Half a croissant, on a plate, with a sign in front of it saying '50c'
h a l f b a k e r y
Is it soup yet?

idea: add, search, annotate, link, view, overview, recent, by name, random

meta: news, help, about, links, report a problem

account: browse anonymously, or get an account and write.



Variable filter double glazing

Double glazing which filters out light of a certain frequency by varying the distance between the panes of glass
  [vote for,

Nice swirly rainbow patterns are seen in oily puddles of water because of destructive interference between the light reflected off the top of the oil and the light reflected off the top of the water layer - different colours in the pattern equate to different thicknesses of the oil film on the puddle.
So my idea is to have semi-reflective double-glazing with absolutely flat, parallel panes of glass. Then, by varying the distance between the two panes so that the extra distance travelled by light which you (on the inside of the window) see after it has been reflected back off the inside pane and then back to you off the outside pane is an odd number of half-wavelengths it would be possible to selectively filter out a specific colour of light. Of course the amount of suppression of this frequency would depend on the reflectivity of the glass and so on.
No practical use, but it would be fun to fiddle with ("I don't fancy seeing purple light today" - <twiddles with controls> - "Ah, there we go. Lovely").
hippo, Nov 19 2001

What's the wavelength, Kenneth? http://www.pbs.org/...n/einstein_sp2.html
[angel, Nov 19 2001, last modified Oct 21 2004]

Iridigm http://www.iridigm.com
Display technology that takes advantage of something similar to what's described here. [bristolz, Nov 19 2001, last modified Oct 17 2004]

Article describing Iridigm's work. http://www.spie.org...ul00/micromech.html
"Micromechanical display uses interferometric modulation" [bristolz, Nov 19 2001, last modified Oct 21 2004]

Daniella Westbrook story (no pictures) http://www.ananova....tory/sm_416759.html
[hippo, Nov 19 2001, last modified Oct 21 2004]

You can get there from here... Liquid-filled_20window_20tinting
[normzone, Feb 24 2015]


       Ooh, pretty! I'd be surprised if you could ever get it to work though, as the difference between one colour and another is very slight. (linky)
angel, Nov 19 2001

       Interesting that you chose purple as your example. I believe (someone more knowledgeable please correct me if I'm wrong) that there's no such thing as purple light. That is, there's no purple wavelength, though of course you can have violet light or light that is a blend of red and blue wavelengths.
beauxeault, Nov 19 2001

       Oh yes, and - this being the HalfBakery - it goes without saying that this idea would also take advantage of subwoofers. As loud thumpy music microscopically bends and distorts the inner pane of glass the Interferometry Double-Glazing™ would selectively filter out different colours in time with the music. Maybe.
hippo, Nov 19 2001

       I don't think this would work just by varying the distance between the panes. You would need some type of filter that could be adjusted (as opposed to the panes themselves being the filter).   

       Perhaps something that emulates differing thicknesses of glass...
phoenix, Nov 19 2001

       No you don't - it just needs to be adjustable to fractions of the wavelength of light. If the two light paths are out by half a wavelength, you'll still get destructive interference. E.g. if the gap between the panes is 1cm, this is 15,822 wavelengths of red (HeNe laser, 632nm) light. Increase the gap by a quarter of a wavelength (158nm) and the two light paths will be 15,822.25 wavelengths and 47,466.75 wavelengths - half a wavelength apart, so you'll get destructive interference.
I accept your point that this may not be especially practical...
hippo, Nov 19 2001

       bx: The blue end of the spectrum does actually end with indigo and then violet (ROYGBIV), before going into the ultraviolet, so a large degree of what we call 'purple' could be excluded by excluding light at those frequencies. It is a bit weird, though, because some 'purples' are composed of red light too. Basically, the shades we perceive map mathematically to a crazy, malformed colour-space rather than that nice round colour-wheel you pick your paint off of.   

       Nice idea, anyhoo, even if it would be kinda hard to bake.
Guy Fox, Nov 19 2001

       shout me down, go on - I thought if you can see purple, then there must be a wavelength that is "purple".
po, Nov 19 2001

       The spectrum is considered as having seven colours only in Western culture, because the guy who 'defined' it (I forget who it was) had some mystical fixation on the number 7. Indian cultures see nine or more colours in a rainbow.
angel, Nov 20 2001

       [OT] I think humans in general have a fixation on the number 7...in psychological tests that I'm pretty sure I didn't just make up, people asked for the first number that came into their head showed a preference for 7 or numbers with 7 in them.
-alx, Nov 20 2001

       In psychology, 7 is a "magical" number because it is the approximate comfortable limit of digit memory. That's why US phone numbers are 7 digits (excluding area code). Most accurately, it is best quantified as "7 +/- 2."

But this is just another case of needing to be clear in our use of language. "Color" is usually used (vernacular) to refer to what we see. In more technical venues, "color" must be cashed out in terms of wavelengths. Once you're in wavelength speak, then the labels "red," "green," etc., have only analogical significance (like a rule-of-thumb), and are not to be taken as technically precise.

As for the idea: COOL!
quarterbaker, Nov 20 2001

       You'd look like dim UK actress and tabloid-fodder, Daniella Westbrook (only 6 holes after the middle bit of her nose succumbed to Cocaine). See link.
hippo, Nov 20 2001

       [bristolz] Very interesting link (I misread the name of the acticle's author as "Sunny Brains" which would be a lovely name) - a sort of microscopic version of this idea.
hippo, Nov 20 2001

       I just thought of a useful application of this idea - or at least something close to this idea. In London, the night skies are not black, they are orange, thanks to the thousands of streetlamps. These lights are sodium vapour lamps and transmit monochromatic light (i.e. of a single wavelength - about 589nm. They're used because this roughly corresponds to the peak sensitivity of the human eye). Therefore if your window glass had a coating with a thickness of a quarter of this wavelength then the light from sodium vapour lights would be subject to destructive interference and filtered out. Everything else would remain visible and the sky would not look orange.
hippo, Jul 23 2005

       I just found this via the random button. Coincidences abound.
1) Two days ago I was talking to yamahito about the colours in soap bubbles and mentioned the work I did at university working out the thickness of microtomed samples by looking at the colour of the slice.
2) Last week I helped a friend with a presentation, one slide of which had all the companies she's worked with over the last year. Qualcomm (the target of bristolz's first link) was among them.
3) I knew Sunny Bains at uni when she was running her own international science journal (why think small), called Holographics International. I worked on one issue with her, and still have a copy of it somewhere.
st3f, Aug 15 2006


Jinbish, Aug 15 2006

       Excuse my ignorance, [hippo], but...   

       ...I thought the thickness of an oily layer on a puddle of water was pretty much constant over any given puddle. So, the distance between the surface of the oil and the surface of the water is uniform, much like the distance between the proposed two panes of glass.   

       So, why would you expect to exclude one specific wavelength using the interference between the two panes? Why wouldn't you get a rainbow effect like the one seen in the puddle?
pertinax, Aug 15 2006

       [pertinax] On a puddle, the thickness of the oil layer is nearly constant. Variations in thickness cause destructive interference at different wavelengths which is why you often get a 'rainbow' effect. Surfaces which have a constant-thickness layer applied to them show a constant colour across the surface - such as is seen on the anti-reflective coating on camera lenses. This idea relies on the interface between light relected off the front of the two panes of glass in double glazing, so if you can keep this constant across the surface (tricky) you'll get a constant colour effect across the surface.

[st3f] Like SkyNet in the Terminator films, the HalfBakery has now become self-aware and knows what you're thinking.
hippo, Aug 15 2006

       Thank you.
pertinax, Aug 16 2006

       And the evil laugh activated hand-drier at last pulls free of the man-made chains that held it back. Its evil laugh now cuts through the souls of every half-baker throughout the lands...
NotTheSharpestSpoon, Aug 16 2006

       I don't think this would work. The reason a bubble has destructive interference is because the exact same image is being reflected from front and back surfaces. You simply won't get this effect once the distance between the two reflecting surfaces is greater than a few wavelengths.
xaviergisz, Nov 10 2006

       I never saw a Purple Cow, I never hope to see one;   

       But I can tell you, anyhow, I'd rather see than be one.   

       Frank Gelett Burgess (January 30, 1866 – September 18, 1951)
popbottle, Feb 24 2015


back: main index

business  computer  culture  fashion  food  halfbakery  home  other  product  public  science  sport  vehicle