Talk:Dichroic filter

Untitled

Specific price ranges would be helpful, as one of the listed traits of dichroic filters is that they're quite expensive, but no concrete price is ever given. Six hundred dollars is expensive; so is ten thousand dollars. Knowing what kind of 'expensive' dichroic filters are is something that this article could use. -Toptomcat 16:08, 21 May 2007 (UTC)

STUPID USER: Raid-ISP-fibre optic-lens = superdichroic holography;;; exostack-flowing where the dichotomy appears to be with holograms are not this future because they are not a storager... —Preceding unsigned comment added by 86.138.211.72 (talk) 11:35, 1 October 2008 (UTC)

Regarding proposed merger

Although both articles currently equate dichroic filters and interference filters, this may be an instance of error arising from viewing a subject exclusively through the lens of the present-day technology without taking the history of its development into account. I would suggest that interested editors contemplate U.S. Patent 1238775 (available through Google Patents), issued to Frederic E. Ives in 1917, which is an early (as far as I am aware, the first) application of dichroic filter-reflectors to color photography, or indeed to any purpose. To roughly summarize: in a beam-splitter system for simultaneously photographing separate red, green and blue component images for color photography, thin coatings of certain aniline dyes are used to preferentially reflect and transmit various parts of the spectrum, avoiding some of the light loss that results from indiscriminately dividing all the light into three beams and then wastefully absorbing the unwanted wavelengths in each with filters. Although the coatings used are literally "thin films", their exact thickness is evidently not critical, so it hardly seems possible that they are functioning as interference filters, but I am no physicist and will leave that for others to determine.

AVarchaeologist (talk) 12:51, 21 March 2011 (UTC)

Well, dyes tend to work on the process of absorption. In interference filters, the thin-film coatings are usually transparant. Although sometimes transparant layers are combined with layers of colored materials, this often greatly reduces transmission. The filtering of colors is controlled primarily by the thickness of the coating, as well as its refractive index, to produce constructive interference in order to boost the desired color while destructive interference eliminates other solors. Often times, hundreds of layers are used to give the proper band pass/band stop.

I have recently added a rather brief history section to the thin-film interference article. That article covers the phenomenon itself in more detail. This was actually one of the great mysteries of science, which helped lead to the acceptance os the wave theory of light. According to the book Thin-film optical filters, the first interference filters were constructed by W. Geffcken in 1939.

I fully agree with the merge. This is simply a case of an object that has two different names getting two articles. I don't care which name is used in the title, but in the mean time, I'll try to link them to the thin-film interference article. Zaereth (talk) 18:39, 31 March 2011 (UTC)

I was not arbitrarily applying the word "dichroic" because the filters somehow struck me as being dichroic. It is consistently used in the patent, q.v.. The thin dye coating (alternatively, but less efficiently, a transparent deposit of gold) is not simply acting as an ordinary color filter. It is reflecting, rather than absorbing, the major part of the spectral region which it does not transmit -- about as good a functional definition as any, it seems to me, of dichroic filter/reflector surfaces. My point is that if this article is not to be limited to the technological "flavor of the week" (dichroic filter = interference filter), it needs to take past and foreseeable future technologies into account. From that perspective, an interference filter is one type of dichroic filter and the two terms are not synonymous, which argues against the proposed merge.

BTW: surely, no history of thin-film interference can be complete without at least a passing mention of the Lippmann color photography process (1891), which recorded color as laminae spaced one-half wavelength apart, producing a complex interference reflector/filter that reproduced true colors by reflection and complementary colors by transmission -- possibly the first practical application of thin-film interference and the true "first interference filters", but with those closing speculations I must again disclaim any qualifications as a physicist or as a historian of physics in general. AVarchaeologist (talk) 00:34, 5 April 2011 (UTC)

I don't have much knowledge of these beyond their use in lasers, but found the very, very brief summary of the phenomenon, provided by the only sources I can find, to be fascinating. That addition to the thin-film interference article is just a summary of the summary provided by those sources that I have available. Changes, additions or corrections are always welcome.

My knowledge of filters in general is very limited. I usually deal with anti-reflection and mirror coatings. It seems to me, though, that the terms dichroic and interference filters are used synonymously. Perhaps there are different kinds of dichroic filters, with interference filters being just one of them. Then, this might be cause to have a "parent article," summarizing them all, which can then link to "subordinate articles" about each one.

Unfortunately, I'm very busy right now, and so I don't foresee myself doing any research on the matter right now. I see that you may be rather new, so I will, however, point out a few things that may help you out. The first is that a patent is not going to suffice a reliable secondary source. As a primary source, it can possibly be used to provide information about the product itself, but you would need reliable secondary sources that can show that the product is somehow noteworthy. Primary sources should be never be the only sources, as this would constitute original research.

A patent can not be used to make any scientific claims. Patents come with no guarantee that a product will even work. (One of the most famous pantent cases ever fought was over a young scientist who drew a rudimentary sketch of a laser and patented it long before the first one was even close to being built. Many patents end up being out-right failures.) This may be especially true if the science is out-dated.

So, if you can find some other, secondary sources on this to show both the notability of the product, as well as its significance to this particular subject, then I sincerely hope that you would be obliged to add something about it to the article. I'm sure others will come along and help move this merge discussion in one direction or the other. (Merge discussion take a notoriously long amount of time.) Thanks for your help. Zaereth (talk) 02:06, 5 April 2011 (UTC)

I am well aware that patents can be misleading, and am also reasonably familiar with WP policy on primary sources and the fairly obvious reasons for caution regarding their use (stacked-deck presentation, novel interpretation, etc.) but straight reporting from cited and readily accessible primary sources (as opposed to unpublished manuscripts, archives of letters, etc.) is not forbidden, as they are verifiable and abuse can be detected by any interested reader. "Product"? There is no "product" being pushed here -- I doubt that Ives made a penny from his patent before it expired in 1934 -- and I have no plans to make any additions to this article or to otherwise edit it. I encountered it in the course of checking WP links for use in another article and was surprised to see the notice of proposed merger, as linguistically there is most certainly a major difference in meaning (check the etymology of "dichroic"). My point is simply that if the basis is that "dichroic filter" and "interference filter" are two names for the same thing, which may well seem to be the case if judged solely by the current technology, just one exception, even from the hoary distant past, explodes that argument and the articles should not be merged, at least not on that basis. Rather than basing a decision on how things seem, a little research (of the scholarly, not the creative "original" variety) might be appropriate before merging, and I have supplied one interesting (or so it is to me) lead to investigate, along with another historical avenue which ought to intrigue anyone actually interested in this topic per se. Adieu, addio and farewell. AVarchaeologist (talk) 05:15, 10 April 2011 (UTC)

Hmmm, your tone seems to suggest that I have offended you. If that is the case, then I must apologize, as that was not my intention. I did not mean to imply that you were pushing anythng, but was merely using the word "product" as a quick way to describe the subject of the patent. I think you have made some interesting points, and I was just trying to be helpful. I have used primary sources myself, like the U.S. Naval Flight Instruction Manual, but these typically require some expertise on their subject for someone to provide a reliable summary, and then can only really be verified if confirmed by reliable secondary sources, like the book Fighter Combat; Tactics and Maneuvering. You've definitely intrigued me, and I hope someone will come along and help sort this out. I may put forth some secondary research on it myself, but just can't afford the time right now. Perhaps in the future. Anyhow, thanks for participating in this discussion. Zaereth (talk) 07:01, 10 April 2011 (UTC)

I guess maybe I should clarify. I'm also just a passer-by, but mainly was reponding to this line, "Although the coatings used are literally "thin films", their exact thickness is evidently not critical, so it hardly seems possible that they are functioning as interference filters, but I am no physicist and will leave that for others to determine." I guess a more direct response would be that, in what is now called a "thin-film filter" or "interference filter," the thickness of the coating is critical. A coating that is called a thin-film is typically less than the wavelength of the light. Usually, the thin-film is either 1/2 or 1/4 of the wavelength. Thin-films are usually less than one micron, and if the thickness is not exactly uniform, the color distribution will not be either. (That's why the oil-film on water is not exactly uniform in color, because of variations in thickness.) This is what makes thin-film interference unique as opposed to other types of interference, such as standing-waves or diffraction.

Also, in the early 1900s, there were two different theories about the thin-film phenomenon. One involved a theory about it being caused by a dye, while the other was about microscpic structures. It wasn't until the 1920s that the structural theory became accepted over the dye theory. Only in the 1930s did vacuum deposition methods become available, depositing the films atom-by-atom and thereby providing precise control of the thickness. I am very interested in this, from a historical context, but am concerned that the coating in the patent may just be an early attempt based on the, later, disproven theory of a "surface color" dye. Still, you have raised some question in my mind that rhe term "thin-film/interference filter" may not be the only kind of dicroic filter, and maybe I'll get a chance to learn something, making this all worth while. Thanks. Zaereth (talk) 11:45, 10 April 2011 (UTC)

As it is plain that you are genuinely interested in the subject, not only in Wikipedia editing, I delay my departure to note that shortly after the elder Ives was dribbling dyestuffs onto glass, his son Herbert was finding scientific applications for the aforementioned Lippmann interference process: HERBERT E. IVES, "LIPPMANN COLOR PHOTOGRAPHS AS SOURCES OF MONOCHROMATIC LIGHT IN PHOTOMETRY AND OPTICAL PYROMETRY," J. Opt. Soc. Am. 1, 49-55 (1917). My digital copy was obtained free of charge from a source not recorded, but a quick new search turns up only for-pay sources, despite the pre-1923 date. Any textbook author who states that the first interference filters were created in 1939 has neglected to qualify his "first" (perhaps the word "constructed" is the qualifier, but it might apply to any artificially created structure), as there can hardly be any doubt about what these are.

Your comment that "in what is now called a 'thin-film filter' or 'interference filter,' the thickness of the coating is critical" is exactly my point with regard to the coatings described in the patent. Being of apparently indeterminate thickness, if thin-film interference were causing the effect the color ought to be varied across the surface due to uneven thickness; at best, if the coating by nature dried as a perfectly even layer, the color would not be predictable unless the thickness were, too, and the color of the dye would be irrelevant in either case. Much scientific water has passed over the dam during the past 94 years, and I would be interested to know what phenomenon was at work there according to current knowledge. Something at the molecular level, perhaps, because of the size of the dye molecules? Whatever the physics involved, unless the patent is grossly misrepresenting the behavior of the filters it appears that their function, regardless of structure, is being accurately described as "dichroic".

My last was a bit bristly because of your lecture on sources and the perils of patents. That would certainly have been deserved if I had injected any of this into the article proper, but in a contribution on a discussion page I was not out of line. I try to confine my actual editing to topics about which I have some longstanding and reasonably deep familiarity. Too many WP articles have been befouled by "helpful contributions" from people who came across an interesting factoid in someone's blog and are eager to share, or who read one chapter in one book and are now instant experts. AVarchaeologist (talk) 18:35, 10 April 2011 (UTC)

Oh, well then, in that case, the answer seems to be fairly straight forward. I took a few minutes to do a little digging. I finally found the patent and was able to compare it to current and older books on similar matters. I have found no sources on the invention itself, other than listings of abstracts in 1917 journals, such as one from the Kodak Company. It appears that Frederic Ives may have been the first to apply the term "dichroic" to filters, but, other than that, his filters seemed to have made no impact on science, as no further literature could be found.

From the patent, it appears that Mr. Ives was using dichroic dyes. This term had long been used to describe dyes that are one color when diluted, but are another color in heavier mixtures. For instance, magenta will often be blue when diluted, but a bright red when saturated. In modern terms, a dichroic dye is one in which the molecules all align in a uniform direction, very similar to the liquid crystal in an LCD. (In fact, dichroic dyes are often used in LCDs where color is needed, because their effect on light is similar.) Because all of the long molecules are aligned, the dye acts as a polarizer. If Mr. Ives used a dye that reflects green light, the dye would literally reflect the P-polarized light while still transmitting the S-polarized light. The claims of ideal transmission/reflection do not seem to be supported by any emperical testing, (which is often the case in patents). In such a dye I would expect a significant amount of absorption in both the red an purple ends of the spectrum, as well as a fair amount in the green. The science at the time apparently did not understand the molecular nature of these dyes. Whatever the reason, Mr. Ives invention did not seem to have caught on.

Most of the book I've looked into, such as The Science of Color, or Handbook of biological confocal microscopy, define filters as being of three basic types; absorbing filters, which absorb unwanted colors, dichroic filters, which, as quoted from the latter book, "are always of the interference type," or a combination of these, such as a thin-film stack on top of an absorbing filter. Whatever the specifics of Mr. Ives invention are, they do not seem to involve interference as the active phenomenon. Nor does it appear that anything ever became of this invention.

On the other hand, Herbert Ives may have just been the first to utilize Lippman color plates as an interference filter. Apparently, through some photo-chemical process that I do not fully understand, the standing waves generated by the mercury mirror somehow set the spacing of the silver grains in the colloid, effectively creating a structure analogous to a thin-film stack. This apparently is detailed in his paper, LIPPMANN COLOR PHOTOGRAPHS AS SOURCES OF MONOCHROMATIC LIGHT IN PHOTOMETRY AND OPTICAL PYROMETRY. I found Herbert Ives' book, An experimental study of the Lippmann color photograph, to be rather interesting, although I have not had a chance to read it fully. This may have actually been the first truely "dichroic" filter. (By the way, the qualifier for the 1939 invention was that they were the first to use vacuum depostion to make filters with dielectric coatings.)

I may be wrong about Frederic Ives invention, but simply don't have any real access to detailed studies of his design. I only did a little bit of research on this, so there may be more out there, but after checking all of the books at my disposal, plus going 10 or 12 pages deep on both google and google books, I could find nothing more about Frederic Ives' invention. Still, there may be something I missed, or don't have access to. Still, I really hope I was able to help. Zaereth (talk) 21:54, 11 April 2011 (UTC)

Thanks for the insight into the physics which may explain Ives' filters.

Although for some reason free access to the "Lippmann color photographs as sources..." article eludes a general web search, the article (which runs to page 63, not 55 -- opticsinfobase.org's citation, pasted above, is in error) is in fact available free via Google Books at http://books.google.com/books?id=PsXVAAAAMAAJ&pg=PA49&dq=%22LIPPMANN+COLOR+PHOTOGRAPHS+AS+SOURCES%22&hl=en

There is a "product" being pushed in this case, rather blatantly for a JOSA article. Per the distinction drawn in the current WP article, they are properly dichroic reflectors rather than filters -- I tend to conflate the two because the coating serves both functions in a beamsplitter, the application which most concerns me -- but no less relevant to a history of applied thin-film interference.

Returning to the original matter of whether "dichroic filter" and "interference filter" should be considered synonyms, searching turned up this example of non-synonymous use, also from the patent literature (US #2403685, 1941, page 1), which applies "dichroic filter" to both the Ives-type filters and interference filters: "Dichroic filters are also known in which a color is reflected other than that which is transmitted. For example, a thin film of gold will reflect a yellowish red light and transmit green. Eosin [the aniline dye cited by Ives] transmits red and reflects green, etc. Such dichroic filters...depend for their characteristics upon the material of which they are composed. It has...been proposed to make a dichroic filter by forming alternate layers of material having a high index and a low index of refratcion on the surface of a transparent medium, these layers having a thickness of a quarter of a wave length of the light which is to be reflected. As described, for example, in the Physical Review for June 1939..." AVarchaeologist (talk) 13:33, 12 April 2011 (UTC)

Well, as interesting as this discussion has been, it still is all OR and speculation. Under that criteria, I might be inclined to agree with you. Some companies, like Edmund Industrial Optics, define "interference filter" as being one of an extremely narrow bandpass, (sort of the opposite of a dielectric mirror), while a "dichroic" filter is one in which the filtering properties are controlled over the entire spectrum. However, the specific terminology varies from company to company. This is why I prefer to rely on peer-reviewed scholarly sources by people who are known to have expertise on their subjects.

Perhaps "dichroic filter" is being confused with dichroic beamsplitter. The coatings in beamsplitters is more akin to mirrors than filters. (There is often a big difference.) Metallic coatings are a whole different ballpark, as absorption can become a major factor.

There is one user here, User:Srleffler, that seems to have exceptional knowledge of optics. This is someone I often turn to when I have questions. Maybe he (she?) could help answer this question. Zaereth (talk) 22:23, 13 April 2011 (UTC)

@AVarchaeologist:@Zaereth: What is the final verdict on merging? If no merge, then there has to be a clearer explanation of the difference between the two. Right now, the intro makes them appear the same. If there are no further comments within the next few days, they will be merged. -- P 1 9 9  ✉ 18:06, 18 March 2014 (UTC)

No jury deliberations, no final verdict. Unresolved. The rather tedious exchange above got started because I simply piped up to say that not all dichroic filters are interference filters, which seems a substantial objection to such a merge. Note that the stubby but good Dichroism article refers to "...mirrors and filters, usually treated with optical coatings..." (italics mine). I am only an interested editor, not an expert, and you would do better to ping someone who is. My correspondent above suggests User:Srleffler, and I highly recommend User:Dicklyon, an honest-to-gosh working physicist with a special interest in, and expertise about, the subject of color. AVarchaeologist (talk) 09:25, 19 March 2014 (UTC)

I apologize for the tediousness of the discussion. I thought I was being helpful in answering your questions.

My support the merger, as it pertains to this discussion, extends from one point: Lack of notability due to lack of sufficient (or any) reliable secondary-sources. The existence of a patent does not, in itself, denote a notable item. I was really interested in this, so I did an extensive search through the internet, trying to find a secondary source showing that these dye-coated filters were ever sold or used in any products. Unfortunately, I found nothing. To include it into an article, there really needs to be something to support its notability, or else we'll end up having an article based upon every failed patent that ever was issued. Zaereth (talk) 10:21, 19 March 2014 (UTC)

Hi Zaereth, no apology needed, in fact you favored me with a few enlightening nuggets, for which thanks. I just wasn't expecting to get caught up in a wearying debate as the result of my modest heads-up. Regarding Ives-type dye-coated filters, the cube beam-splitters in the earliest (1930s) Technicolor three-strip cameras (and possibly the very different prism in their earlier two-color cameras) may have made use of patents licensed from Ives -- anything that allowed more efficient use to be made of the available light was highly desirable -- but that is based only on some WP-useless vague references I've encountered in the related literature. Patent-protected or not, Technicolor treated such intimate details as trade secrets. By the end of the three-strip era (1955), a dichroic filter/reflector coating of the then-modern type was being used.

I still oppose, on the basis of fundamental non-equivalency. AVarchaeologist (talk) 14:06, 19 March 2014 (UTC)

Unfortunately, fundamental non-equivalency is a matter of opinion, and here we have to deal with the verifiable information found in reliable secondary-sources. There is a big problem in relying on primary sources alone, which I think is highlighted here. For example, I could scribble a diagram of a perpetual-motion machine onto a napkin with a crayon, and then receive a patent on it. I don't have to have a working prototype, and the patent office doesn't verify and really doesn't care whether the machine will work or not. It simply has to be a novel design, and I can get a patent. (For an example, see Gordon Gould and his laser patent. Although his worked out, that is not the case for most patents.) Without reliable secondary-sources, there is no way to even verify that these things actually did what Ives claimed, that they ever got passed the design stage and produced a prototype, or that they were ever marketed at all. Primary sources can be wonderful sources of information. I've used them myself in many articles, like Basic fighter maneuvers, Barrel roll, Slow roll, and Wingover. However, they really require someone who has some substantial background knowledge of the subject in order to interpret them correctly, and this is why we need secondary sources to back them up. That way we are not providing the reader with sources that can be easily misinterpreted.

[interleaved comment] Here we go again. I really don't appreciate another lecture about the perils of using primary sources and patents in WP articles, which I have not done in this case and never had any intention of doing. This is a discussion page, and it seemed that just possibly by posting those tidbits here I might, as one effect, jog someone's memory about a more suitable source that could be used and cited in the article. Mission not accomplished, evidently. It must be said, too, that in the case of some arcane information, unless it is to be declared forbidden knowledge for WP purposes, there is no alternative but to cite respectable primary sources (such as peer-reviewed journals) because they are either not compiled or analyzed in any known secondary sources or have been seriously misreported by the all-too-human authors of the latter. AVarchaeologist (talk) 22:00, 19 March 2014 (UTC)

P.S. As a matter of fact, at least as late as the 1970s, the US Patent Office famously still required the submission of a working model in the special case of alleged perpetual motion machines, nearly a century after models were no longer routinely required. A few slipped by because they were labeled as amusement devices or the like, but violations of known physical laws and mechanical impossibilities were normally caught. Historically, the US system was no pushover and its examiners were usually well-informed within their fields and endowed with a healthy degree of skepticism. Very possibly things are different in modern times due to the vastly increased volume of corporate patent applications. AVarchaeologist (talk) 00:05, 20 March 2014 (UTC)

I'm not trying to lecture, and I really don't appreciate the insinuation. I'm merely explaining my position for others one final time so they don't have read through the "tedious" discussion above. Zaereth (talk) 23:47, 19 March 2014 (UTC)

We can nit-pick hypotheticals all day, but I think Gould's example speaks for itself. Listen, I think this is really interesting stuff, and truly wish I could have found enough information to warrant a separate article about these first "dichroic filters." I honestly do, or else I would not have taken the time research it and try to have a nice discussion with you about it. Had I found enough, I would gladly separate out all of the "interference filter" stuff out to it's rightful article and make this article solely about dye-coated filters.

I really don't care what name the article ends up being titled. "Interference filter" and "dichroic filter" are equally valid synonyms. This is an encyclopedia, and it's not about words; it's about things. Things that are the same but are called different names are given one article. Things that are different but share the same name are given different articles. In order to warrant giving Ives' invention its own article, I would need enough independently sourced information to be able to write such an article. Lacking that I cannot see the reason why we should not make one name the title of the article, make the other name a redirect, and place all of the information in one spot where the reader will be able to find it.

As it is, I have found enough information on Ives' invention to warrant a mention in a history section, or possibly an etymology section, but not enough to warrant its own section, let alone dividing it into its own article. I'm sorry about that, because I had truly hoped to find more, but I failed in my attempt. Zaereth (talk) 00:36, 20 March 2014 (UTC)

@ P199, this is pretty much the impasse we are stuck at, so I was hoping that someone else would come along and give some input of their own. I'm familiar with and very much respect Dicklyon, and suggested Srleffler because they are the most knowledgeable person I know when it comes to optics. Zaereth (talk) 19:30, 19 March 2014 (UTC)

I am not familiar with these topics, I'm just cleaning up the merge backlog. It seems that the editors you recommend are not interested (actually User:Srleffler is hardly active on WP). So that means that this merge proposal will result in "no consensus" = "no merge". If in the future more info is found to warrant a merge, just open a new merge proposal. @AVarchaeologist: Since you argue that there's a difference between the 2, can you please explain that in the introduction of each article? Thanks. -- P 1 9 9  ✉ 19:43, 19 March 2014 (UTC)

As I freely admitted to Zaereth above lo nearly three years ago, I do not consider myself competent to do any actual editing of this particular technical article, so I must demur. AVarchaeologist (talk) 22:00, 19 March 2014 (UTC)

I'm still around. I hadn't been watching this page until today. I don't know anything about the Ives filters. It seems to me that even if AVarchaeologist is right, Interference filter could still be merged into this article as long as the lede was adjusted to not say that the two terms are synonymous. Instead, we would be merging a more specific article up into the more general, broader article. To broaden this article to discuss the Ives filters, we would need a better source, though.--Srleffler (talk) 05:51, 20 March 2014 (UTC)

Hi Srleffler. I have one question for you, since you mentioned that interference filters are a more specific type of dichroic filter. Most of the books I've read are rather vague about any difference, and many often do use the terms synonymously. Some companies, like CVI Laser Optics, also use the terms synonymously, but others, like Edmund Industrial Optics, use the term "interference filter" only for filters with a very narrow bandpass range. Can you clarify if there is a generally recognized difference in the field of optics? Zaereth (talk) 18:55, 20 March 2014 (UTC)

I'm only familiar with the two things as synonymous. My point was that even if AVarchaeologist is right, the two articles could be merged on the basis I describe. --Srleffler (talk) 04:25, 21 March 2014 (UTC)

I always thought of these as two terms for the same thing; I'd take the "very accurate" out of the lead and merge them. If there are instances of sources that draw a distinction between them, that subtlety of usage could be pointed out in the article. I'm not aware of any; I'd have to study the sources for the "aniline dyes" thing that User:AVarchaeologist mentions, but I think it's not very relevant to current usage and could be covered in a short section. By the way, I used the terms "dichroic mirror" and "dichroic reflector" interchangeably for the interference filters in this paper; the point is that one wavelength range is transmitted and the other is reflected, and you can use both, which is a little more general than the "filter" concept. Dicklyon (talk) 02:45, 21 March 2014 (UTC)

Thanks for your response Srleffler. That's kind of what I thought, but I wanted to be sure. And thank you too Dicklyon. Different companies often use different terminology for the various types, but the only real difference is usually the number and thickness of the coatings. To give an example of where some of the confusion lies, here is a quote from the Edmund catalog:

Interference filters These filters have extremely narrow band (= or - 10 nm), broadband (+ or - 50 and 80nm) and linear variable transmittance across the substrate. They are extremely angle-sensitive, so care should be taken when mounting them in an optical train. Although highly selective, they do reduce throughput of the peak wavelengths significantly.

Dichroic filters These are substrates which are coated with thin-films to acquire the desired reflection/transmission characteristics throughout the spectrum.

Although found in catalogs, it's difficult to find these distinctions in books. Plus thee are other terms such as "color-substrate filters" (these may or may not have thin-film coatings), neutral density filters, longpass and shortpass filters. I think this article should be merged and then expanded to describe all that are relevant in the way Srleffler and Dicklyon describe. (Thanks again for the responses.) AVarchaeologist, does that sound like a fair plan to you? Zaereth (talk) 04:56, 21 March 2014 (UTC)

Off topic text in summary

In summary text: "This improves whiteness by removing excess red, however it poses a serious fire hazard if used in recessed or enclosed luminaires by allowing infrared radiation into those luminaires. For these applications non cool beam (ALU or Silverback) lamps must be used."

The text marked in bold is really off topic. Come on, suggesting using a silver backed lamp to prevent fires? This is a dichroic filter entry, not lighting installation and safety! I'll work on modifying in a few days if there aren't any good reasons to have this here. 71.202.73.172 (talk) 03:10, 14 August 2011 (UTC)

Details

This article is missing technical details. There are some details on Dielectric mirror but even then it only shows the basic principal. I'm curious how they can be made to work across a range of wavelengths, how they can have such a sharp cutoff, and (looking at the diagram on dielectric mirror) how the reflections off the back sides of interfaces get dealt with. —Ben FrantzDale (talk) 17:29, 14 December 2011 (UTC)

Color temperature

Removing the infrared content of a light beam does not change its color temperature, even if it does reduce the temperature of bodies that absorb the light. Color temperature literally is a description of the color of the light source. It comes from color photography: The color of the light emitted by an indcandescent photo-flood light bulb was dependent upon the actual temperature (typically measured in Kelvins) of the tungsten filament.

Photo floods are mostly obsolete now, but lamps are still assigned a "color temperature" of so-many Kelvins if the color looks about the same (either in a photograph, or in-person) as the color of a photo-flood that burns with that same actual temperature. If a dichroic filter is used to strip out only invisible wavelengths, then it should be obvious that that will not change the visible color of the light. 75.149.30.179 (talk) 20:00, 25 July 2019 (UTC)

Xenon flashtube ion-spectral radiation.

A hot-mirror to reduce red-eye in a camera

I agree, to some extent. What is being described there is what is commonly called a "hot mirror", used to reflect infrared light in order to reduce heating of components. A dichroic mirror and a filter are basically the same things, just used in reverse.

However, what's invisible to us may be completely visible to a camera, which can often clearly "see" wavelengths ranging from the near IR to the near UV, especially if it has plastic or quartz lenses. (For example, see the photo of the xenon flashtube to the right. The greenish-blue of the low-current arc is how it appeared to the naked eye, but all that bright blue was not visible. Those are the wavelengths around 900 nm, which the flashtube emits intensely at those currents, that the camera interprets as the second-generation harmonic of 450 nm (same "note" but the next octave up, A# or deep blue).) Removing a color from white light tends to change it to the opposite color, ie: filter out red and it becomes green, but color temperature is a term related to blackbody radiation. (ie: At low current the flashtube has no "color temp", but add enough current to the flashtube and the eerie, greenish-blue, haunted-house lighting will change into white greybody-radiation, then to the blackbody radiation of a blue-giant star.) Filters can be used to alter the color of the light, such as the yellow filters used by CSI Miami, or those horrid blue filters used in other shows, to give everything a drab look. But altering for color temp usually requires full, broadband filtering to try to match the right color.

That's not to say that invisible wavelengths can't interfere with visible light; they most certainly can. And do. That's why most mammals, and many other animals, have built-in filter/mirrors within their eyes called the tapetum lucidum, to reduce this interference and enhance contrast in the image. Some high-end camera manufactures use such hot-mirrors (pictured to the right) which exactly matches the reflectance of our own tapetum lucidum (usually placed in the same spot; just above the sensors within the "cone of least confusion" where al the jumbled, chaotic wavefronts interfere with each other to form the image), to both reduce interference and filter out human eyeshine (red-eye) which occurs at wavelengths just below the human visual-spectrum.

All in all, I agree that the wrong tem is being used, but will have to take some time to sit down a think of how to rephrase it. If you want to give it a shot, you're welcome to go ahead and fix it yourself. Zaereth (talk) 20:45, 25 July 2019 (UTC)

Merge proposal

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.

To merge Dichroic filter into Interference filter as a subtype or synonym; overlap and short text. Klbrain (talk) 12:57, 18 November 2023 (UTC)

---

This article needs to be merged with Interference filter, as both identify the other title as a synonym. Anyone have a preference for which title to merge to?— Srleffler (talk) 14:15, 28 May 2023 (UTC)

Seems like these are not synonyms but very closely related:

https://www.olympus-lifescience.com/en/microscope-resource/primer/lightandcolor/filtersintro/

"Dichroic filters are manufactured by coating either optical-grade or lower quality substrates, including polymers and glass, with thin films in a manner similar to interference filters to achieve specific wavelength transmission characteristics. However, dichroic filters are not as sensitive to the incident illumination angle as interference filters, and they are also not as wavelength-selective. In most cases, the term dichroic is reserved for filters having passbands of 100 nanometers or more with reflected bands being about twice as wide and containing wavelengths comprising the complementary color. Thus, a characteristic of dichroic filters is that they produce different colors when illuminated by reflected or transmitted light. These filters are often utilized as either additive or subtractive color filters for contrast enhancement, machine vision, or color separations. In general, dichroic filters provide a wider aperture than narrow bandpass interference filters and are more suitable for applications that do not involve image formation, such as traditional photographic illumination, printing, and stage lighting."

To me the best solution would be a merger and a renaming to make "Dichroic and Interference Filters". Basically this would be the current Dichroic page renamed and the Interference Filter page redirected to it. Johnjbarton (talk) 17:11, 29 May 2023 (UTC)

I don't like the idea of having such a long spliced-together title. I would rather go with "Interference filter" as an accurate description of the technology and then explain differences in terminology in the lede.--Srleffler (talk) 17:32, 29 May 2023 (UTC)

I think "Dichroic and Interference Filters" is the clearest, most direct way to describe it. Just "Interference filter" might confuse people looking for dichroic filters. Putting them together in the name also communicates the fact that they are closely related but not the same. It's also not the longest name on Wikipedia. Science-is-real (talk) 03:18, 27 September 2023 (UTC)

I disagree. The problem I find in these situations is that terms like these are often used as jargon by specific companies to describe certain products, but those same terms are then used differently by other companies. Here, the source is Olympia Cameras. Check Nikon, Sony, and others, you'll find they use these terms slightly differently. You can even go to optics manufactures themselves, like Edmund, Thor Labs, or CVI, and find they too have their own specialized jargon that relate more to their specific products than some wider scientific consensus. It's really all just a game of semantics at that point. Primary sources like these can often be wonderful sources of information, but this illustrates why we must be very, very careful in our interpretations of them.

However, when you actually look up these terms in very reliable secondary-sources, you find that they've already sorted this all out into much broader strokes. That's why we prefer them. For example, from the book Handbook of Biological Confocal Microscopy by James Pawley (page 212): "The various filters cause the largest intensity losses. Filters come in three general types: those made from absorbing/colored glass, interference filters made from transparent optical flats coated with several dielectric layers, and hybrids made from absorbing optical flats coated with dielectric layers. Dichroic filters are always of the interference type."

Thus, taking what I know about it from reliable secondary-sources, I would merge the article and use "interference filter" as the title. "Dichroic filter" can then just redirect here, and in the article we can explain whatever subtle difference exist in both products and their uses. We can even have it redirect to a specific section if necessary. Interference filter is just a broader, more all-encompassing term (something like "all liquids are fluids but not all fluids are liquids"). Zaereth (talk) 20:36, 27 September 2023 (UTC)

I agree. "Interference filter" is the broader topic. A dichroic filter is a specific type of interference filter. --Srleffler (talk) 04:16, 28 September 2023 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

  Y Merger complete. Klbrain (talk) 12:57, 18 November 2023 (UTC)