above. The numbers are arranged so that at t=0 (0 meters into the transition zone) Ltr is almost exactly 0.4, thus taking over from the 40% linear fall in the second half of the threshold zone.

How do we calculate the transition zone length? Contact me for a detailed explanation.
How do we calculate the stopping distance? Contact me for a detailed explanation.




 

Wednesday, November 16, 2016

Unified Glare Rating (UGR) basic explanation.

Although the formula looks very complex, scarily complex basically it is actually quite simple, taken a step at a time. First a diagram:


Very roughly it is the luminance from the lamps divided by the background luminance from the room's walls and ceiling. 

(For an explanation of what luminance means, this book...

Buy Candelas Lumens and Lux as a paperback

...will help you)


Have a look at this detailed explanation of the formula for UGR:



Log is log10 by the way.

Lb, the background luminance or cd/m2rd, L is the luminance of the luminaire. 

Looking at L and Lb, glare increases with stronger lamps and lower background lighting, whereas it decreases with weaker lamps and more background illumination. 

p is the Guth index, which increases with distance from direct line of sight. So as the lamp moves further from the line of site, p increases and so the glare, as measured by the UGR, decreases.

Some useful to know limits:

  • UGR < 10 : Glare is so insignifigant it can be ignored.
  • UGR > 30 : Lots and lots of glare!
When you have that big table of UGR values “S” is the interdistance spacing of the luminaire centers in a grid in the room. You can see it has been set to 0.25 in the example below.


Note that there are many UGR ratings for the same luminaire but in different rooms. The table above gives UGRs for rooms with different wall, ceiling and worksurface reflectancies, as well as different room sizes (described as dimensions in MH (mounting height) units.)

The luminaire whose UGR is shown above is quite a low glare device, most of the room dimensions sizes give a value of around 10, which is considered a good value.

Notice also that since the luminaire is rotosymmetrical whether it is viewed endwise or crosswise does not change the UGR much. It's a different story with an assymetrical luminaire:







Understanding VH lighting photometric diagrams

VH lighting diagrams are cartesian rather than polar diagrams, and are normally used for floodlights. They show a hemisphere of intensities', it is assumed that no light is projected out backwards. In the real world "out backwards" would be towards the sky or the spectators of a football match.

The V and H angles are explained graphically here:


And here is an example:


In the icon top left the flat side of the hemisphere is the one which emits light.
The diagram above will be easier to understand if you also look at this annotated photometric solid:




Just to explain a bit more, imagine this luminaire is at the end of a playing field, above the goal, and imagine it is pointing into the center of the field:




As the V angle increases you are pointing further and further away from the goal, and closer to the center of the field. You could imagine that
  • V=-70° points directly at the goal
  • V=0 points at the center of the field
  • V=+70° points at the opposite goal. 
V is the Vertical angle. H is the horizontal angle and points to the left and right of the line between the two goals.

(In practice of course it is rare to have a floodlight above the goal, but it helps us to imagine what V and H mean.)

And here is one more graphical explanation of VH diagrams


(For an explanation of what intensity, luminance, and illuminance means, this book...

http://www.ransen.com/photometric/Candelas-Lumens-And-Lux.htm
 
  ...will help you)

Thursday, November 10, 2016

How to understand photometric polar diagrams

If you are working in the lighting industry sooner or later you will come across photometric diagrams and you must know how to interpret them. This blog entry page is quick introduction on how to look at a photometric diagram and get important information from it.

Often photometric diagrams use the C-Gamma system. Gamma=0 points downwards towards the floor or road. Gamma=180 points upwards to the ceiling or sky.

Here is a C-Gamma diagram with some of the luminous intensity “rays of light” left in:


The "rays" make the diagram more confusing than it needs to be and photometric diagrams always leave out those “rays” to give you a simpler diagram as shown below:


The point to remember is that the distance from the center of the diagram to one of the points on the “outline” corresponds to a luminous intensity value, often in candelas, in the given direction.

These diagrams tell you immediately if most of the flux (the lumens, the “flow of light”) goes upwards downwards or sideways. In the example above all the light "flows" in a downward direction. 

With C-Gamma photometries the gamma is the “elevation angle” and gamma=0.0 corresponds to a ray of light pointing downwards. 

The C angle, the angle of the “C-Plane” is usually represented as C=0 going off to the right along the positive x axis, and C=90 going along the positive y axis. 

The luminaire whose polar diagram is shown below therefore, shoots most of its flux “out to the left” and is symmetrical in the C90 - C270 plane (the dotted line):


A concrete example might explain better the concept of C-Plane. If mounted inside a room, you could put the C0 plane pointing north, then the C180 would point south, and so on. The 3D view below should help you orient yourself. Remember though that this is the default positioning, the luminaire can be rotated and tilted in real life.



 

The images below show you two different sorts of luminaires in a polar C-Gamma diagram. The first luminaire shoots all of its flux upwards, presumably it is used for indoor indirect lighting, when the light is first reflected by the ceiling before arriving at the worksurface. All the light is in the gamma = 90 to 180 degrees. The second luminaire shoots some of its flux upwards and some downwards, a “direct-indirect” method of lighting an indoor environment.


Sometimes, if the luminaire distributes light very unevenly or assymetrically it is useful to see a complete “photometric solid.” An example is given below:

These polar diagrams are sometimes called C-Gamma diagrams and are normally used for roads and interior lighting. Click here for floodlights and VH diagrams.

http://www.lulu.com/shop/owen-ransen/candelas-lumens-and-lux/paperback/product-20680738.html


Wednesday, November 9, 2016

Spatial chromatic nonuniformity in LED lighting, Delta u' v'.

You cannot assume that a LED light source emits the "same color light" in all directions. Some do and some don't. And then what does "same color light" mean? Some answers are given in IESNA document LM-79-08, with the calculation of Delta u' v'.

Here are the results of measuring the spectrum of a LED light source at various angles from the horizontal:




If you compare the spectrum at 90° with that at 0° you can see they are quite different, as are the u'v' values. But the u'v' values are not wildly different, so can an average be taken and then used to assign a single u'v' value to the whole light source? According to
LM-79-08, within very specific limits, yes, but you must also give a number for the variation in color.

Imagine you have a whole set of measurements around the light source, in all directions. In many directions there'll be little or no light at all, so they should be ignored. The samples which should be ignored are those which have less than 10% of the peak intensity.

A weighted average of the remaining samples is made. The weighting is based on the intensity (candelas) and solid angle "occupied" by the sample. In the formula below you can see that the weighting depends on I (intensity at theta) and omega (solid angle at theta)...




In the above formula xa will be u' or v', the same formula is used for both coordinates of course.

So now you have a weighted average u'v' value. Delta u'v' (the spatial chromatic nonuniformity value) is the largest difference (in the 2d space of the u'v' diagram) in the samples from the average. Low delta u'v'means the color of the light does not change much with direction.

If we assume for the sake of argument and simplicty that the u'v' values in the image above have already been weighted, then the average u'v' is 0.222 and 0.495. That average point is shown in green in the image below. The five samples are shown in red. The distance to the most distant sample from the average is shown by the white line, and is 0.0227, which is the Delta u' v' value.






So Du'v' is a measure of non uniformity of color of a LED in different directions. (Du'v' is not to be confused with Duv, as explained here.)


The new edition of my book Candelas Lumens and Lux now has a chapter on the spectrums of light sources:

http://www.lulu.com/shop/owen-ransen/candelas-lumens-and-lux/paperback/product-20680738.html
 

Tuesday, November 8, 2016

How to calculate spread and throw from isocandela diagrams

Photometric isocandle diagrams show what you can think of as a "sphere of light" with isocandles plotted on its surface. They are contour maps of the intensity of the luminaire. 

There are many projections of the sphere, you must use the equal-area (also called equivalent, equiareal or authalic) projection to calculate throw and spread, which are, by the way, purely "graphical" calculations.


To calculate spread:

Draw a line from the C=0° G=90° point and make it tangent to the 90% isocandela contour. In the above example this happens at about 17° "East" and 62° "South"

The angle at which it hits the edge of the "sphere" is the spread. It is the red diagonal line in the image above, so the spread is 27.2°

To calculate throw:


The throw is calculated by finding the longitude of the maximum candle intensity, then drawing a "vertical" (longitudinal line) from the intersection of the 90% contour with the longitude at maximim intensity. Here's the above image zoomed:




In the image above you can see that the maximum intensity is at C=10° ("longitude"=10°), and you can see the short red line inside the innermost 90% contour.

Now, once you have that line the throw is halfway along it, and in the first image on this pageis the long curved red line touching the edge of the sphere at Gamma = 62.3°

To be honest I don't find spread and throw very intuitive. Spread is considered to be how far a luminaire gets its light across a road. While throw is supposed to give you an idea about how far a light shines along a road. I find the newer Type 1 etc. classification much easier to understand. 

Very occasionally another problem with this classification method is that the isocandle contours don't always follow the neat "all roughly elliptical and centered on a single maximum" idea. Which can lead to ambiguous calculations and/or results.

Here's another example for you to test yourself on:





Buy Candelas Lumens and Lux as a paperback


How to calculate Beam Angle, Field Angle and Nema class


In floodlight VH (large area) photometries the field angle is the opening in degrees at 10% of the maximum intensity. The beam angle is the opening in degrees width at 50% of the maximum intensity.

Buy Candelas Lumens and Lux as a paperbackBuy Candelas Lumens and Lux for the Kindle on Amazon


So the field angle is wider than the beam angle, it requires less intensity. As a mnemonic remember that a field (of corn for example) is much bigger than a beam of wood (in a barn on the same farm). Well. Hey. That is how I remember it!

Here is an example (from OxyTech's PhotoView program): 


Here's a graphical explanation of where those numbers come from (screenshots taken from OxyTech's photometric program PhotoView):



In the above image the calculation for beam and field angles for H (=horizontal) has been made explicit (blue arrowed lines). See if you can roughly confirm the beam and field angles for V (in red) from the photometric polar diagram.

The NEMA classification for floodlights is simply some text with two numbers which specify the field angle. It is a simple lookup into a table to find out the numbers;

Field Angle NEMA Type
10° to 18° 1
> 18° to 29° 2
> 29° to 46° 3
> 46° to 70° 4
> 70° to 100° 5
> 100° to 130° 6
> 130° 7

You can see from the table that the bigger the NEMA Type the wider is the field angle. The types are found for both V and H angles. First the H Type is listed then the V Type. For example take the image at the top of this page. The H field angle is 113, so from the table the H Type is 6. The V field angle is 97.7, and from the table this is type 5. So the whole NEMA classification is:
6x5

as shown at the bottom of the first image.

Here is a 3D view of the photometric solid of a flood-light photometry with a much wider beam horizontally than vertically...


 
 ...not surprisingly it is a 7x4 (7 horizontal 4 vertical):