Talk:Ray (optics)

Diagram request

Perhaps this article could use diagrams illustrating some or all of the named rays.--Srleffler 06:14, 9 December 2005 (UTC)

Questionable edit

I'm concerned about this edit. The editor appears to be trying to clarify or correct the definitions, but the new material is not so clear to me. I'm not sure what he/she means by "starts at the object". Assuming the object is not a point, to define a ray you have to specify where on the object the ray starts. The old text did so (although it also made some assumptions and was not so clear). It is not clear to me that these new changes are correct. --Srleffler 06:28, 29 November 2007 (UTC)

Expanding on the above, the new definitions given for marginal and chief rays are not in agreement with those in Greivenkamp's book (reference in article), and are uncited. I will revert for now to the definitions that have citation support. I particularly note that the claim that images are formed where the marginal ray crosses the chief ray (as opposed to the optic axis) appears to be wrong. Perhaps we just have conflicting definitions of the same terms, used by different authors. If that is the case, a citation needs to be provided in support of the other definition, and the article needs to be broadened to explain and compare both usages.--Srleffler 20:44, 2 December 2007 (UTC)

principal rays

As far as I can tell, only principal rays (= chief rays?) are used in diagrams related to image perspective; including angle of view, entrance pupil, etc whereas diagrams showing how the image is focussed use marginal rays (?) as well. Please could someone explain / elaborate on this ? If I'm on the right lines this should be quite an important "key" to avoiding misinterpretation of ray diagrams. Redbobblehat (talk) 14:33, 2 September 2008 (UTC)

I don't know about "only" such rays being used, but you're correct that there is a connection. The perspective of a lens is controlled by the entrance pupil, and ultimately by the aperture stop. As noted in this article, the chief ray identifies the locations of the lens's pupils—it crosses the axis at each pupil location. The marginal rays identify the location where the the image will be formed.--Srleffler (talk) 03:46, 3 September 2008 (UTC)

"The marginal principal ray begins at the edge of the covered object and travels through the centre of the aperture stop and the edge of the field stop." (Reidl 2001 Optical design fundamentals for infrared systems p.1-2). This could be interpreted as contradictory to the article's current definition of "marginal ray" ... can anyone clarify ? (The significance, BTW, of this special principal ray is it traces the edge of the field of view.) —Preceding unsigned comment added by 83.100.248.38 (talk) 20:41, 30 January 2011 (UTC)

Not contradictory, just slightly different. Riedl's "marginal principal ray" is what this article calls simply the "principal ray". Riedl also has a "marginal axial ray", which is what this article calls just the "marginal ray". Riedl's formalism allows every object point to have a "principal ray", which is useful for detailed modeling of optical systems. In simpler analysis, one is concerned mainly with the principal ray from the edge of the object.--Srleffler (talk) 01:11, 31 January 2011 (UTC)

I added Riedl's definitions to the article. Thanks for providing a reference!--Srleffler (talk) 01:21, 31 January 2011 (UTC)

Definitive difference between chief ray and principal ray : "One exception is the case in which the aperture stop is at the lens, so that the chief ray is, in effect, the principal ray (i.e., it passes through the principal points, here coalesced at O) .... The point at which the ray crosses the axis is the optical center of the system, but since this is a chief ray, it is also the center of the aperture stop. " (Hecht 2003 Optics p.266-7 - see "Fig 6.34"). That's been bugging me for ages - woot! --Redbobblehat (talk) 20:50, 5 February 2011 (UTC)

Pharoid ray ?

I came across this definition of a third conjugate ray :

• Rays from object points at the of the object field which limit these ray bundles are called pharoid rays (PR). They are in analogy with the marginal rays (for object points on the optical axis) and thus determine the image brightness for points at the edge of the field of view. Note: By our definition, the pharoid rays are conjugate rays, that means corresponding to the same ray pencil. Sometimes rays ... which characterize the totality of all rays passing through the optical system are called pharoid rays. In that case these rays are of course not conjugate rays... It passes from the edges of the object to the edges of the aperture stop up to the edges of the image. [1].

From the diagram on p238, a pair of PRs flank the chief ray through the system - diverging from the same object point and converging to the same image point as the chief ray, but passing through the edges of the aperture instead of the centre. Something similar is depicted (but unnamed) in Fig.2 of van Walree's on Distortion [2] ...?

Google can find no other reference to a "pharoid ray", and "pharoid" (or similar) does not appear in any of the dictionaries I checked. Is this concept - possibly going under a different name - familiar to anyone ? --Redbobblehat (talk) 16:57, 14 August 2009 (UTC)

Another reference to the same definition of "pharoid rays" : [3]-p.17. --Redbobblehat (talk) 20:02, 14 August 2009 (UTC)

Parabasal rays

The new entry on parabasal rays doesn't accurately capture what the source is saying. The source is poorly worded, which doesn't help. What distinguishes these rays is not that they are close to the chief ray, but that they are "real" rays instead of "paraxial" ones. In Zemax and other optical design software, rays can be traced in two ways: a paraxial raytrace makes the paraxial approximation and ignores the physical shape of the surfaces, treating them as flat interfaces with a given optical power. This is a crude assumption, but was traditionally used because the raytracing calculation was very fast. It's also the appropriate assumption for analyzing first-order properties of the lens. Alternatively, Zemax can trace the real path of any given ray through the system, taking full account of the shapes of the interfaces and using Snell's law without approximation. On a modern computer there is little reason not to do this, so "real rays" are used for most purposes.

The author presumes that the chief and marginal rays are treated paraxially. (I forget whether this usage is typical in Zemax.) His parabasal rays are real rays traced in the paraxial limit, but without making the paraxial approximation.

Confusingly, the author's "chief rays" originate at the point where the object plane intersects the optical axis, rather than from the edge of the object. I'm not sure if this is a Zemax-specific usage or if it has wider support. What is going on, is the author would consider an optical system to have many "chief rays", one from each point on an object. (In Zemax terminology each field point has its own chief ray.) The (cited) definition used here, on the other hand, considers an optical system to have a single chief ray, defined by the most extreme object point.

I can try to rewrite the new entry, but I'm not sure I can do that without improper synthesis. It would be good to have a better source.--Srleffler (talk) 17:30, 17 August 2009 (UTC)

I agree the source uses some acrobatic ellipsis (namely parabasal = Snelly refraction and chief ray = base ray), but I didn't find it confusing at all!. I think we might be in danger of conflating the term "paraxial ray" with the "paraxial approximation" of small angles of incidence[4], [5], [6]. If the object point is close to the axis of symmetry, its chief ray will also be close to the axis. Doesn't this mean that the path of this chief ray is paraxial, regardless of how thoroughly it's angles of refraction are calculated ? A paraxial ray is only a good candidate for the paraxial approximation of its incident angles.

A parabasal ray is one which is 'almost parallel to' an arbitrarily chosen "base ray"[7], [8], [9]. These sources offer no indication that "parabasal" describes anything other than the general path of the ray, and no reason to suggest that "parabasal" specifically entails either Snelly or Sineless refractings. The obvious implication, however, of specifying a parabasal ray is "don't assume this is a paraxial ray", and then of course, if this ray is not paraxial, it is not a good candidate for paraxial approximation of its angles. --Redbobblehat (talk) 23:22, 17 August 2009 (UTC)

I think you're thinking along the right lines. The conflation of "paraxial ray" with "paraxial approximation" is deliberate, though. In optics modeling, and particularly in the context of Zemax, a "paraxial ray" is explicitly a ray that is treated using the paraxial approximation, regardless whether that is actually a good approximation for that particular ray's path. If I follow what the author of that reference is saying, a "parabasal ray" is a ray that is paraxial in the sense of having small angle to the axis, but which is not treated using the paraxial approximation. The new references you found make it clear that one can have parabasal rays around any defined "base" ray, not just the chief ray. I'm not sure that these references are all describing exactly the same thing, however. The textbooks appear to be describing an approximate ray propagation model involving an expansion about some known ray path. The Zemax treatment, on the other hand, is describing a full Snell's law tracing of the ray's path. I believe the same is true of the parabasal rays in ASAP, although they are used for a different purpose.--Srleffler (talk) 01:58, 18 August 2009 (UTC)

Finite rays

It seems that the term "finite ray" is commonly used to describe rays which are traced without the paraxial approximation : [10],[11],[12], [13]. --Redbobblehat (talk) 12:35, 18 August 2009 (UTC)

Warren J. Smith defines trigonometric ray as "A ray the path of which is traced according to Snell’s law (which see) as opposed to a paraxial ray. Also called an exact ray." (Modern optical engineering Fourth Edition 2007, p.731 [14]).--Redbobblehat (talk) 17:24, 18 August 2009 (UTC)

Oscillation figure

I think we need to get rid of the figure and caption that refers to light as "oscillating in amplitude" as it travels. First of all, it doesn't. It oscillates in phase. The electric field oscillates, but when one refers to "amplitude" in optics, it's understood to mean something other than the instantaneous magnitude of the electric field. Second, that figure adds absolutely no information, and is meaningless and confusing. Why does it help one understand ray optics? The wave nature of light doesn't even come into account in ray optics, so why is this even here?

If nobody objects, I'll delete it. Birge (talk) 22:20, 23 September 2009 (UTC)

Rays bending in the atmoshere

I added a link to this question in the section "See also". It is mentioned at the end of the section on boundary value problem of the article Computation of radiowave attenuation in the atmosphere --Thuytnguyen48 (talk) 15:08, 18 October 2010 (UTC)

I removed the link. An article on radio wave attenuation is not relevant to a general article on rays in geometric optics, even if that article does happen to mention propagation of rays.--Srleffler (talk) 02:12, 19 October 2010 (UTC)

Alternate Ray Names

A significant proportion of optical engineers (especially those trained at the University of Rochester) and optical texts refer to marginal rays as a rays and chief rays as b rays. I am tentatively going to plug these names as parenthetical notes in to the main article, unless anyone sees a reason not to. I'll provide a textbook reference when I have one in front of me. Most of them acknowledge a ray and b ray as alternative terms. 128.151.226.73 (talk) 19:32, 20 October 2010 (UTC)

Sounds fine. I'm not familiar with that usage myself. Please do provide a textbook reference to support it.--Srleffler (talk) 02:36, 21 October 2010 (UTC)

Newton's definition

The first definition in Newton's Optics is the ray. It is a least part of light moving in a line. His definition appeals to me because it is intuitive and practical rather than formal. It seems nice to have this around.

from: http://www.gutenberg.org/files/33504/33504-h/33504-h.htm

DEFIN. I.

By the Rays of Light I understand its least Parts, and those as well Successive in the same Lines, as Contemporary in several Lines. For it is manifest that Light consists of Parts, both Successive and Contemporary; because in the same place you may stop that which comes one moment, and let pass that which comes presently after; and in the same time you may stop it in any one place, and let it pass in any other. For that part of Light which is stopp'd cannot be the same with that which is let pass. The least Light or part of Light, which may be stopp'd alone without the rest of the Light, or propagated alone, or do or suffer any thing alone, which the rest of the Light doth not or suffers not, I call a Ray of Light.

AJim (talk) 03:40, 14 April 2011 (UTC)

In all the discussion below, I somehow missed your comments about Newton, above. I read and replied to only the paragraph below. I think it's important to point out here that Newton's conception of light is not just outdated, but fundamentally wrong. Newton strongly advocated the particle model of light, and opposed the idea of light as a wave. In his definition, he imagines that when you block part of a beam of light, the part you block is blocked and the part you don't block is unaffected. Light doesn't actually behave that way. When you block part of a beam, diffraction changes how the unblocked part propagates. If you try to find the "least part of the Light, which may be...propagated alone" what you get is not some ideal narrow beam, but rather a beam that spreads into a broad diffraction pattern. Newton's whole conception of light was wrong, and the error begins even before his first definition.--Srleffler (talk) 03:32, 1 May 2011 (UTC)

What bothers me about the current definition in the article is that it is narrowly focussed on ray tracing. I think it excludes talking about the polarization of a ray, for instance. I fear that this narrow definition may exclude some historic uses of the word, which may be confusing. Faraday, for instance, famously wrote, "...I have magnetized a ray of light...", and that ray was linearly polarized. AJim (talk) 04:09, 14 April 2011 (UTC)

I'm not sure why you think the current definition excludes polarization of a ray. Can you elaborate? Rays can certainly be polarized, and this is crucial in ray tracing.--Srleffler (talk) 00:18, 15 April 2011 (UTC)

I am glad you agree that polarization is important for rays. But consider this statement from the introduction: Ray theory does not describe phenomena such as interference and diffraction, which require wave theory (involving the phase of the wave). Or this statement from above: The wave nature of light doesn't even come into account in ray optics .... It seems to me that these discourage the kinds of description one needs to use to talk about polarization. The introduction mentions that rays obey Maxwell's equations, but people seem to be ignoring that fact that the solutions are waves, and that these can be two-dimensional, transverse waves. I think this gets in the way of explaining and understanding polarization of rays. AJim (talk) 17:12, 20 April 2011 (UTC)

Rays only approximately obey Maxwell's equations. Rays are an abstraction, a simple model for the behavior of light that throws away all of its wave properties except polarization, and treats it as traveling through free space in a straight line, like a particle. --Srleffler (talk) 02:45, 21 April 2011 (UTC)

Abstractions are good. However, I just checked, and the article does not seem to contain any word like polarization (pol*). How do you explain a circularly polarized ray? Would you agree that a plane wave, another popular abstraction, can be considered an infinite collection of parallel rays? I am still concerned that this article is focussed only on a modern understanding of a ray. I am interested in making it easier for beginners to understand polarization. I think discussing ray polarization is a very promising route. I think it is easier to understand, for a real novice, than the plane wave model I was taught. --AJim (talk) 03:51, 21 April 2011 (UTC)

I'm not sure where you're going with this. If you want to explain wave properties to novices, you ought to be using a wave model such as plane waves. Are you thinking of a ray as a narrow bundle of waves or something? I'm not getting how you would explain polarization to a novice using rays, and how that would be better.

If I had to explain polarization of a ray, I would probably think of it as a beam of photons, which have either clockwise or counterclockwise spin. This is, in fact, the underlying mechanism for polarization of light in quantum mechanics.--Srleffler (talk) 03:00, 1 May 2011 (UTC)

mention that incident light is incoming light

I want some advise. I was reading an article on colors and the term "incident light" is mentioned. As a good but not native english language speaker and reader, the word "incident" is unknown to me. So I edited the article, put 2 square brackets around it and am happy the term turns blue. So I hoover over it to see what it means, unfortunately it redirects to "ray". That's not useful. In the article on rays the term isn't mentioned either, only "incident ray" is mentioned with a drawing and I derive from it incident means incoming light. What should we do? Is that just a bummer for non-native english speakers and they just have to look it up in a translator or do we mention it under the redirect code of the incident light article or should it be mentioned in the article on "ray"? Thx --SvenAERTS (talk) 10:38, 9 August 2011 (UTC)

Probably the best fix is to explain the term where it is used. In what article did you see the term used? Putting "(incoming)" after the word incident would probably be sufficient.

The next-best solution is to link to Wiktionary for a dictionary definition of the term, like this: incident. Note that by policy, Wikipedia is not a dictionary. Wikipedia articles should not be linked to to provide definitions for words. That is the function of a dictionary, like Wiktionary.--Srleffler (talk) 02:23, 10 August 2011 (UTC)

merge with light beam

Light beam seems to describe the same thing, but artificial. The terms are interchangeable, right?--عبد المؤمن (talk) 18:36, 1 March 2014 (UTC)

Oppose. Did you bother to read even the first sentence of each article? They are not at all the same thing. A light beam is a physical, narrow beam of light, like from a spotlight or a laser. A ray is an abstract mathematical object used in analyzing how light propagates.--Srleffler (talk) 07:27, 2 March 2014 (UTC)

'Oppose' I also disagree. If anything it should be merged with Geometrical optics.Tomgee (talk) 18:25, 9 December 2014 (UTC)

Oppose. A beam is real, a ray is an abstract concept. I think it would be quite helpful, however, to explicitly address the fact that many people, not just the original poster, conflate the two. I think this is important because the confusion is deeply embedded; prominent scientists, Newton and Faraday come to mind, have used the term "ray" when they were describing a beam of light. --AJim (talk) 00:33, 10 December 2014 (UTC)

It's not wrong to use the word "ray" when describing a physical beam of light. It's just that that's not what we are referring to when we talk about "rays" in optics, and such physical rays are not the subject of this article. Some of the confusion may be caused by problems with the redirects and dab page that link to this article, which I am fixing.--Srleffler (talk) 03:43, 10 December 2014 (UTC)

Merger proposal

Out of incident, reflected, and refracted rays, only "angle of incidence" currently has its own article. The article is just a definition of "angle of incidence" and "glancing angle". I think it would be better explained in context, as a part of this overview article on the ray formalism. They're both pretty short, and there's some repetition of content. Main concern is WP:DICT. Forbes72 (talk) 21:32, 22 March 2016 (UTC)

• Weak oppose. Angle of incidence is an extremely important concept, and is referred to in many articles. Given that the meaning of the term may not be clear to many readers, it seems useful to have an article that explains the concept without forcing them to wade through an exposition on the ray formalism in general first. --Srleffler (talk) 01:09, 23 March 2016 (UTC)

It's a fair point about accessibility, but sounds like support for the advantages of a dictionary entry. What purpose does the angle of incidence article fulfill that is not served by the wiktionary entry? I agree on the importance of the term: we can always set up a "critical angle" subsection to link to directly from the search bar. Forbes72 (talk) 19:34, 23 March 2016 (UTC)

• I agree, leave separate it separate - to me, it's about discoverability of concepts, and links between concepts. Maybe over time wikipedia itself could gain a better mechanism for handling simple definitions (automatically embeddable in a parent article?) but in the meantime smaller articles seem better, even if there is a little repetition. In future the link structure makes wikipedia a valuable AI resource. another option would be a page 'reflection/refraction' which defines *all* the terms and 'angle of incidence', 'angle of reflection' 'angle of refraction' redirect into sections of it. Fmadd (talk) 10:03, 4 June 2016 (UTC)

Wording changes

I'm going to revert some of the wording changes, as they are more complicated, less clear, and not more accurate.

For marginal rays:

This ray is useful, because it crosses the optical axis again at the locations where an image will be formed. The distance of the marginal ray from the optical axis at the locations of the entrance pupil and exit pupil defines the sizes of each pupil (since the pupils are images of the aperture stop).

is better than

This ray is useful, because it crosses the optical axis again at the location of the image plane where the image is formed. At the location of the entrance pupil (exit pupil), the lateral or vertical distance of an extended straight line, along forward (backward) the marginal ray before the first optical element (after the last optical element), from the optical axis defines the size of the entrance pupil (exit pupil) since the pupil is the image of the aperture stop formed by the optical elements before (after) the aperture stop.

Firstly, images can be formed at more than one location. The marginal rays cross the axis every time an image is formed, not only at the final image plane of the system.

I'm not sure what User:Goodphy was trying to achieve with the rest of this change. The rest of the paragraph is just less clear with the new wording, and may not even be correct. There is no need to extend straight lines along anything. The marginal ray bends when it passes through optical surfaces. Its distance from the axis at a pupil defines the size of the pupil. -- Srleffler (talk) 22:49, 2 September 2023 (UTC)

Similarly for chief rays:

This ray crosses the optical axis at the locations of the pupils. As such chief rays are equivalent to the rays in a pinhole camera. The distance between the chief ray and the optical axis at an image location defines the size of the image. The marginal and chief rays together define the Lagrange invariant, which characterizes the throughput or etendue of the optical system.

is better than

As such chief rays are equivalent to the rays in a pinhole camera. The lateral distance between the chief ray and the optical axis at the image plane location defines the size of the image. An extended straight line, along forward (backward) the chief ray before the first optical element (after the last optical element), crosses the optical axis at the location of the entrance pupil (exit pupil). The marginal and chief rays together define the Lagrange invariant, which characterizes the throughput or etendue of the optical system.

There is no need to extend straight lines along anything. The chief ray simply crosses the axis at the locations of the pupils. (The ray's path bends when it passes through optical surfaces.)--Srleffler (talk) 23:03, 2 September 2023 (UTC)

Hi, Srleffler.

I agree that there are multiple locations where images are formed. That's a good thing to take.

However, I disagree the rest of your opinion because margin l and chief rays themselves do not directly determine the entrance and exit pupils if the pupils are virtual images. For real images, you are right that the marginal and chief rays define pupils.

I will modify that the contents in the Wikipedia page Ray (optics) by taking your points. And my intension for Wikipedia updates is to make it better for everyone. Goodphy (talk) 23:54, 4 September 2023 (UTC)

I see your point about virtual images. I couldn't see what you were driving at from the text I reverted though; it just wasn't clear enough. I'm not sure how to work an explanation that encompasses the possibility of virtual images into the text. The article needs to stay approachable to a reader who doesn't have much background in optics and is just trying to get the basic concepts. The technique of tracing virtual rays backwards is too big a concept to squeeze into the definitions of the different special rays, without making those definitions incomprehensible to many readers.--Srleffler (talk) 03:47, 5 September 2023 (UTC)

Hi. I have reviewed your changes, and tried to refined sentences for more clarification. You may review the contents. Goodphy (talk) 23:52, 5 September 2023 (UTC)