Talk:Lux

From Wikipedia, the free encyclopedia

WikiProject Physics This article is within the scope of WikiProject Physics, which collaborates on articles related to physics.
Start This article has been rated as start-Class on the assessment scale.
??? This article has not yet received an importance rating within physics.

Help with this template This article has been rated but has no comments. If appropriate, please review the article and leave comments here to identify the strengths and weaknesses of the article and what work it will need.

Contents

[edit] Disambiguation Page

I think there needs to be a disambiguation page for lux in terms of light and also the lux gene/operon in bioluminescent bacteria. —Preceding unsigned comment added by 99.225.24.69 (talk) 16:20, 12 February 2008 (UTC)

There is a disambiguation page for lux. It is at Lux (disambiguation).--Srleffler (talk) 16:38, 12 February 2008 (UTC)

[edit] Units and symbols

From the SI article:

  • Symbols are written in singular, e.g. 25 kg (not "25 kgs").

Similarly it is lumen not lumens

213.51.209.230 11:09, 9 Oct 2004 (UTC)

The symbols lm and lx are written in the singular. Lumen and lux are not symbols, they are ordinary words and follow ordinary grammatical usage. Lux is a special irregular case, because its plural happens to be lux.

OK, point taken 213.51.209.230 16:34, 9 Oct 2004 (UTC)
Bobblewik  (talk), whom I invited to comment (see below), is very knowledgeable on SI. Oh, by the way, welcome to Wikipedia, Sonett72. [[User:Dpbsmith|Dpbsmith (talk)]] 21:51, 9 Oct 2004 (UTC)

(NIST) Rules and Style Conventions for Spelling Unit Names

Plural unit names are used when they are required by the rules of English grammar. They are normally formed regularly, for example, "henries" is the plural of henry. According to Ref. [8], the following plurals are irregular: Singular - lux, hertz, siemens; Plural - lux, hertz, siemens. (See also Sec. 9.7.)

[[User:Dpbsmith|Dpbsmith (talk)]] 14:32, 9 Oct 2004 (UTC)

P. S. Although klx and klm are properly formed and presumably valid combinations of prefix and symbol, I don't recall ever, ever, ever having seen either of them. Are you aware of this as being real everyday usage (as opposed to, say,

1.00 × 103 lx ?

I would write klx, in preference to "1.00 × 103 lx". --Srleffler 05:19, 13 January 2006 (UTC)

[edit] Plurals

The official SI website says "Unit symbols are unaltered in the plural" (http://www1.bipm.org/en/si/si_brochure/chapter5/5-2.html). This applies to the symbols, not to the unit names. So unit names can follow the local language and spelling when used in full. English language forms are:

  • 50 metres and 50 m, but not 50 ms
  • 50 kilograms and 50 kg, but not 50 kgs
  • 50 seconds and 50 s, but not 50 s's
  • 50 volts and 50 V, but not 50 V's
  • 50 lumens and 50 lm but not 50 lms

Bobblewik  (talk) 14:58, 9 Oct 2004 (UTC)

[edit] Prefixes

According to the official SI website, SI prefix names and the prefix symbols can be added to any SI unit (http://www1.bipm.org/en/si/prefixes.html). Therefore klx and μlx are indeed valid and unambiguous forms. Some prefixes and unit combinations are less common, but SI does not address the issue of familiarity. That is a matter between authors and their readers.

Computing has made more people familiar with small (micro, nano) and large prefixes (mega, giga). I think it is interesting that potentially useful unit multiples Mg and Mm are not more common.

Bobblewik  (talk) 15:16, 9 Oct 2004 (UTC)

[edit] Less dictionary/science, more real-world?

I came to Wikipedia to look at the entry to lux because I was wanting some idea of how a 0.01 lux camera would compare in low light to a 0.1 or 1 lux camera. Like a comparison of what the human eye would see, versus what a camera at 0.01 lux would see. This may be one of those area too specific for Wikipedia to really cover, but it would be nice if the lux definition had some more real-world data than scientific data. --Jmccorm 04:18, 13 January 2006 (UTC)

You're not really looking for information on lux, you are looking for information on camera technologies and human vision. Maybe you'll find what you are looking for at visual perception, visual system or camera, or one of the related articles. I presume that a "0.01 lux camera" is a camera that the manufacturer claims will take a viewable picture at that level of light. This sounds like something that might be pretty poorly defined—more a matter of marketing hype than detailed specification. You might want to check the manufacturer's website and see if they provide an explanation of what it means. --Srleffler 05:19, 13 January 2006 (UTC)


[edit] 60 watts bulb example & a square meter?

the article says 1 lux = 1 lunem per square meter.

shouldn't the square meter be cube meter??

and, how many lumens is a average 60 Watts light bulb?

Xah Lee 07:21, 2004 Oct 12 (UTC)

1. No.
2. About 870: 14.5 lm/W (according to our light bulb article) × 60 W.

--Heron 08:25, 12 Oct 2004 (UTC)

Just to avoid confusion for someone reading this who is not familiar with these units: note that the 14.5 lm/W ratio is particular to incandescent light bulbs, and even then isn't all that accurate and only applies for bulbs around 60 W. Other types of light source have different ratings, and in fact the exact ratio varies from one type of light bulb to another. The article on incandescent bulbs has a nice table comparing the luminous flux of various light sources.--Srleffler 05:19, 13 January 2006 (UTC)

[edit] Human perception?

I think the remark below is wrong, for the following reasons:

  • Human perception of light intensity is not logarithmic, but follows a power law. That is, equal ratios of light intensity correspond to equal ratios, not equal increments of sensation. See Stevens' power law.
  • Although ratios of illuminance can be expressed in decibels, in fact I don't believe they ever are. That is, the comment verges on original research.
  • Something's wrong here, anyway, as lux is a unit of illuminance. I don't remember offhand what the unit of light-intensity-reaching-the-retina is, apostilbs or some damn thing like that? but in any case lux is not the right quantity. If this remark belongs anywhere it would be in the article about some other photometric unit, not the lux.

Radiometric measurements involving actual power, i.e. measurements derived from watts, might well use decibels (e.g. to express signal-to-noise ratios).

If I'm wrong about this, apologies. But I'd like to see a source. Dpbsmith (talk) 16:06, 11 May 2006 (UTC)

--I will try to provide a source, but I am starting to doubt. I think I heard it a few years ago at the Uni. I agree, it should be left out untill I verify my information. —Preceding unsigned comment added by 85.144.129.243 (talk • contribs)

[edit] Human perception

Human perception of light, like that of sound, is logarithmic[citation needed], so an increase by a factor of 10 in lux can be expressed as an increase of 10 dB in illuminance. With this notation, an increase from 50 klx to 100 klx is an increase in illuminance of about 3 dB.

Wrong. You need to read the famous paper by S. S. Stevens, "To Honor Fechner and Repeal His Law". He updated the Weber-Fechner law (logarithmic psychophysics) with the more accurate Stevens' power law. It's OK to cite decibels, but they should not be regarded as perceptually equal units. Dicklyon 07:26, 16 May 2006 (UTC)

[edit] decibels in optics and photometry

The use of dB to compare optical powers is always risky, since the dB definition, 10log(power_ratio), conflicts with the definition used in video and image sensors, 20log(voltage_ratio) where the voltage is proportional to the optical power. So when you see dB, you often can't be sure which it is. No such problem arises in audio and most other uses of dB, where power is proportional to the square of voltge. Dicklyon 05:47, 21 May 2006 (UTC) Ů

[edit] Definition

I changed the definition to make it clear the Lux is NOT a measure of perceived intensity, but rather a physical measure. To make the definition definite and complete, we should really include a table and/or plot of the Vlambda curve of luminous efficiency, which I expect is in the public domain, or at least widely copied from the IEC standard. Dicklyon 07:50, 16 May 2006 (UTC)

  • Yeesh. When did THAT creep in? Even if it were a psychophysical unit it would be the wrong one, as it measures illuminance on a surface, not on the retina.
    • Actually, illuminance of the retina (not of the scene being viewed) is what drives the retinal sensors and hence is an OK measure of the stimulus that is being perceived at that level. Similarly, total exposure of a photograpic sensor or film (in any small uniform patch) can be quantified as lux seconds. Dicklyon 14:49, 16 May 2006 (UTC)
      • No, that wouldn't be lux. I can never remember exactly what it would be--apostilbs or something like that. Lux measures illuminance of the scene. You can quantify the light falling on the retina in many ways... just as you can quantify the light output of a light bulb in "watts"... but lux is not the appropriate. unit. Sorry, apostilbs is a unit of luminance. The unit I'm thinking of takes into account the aperture of the iris and the differential effectiveness of light falling at angles that are not normal to the retina. That, by the way, is one reason (maybe the reason?) why lux is an inappropriate unit. The sensitivity of the retina to light does not obey Lambert's law; light coming in at an angle is "worth" less than when it falls on a Lambertian surface. The unit I'm thinking of is one of those obscure photometric units I use about once a decade if that... Dpbsmith (talk) 15:25, 16 May 2006 (UTC)
      • Stiles-Crawford effect, that's what I'm trying to think of... the appropriate unit for measurement of light on the retina takes account of the Stiles-Crawford effect... but what the heck is the name of that unit? Dpbsmith (talk) 15:31, 16 May 2006 (UTC)
      • Troland! Retinal illuminance is measured in trolands, not in lux. But I was out to lunch above: the Troland takes into account the pupil size, but does not take into account the Stiles-Crawford effect. (Unless, of course, the Stiles-Crawford effect is included in determining "effective pupil size.") Dpbsmith (talk) 15:40, 16 May 2006 (UTC)
        • Disagree! -- As the Troland article says, that's a measure of conventional retinal illuminance, which is not the same as actual retinal illuminance, which would be measured in lux if that's what you wanted to know. In cameras, for example, which are not so different from eyes, the film or sensor illuminance is quantified in lux, and the total exposure in lux-seconds. The actual illuminance is the actual physical stimulus that the eye or film reponds to (although for film it's not such a great measure, since the spectral sensitivity of film is not very close to the luminosity function as it is in the eye; nevertheless, it is used). Dicklyon 21:51, 17 May 2006 (UTC)
  • We shouldn't include a plot of luminous efficiency, because the curve is explained at luminosity function. The article links there. Sorry about the "perceived intensity". That was probably me. I was trying to distinguish it from direct physical measurement of power, not to imply that it was a measure of retinal illumination. Photometric terms are hard to explain.--Srleffler 04:45, 17 May 2006 (UTC)
    • Oh, don't make it so complicated, I just want to know the answer to a simple question: which is brighter, one watt or 10,000 candlepower? :-) Dpbsmith (talk) 13:51, 17 May 2006 (UTC)
      • Yes, let's get rid of all the silly Nit-picking. (If you think I'm going to sign that, you're crazy...)
      • That's easy. 1 watt can't make more than 683 candlepower (modern unit, candela), best case. So 10,000 candlepower is also brighter than 1 watt. Usually MUCH brighter. Now if you had asked 1 watt or 100 candlepower, the answer would require a lot more waffling. Dicklyon 05:56, 21 May 2006 (UTC)
        • ???? One watt can't make more than 683 lumens. Can't it make as many candlepower as it likes (I deliberately used "candlepower" because that is the unit that is always used, usually without the word "beam," commercially in the U.S., e.g. [1], [2], etc. If you put it in a parabolic reflector and scrunch it down to 0.0683 steradians, you get 10,000 candlepower, don't you? I'd have thought a one-watt laser could easily exceed 10,000 candlepower. Dpbsmith (talk) 10:34, 21 May 2006 (UTC)
          • Oops, you're right. Your intentionally-silly question didn't admit to the answer I thought it did. Thanks for the clarification, or intensification. Dicklyon 02:38, 22 May 2006 (UTC)
            • Oh, good. I'm never completely sure about anything when it comes to this stuff. For me, this is all half-remembered stuff that I one three-quarters-understood about half of. I always need to go to some reference source and do the mental equivalent of "sounding it out." But at least I understand that there are reasons why there need to be so many different units, that that when it comes to measuring light, the usual answer to "how can I convert X to Y?" is "you can't."
            • Take a look at User:dpbsmith/photometry... Dpbsmith (talk) 09:55, 22 May 2006 (UTC)

[edit] Biological clock

This seems wrong to me. I'd like to see a citation. In Czeisler's work, 10,000 lux is used to reset the clock. 1,000 lux is typical of ordinary workplace illumination and it is not sufficient to synchronize shiftworkers to their schedules, as witness Bhopal, the Exxon Valdez, and numerous other accidents that happened in the wee hours... Dpbsmith (talk) 12:46, 16 May 2006 (UTC)

I just removed the biological clock material from the table. An anon editor correctly noted that the cited article claims small adjustments of the biological clock at illuminances as low as 180 lux. That's fine, but it renders this issue not interesting or important enough to include in the table. Rather than a sharp threshold for resetting the clock, the reference shows a gradual increase in clock-resetting from very low levels of light to very high levels. One would presume that this variation is continuous, so there is no particular value that is distinct enough to be worth listing here. The clock-resetting level was only really interesting when it was higher than normal room light anyway. Light makes you feel more alert. Film at 11. (yawn).--Srleffler 23:58, 19 September 2006 (UTC)

[edit] Picking Nits... just call me an apostilb, I mean apostate...

I don't believe the section "Relationship between illuminance and power" belongs here at all. It should go in the article on lumen. Or, it should go in an article about luminous efficacy or the luminosity function and all of the articles on matching photometric/radiometric pairs should link to it.

The current text:

Illuminance is not a direct measure of the energy of light, but rather of the illumination it provides as perceived by the human eye. Thus, the conversion factor varies with the wavelength composition or color temperature of light. At a monochromatic light frequency of 540 THz (approx. 555 nm wavelength; the middle of the visible spectrum) the power needed to make one lux is minimum, at 1.464 mW/m²; that is, the peak of the luminosity function is 683 lumens per watt, falling to zero in the infrared and ultraviolet wavelengths. Typical light sources have a luminous efficacy much less than this theoretical maximum (luminous efficiency much less than 1). For example, a typical incandescent light bulb has a luminous efficiency of only about 2%.

mixes up illuminance, energy, and power. All of the very confusing photometric (and radiometric) units have different dimensions and different factors that affect them. We really shouldn't be trying to compare thingies that differ by more than one doohickey at a time.

This section is really about the relationship between photometric and radiometric units. It applies to power (watts) versus luminous flux (lumens). It applies to other corresponding pairs of photometric and radiometric units.

It doesn't apply to energy versus illuminance.

Illuminance is not a direct measure of the energy of light, for many reasons other than the luminosity function. For example, a strobe light may put out the same amount of energy in 1/1000th of a second that a floodlight puts out in ten seconds... but for that thousandth of a second its illuminance will be ten thousand times higher. Similarly, a bare bulb and a bulb with a parabolic reflector around it may put out the same amount of luminous flux, but the parabolic reflector results in a much higher illuminance over a much smaller area. Dpbsmith (talk) 19:02, 17 May 2006 (UTC)

I agree; make sure there's a good explanation in the appropriate article, and replace this section by a brief statement about photometry being different from radiometry, and a link. Dicklyon 21:43, 17 May 2006 (UTC)
I agree that section isn't very good, but I think it should be replaced rather than removed. Because of the confusion between photometric and radiometric quantities, every article on photometric quantities and units should contain a paragraph explaining the difference between the two (either a general paragraph, or one comparing that specific quantity/unit and the corresponding radiometric quantity/unit). The articles on luminous efficacy and the luminosity function are (and should be) more technical articles for readers interested in a greater level of detail. The articles on the quantities and units should give a naive reader a reasonably clear qualitative picture of the difference between the photometric and radiometric quantities, without recourse to the more technical articles. --Srleffler 22:10, 17 May 2006 (UTC)
I agree. I'm not sure I exactly know how to do it... Dpbsmith (talk) 22:24, 17 May 2006 (UTC)
I'm not too pleased with the new long-winded version. The long discussion of the relationship between lumens and watts might be OK in the lumen article, but not here. Then we could reference that with a short paragraph. Also, I believe you've inverted the meaning of luminous efficacy and luminous efficiency, if the article on those is correct. Dicklyon 05:57, 18 May 2006 (UTC)

[edit] Another nit: 540 THz versus 555 nm

I admit I was confused and some of my edits may need to be corrected. There's a difference between the standard frequency at which the 683 lumen/watt is defined (540 THz), and the peak of the luminosity function as adopted by the CIE, which is 683.002 lumen/watt at 555 nm. Dicklyon 21:08, 18 May 2006 (UTC)

[edit] Starlight example

Currently, the article states starlight causes an illunminance of 0.00005 lux. In my opinion this greatly depends on which star you look at and from where you do it. --Abdull 19:42, 19 May 2006 (UTC)

No, the illuminance provided by starlight on earth has no influence from where anybody looks; it's a physical quantity that you can get by putting a lux meter on the ground on a clear moonless night, away from a city. There may be some other dependencies, like some parts of the sky being brighter than others, but that's a pretty small effect. Dicklyon 21:03, 19 May 2006 (UTC)
"Which star you look at" suggests a misunderstanding of what "illuminance" is. I'm going to try to keep this simple which (means I'm going to introduce some inaccuracies of my own) but... here, in this context, "starlight" means how brightly the ground is lit up by all of the stars in the sky shining on it at the same time. It is about light falling on the ground; it is something you can assess very roughly by looking at the ground, not at the sky.
(Of course, now I'm wondering how much difference there is in how many bright stars are contained in various celestial hemispheres. I'm quite prepared to believe that starlight when viewed from a point on the earth on the galactic equator might be a little brighter than starlight viewed from a point under the galactic pole...)
To take another example: in "A Visit from Saint Nicholas" ("'Twas the Night Before Christmas"), Clement Moore writes:
The moon on the breast of the new-fallen snow
Gave the lustre of mid-day to objects below.
When we say that "moonlight" is 1 lux, we are not talking about how bright the moon appears when we look at the moon. We are talking about the moon as a light source for illuminating other things (hence "illuminance.") That is, how well it lights up "the objects below." Dpbsmith (talk) 19:02, 20 September 2006 (UTC)

[edit] Definition: bad math?

The current definition doesn't make sense to me: 1 lx = 1 lm/m2 = 1 cd·m2·m–4. Isn't this the same as saying: 1 lm·m–2 = 1 cd·m2·m–4, and therefore 1 lm·m–2 = 1 cd·m–2, leading to 1 lm = 1 cd , which is wrong.

But since the definition has survived for so long in this article, I suspect that I misinterpret it. How can something that looks so simple also look so wrong, and still remain uncorrected? In a similar fashion, the definition for Lumen (unit) is fishy. --HelgeStenstrom 08:43, 24 April 2007 (UTC)

This stuff always confuses me, too. I think the issue is that in the definition of lumen, there's a steradian that got dropped becuase it's 'non-dimensional'. Something like that applies to the lux, too. I usually end up back in books or tutorials that try to explain it better. I see what I can figure out the help resolve this... Dicklyon 14:48, 24 April 2007 (UTC)
I feel confused too. I quite suggest that the definition should be changed to: 1 lx = 1 lm/m2 = 1 cd·sr/m2. Yufeng.Zhang 13:51, 8 August 2007 (UTC)
This just illustrates the pitfalls of trying to do "math" on units. The results don't always mean what you think they do. You have to think about the physics, not just the math. In your derivation, you dropped a factor of 1 steradian, since 1 sr = m2·m-2 = 1. You can drop the steradian for purposes of dimensional analysis, but you have to keep it if you want to deal with units rather than dimensions. 1 lm ≠ 1 cd, even though they have the same dimensions, because these are units of different physical quantities. You can't equate units of distinct physical quantities, even if they happen to have the same dimensions. The correct interpretation of the correspondence between 1 lm and 1 cd derived above is the one stated in the article: "If a light source emits one candela of luminous intensity into a solid angle of one steradian, the total luminous flux emitted into that solid angle is one lumen." --Srleffler 00:15, 10 August 2007 (UTC)

[edit] convert footcandles and lux to candelas

"Candelas are equal to the square of the distance multiplied by the number of footcandles. For example, if your meter is ten feet away from the light source and your meter reading is 10 footcandles, the equivalent candelas equals 10 feet squared (e.g. 100) times 10 -- which equals 1,000 candelas.

The conversion of footcandles and lux to candelas is most accurately achieved in the context of a single point source of light measured in the dark. If the light source is diffused, you should take several readings at different angles in order to calculate an "average" candela measurement." [3] -69.87.200.181 16:22, 24 May 2007 (UTC)

"The name "footcandle" conveys "the illuminance cast on a surface by a one-candela source one foot away." As natural as this sounds, this style of name is now frowned upon, because the dimensional formula for the unit is not foot · candela, but lumen/sq ft. Some sources do however note that the "lux" can be thought of as a "metre-candle" (i.e. the illuminance cast on a surface by a one-candela source one meter away). A source that is farther away casts less illumination than one that is close, so one lux is less illuminance than one footcandle. Since illuminance follows the inverse-square law, and since one foot = 0.3048 m, one lux = 0.3048^2 footcandle ≈ 1/10.764 footcandle."

So, lux can be converted directly to candela, without computing footcandles. Just multiply the lux reading by the square of the distance, measured in metres. -69.87.200.181 16:38, 24 May 2007 (UTC)

[edit] Lux to take into account observer distance?

Lux takes into account the area of the light source, and its intensity. The observer's distance or own size (that the light shines upon) is not a consideration. With this in mind, why is "Candle at a distance of 30 cm (1 ft)" stat specifying a "distance of 30 cm"? Surely the lux would remain constant no matter how far away it is? 50,000cm or 50cm, does that same candle remain at 10 lux? At the moment, it sounds more like a measure of nits. --Skytopia 15:38, 4 October 2007 (UTC)

Not quite. There are two distinct things that are measured in lux: illuminance (light falling on a surface) and luminous emittance (light emitted from a source). You're thinking of luminous emittance, which is the luminous flux per unit area on the source. Illuminance is the flux per unit area at the location of the observer. The luminous emittance of a source does not depend on the observer, but the illuminance received by an observer decreases with distance. Two different things, but they both use the same unit.
The nit is a measure of luminance—an entirely different physical quantity.--Srleffler 17:40, 4 October 2007 (UTC)
Fair enough. Yes, the fact that Lux is used in two slightly different ways was confusing at first. The "SI photometry units" section also shows two 'types' of lux to support what you say. However, given all of this, would it not be a good idea to further highlight this on the page about the two different types. In particular, it would be a good idea to mention that the benchmarks given in the table are representing illuminance rather than luminous emittance. For example "Moonless clear night sky" actually refers to the density of light (i.e. luminous emittance) received by Earth. It would seem the current definition on there now could be ambiguous to the reader. --Skytopia 06:30, 10 October 2007 (UTC)
The column in the table is clearly labelled "illuminance". You're right though that the article should call attention to the two distinct uses of this unit.--Srleffler 10:54, 10 October 2007 (UTC)

[edit] When invented?/history

When was Lux unit used from? --maxrspct ping me 15:36, 5 November 2007 (UTC)