Noise canceling headphones work by creating an audio wave with opposite phase to the noise you are hearing. Does this work for light? If so that would let you “project dark” which would be great for increasing contrast on projectors and have lots of applications in the AV world.
I suspect that if this is theoretically possible the problem is that computer chips and projection devices can’t switch fast enough to match the waveforms that they sense so no canceling is possible in practice. Am I right or is this not even theoretically possible?
You have to remember that light acts as both a particle and a wave. The eyes do their sensing by absorbing the particles (photons). So you can block those by closing your eyes or wearing a mask.
It is theoretically possible, but in practice, not even close. It could be done with radio frequencies.
But remember that the noise cancellation effect is beyond the headphones. If there were a material which was a perfect reflector for audio sound waves, they could just use that. For light, there is; you could just put aluminum foil to block out the light you don’t want.
You’re missing the point. You can create noise canceling projectors which work the same way as the Bose headphones but block noise within a cone of projection, not just for the person wearing them.
What I’m asking is can we create a light canceling projector that works by detecting the photons in a space and projecting light with same frequency and opposite phase.
You can certainly create interference patterns with light where the opposing peaks and valleys of the waveforms cancel each other out. But I can’t think of how you could create a light-canceling device. It would take time to detect the waveform and calculate it’s opposite and since the signal is moving at the speed of light any opposing waveform your projected would never be able to catch up.
Fairpoint, I thought of this, but then perceptually is there a threshold within which the human visual system might not notice the delay in which a light canceling system can work?
It’s not the reaction time of the human visual system you have to worry about; it’s the reaction time of the light itself. The period of visible light is about 10[sup]-15[/sup] seconds, so you’d have to react faster than that to be able to achieve any cancellation at all. You could, in principle, anticipate what the light wave is going to do before it does it, and cancel it out that way, but that would only work with an extremely consistent source like a laser, not something like the Sun or a light bulb.
The sensor would have to be far enough away to give the machine time to do it’s work. Depending on how far away that is (feet, miles, 100’s of miles? I have no idea), the machine might even have to interpolate what distortions happens to the light between the sensor and itself and adjust for that as well.
No amount of distance would give the machine any extra time; you can’t transmit the detected signal any faster than light. I don’t think ‘slowing down’ the light from the source by inserting something in-between would work either (and an opaque wall would accomplish the desired result anyway).
Actually, I should point out that there are systems which can and do react to light fast enough to cancel it out, but such systems are precisely what an opaque wall is. The incoming light excites the electrons in the wall in just such a way that the light wave on the far side of the wall is canceled out, and there’s a reflected light wave on the near side of the wall.
The other major obstacle is the short wavelength of visible light.
Even with sound it is difficult to do cancellation unless you have one of two situations:
The source of the noise is highly localized and you can put a cancelling transducer within a fraction of a wavelength of the source.
The place where you want to recieve the audio is very localized and you can cancel the noise within a fraction of a wavelength of the reciever.
Case 1 is the basis for some commercial machinery which is made quieter by active noise calculation. This is usually working at low frequencies with large wavelengths.
The one situation where case 2 works really well is headphones. The wavelegth of middle C is about 4 feet. The highest frequency you can hear has a wavelength of about an inch. The headphones are placed precisely with respect to the reciever (your ear drum) and within a fraction of a wavelength over the audible band.
What if the device split the light in two, then let one path travel 1/2 wavelength further than the other before it recombined? Of course it would only work at one wavelength.
You don’t need to catch up with light. You can expect your light source to be emitting light with the same properties into the next nanosecond.
With coherent, monotonal light (ie, a laser), it is trivial. But actual light has an enormous mix of frequencies, and each source has its own phase (and i’m not sure, does polarity count too?) I think it could be done, but the amount of information (‘bandwidth’ in the technical sense) involved is huge. Can someone calculate it for us? It would be terabits, no? Audible sound is kilobits, maybe megabits.
And that is just for one point! For sound, you’ve got 2 point receivers. For light, you’d have to do it at each pixel.
I don’t think it’s possible even in principle for an incandescent light source, like an old-fashioned light bulb or the Sun, and it might not even be possible in principle for spectral-line sources like fluorescents or LEDs.
Oh, and the words you were using should be “monochromatic”, not “monotonal”, and “polarization”, not “polarity”.
You don’t think you could split the beam in two, run it through a prism, reflected through specially designed mirrors such that every path is 1/2 wavelength greater than it’s pair and then recombine them?
For a laser, sure, you could do that, and in fact something like that is involved in most practical applications for lasers. For a non-laser, though, 1/2 of which wavelength? A light bulb produces light at all wavelengths. At best, you’ll filter out a few specific wavelengths, but everything else will get through.
On thinking about it some more, you could pull it off for any monochromatic source, but it’d be a lot harder for a non-coherent source.
Optical interference is readily observed in oil films and insect wings. The colors you see are partly due to selective cancellation of specific wavelengths of white light, and partly due to selective reinforcement.
The trick would be to use metamaterials - engineered materials with a negative refractive index. You could create a mirror where light reflects on the back along the incident path, with a shifted phase. Current metamaterials are far too frequency/wavelength specific to allow general solutions, but in the future…
Metamaterials may help with the response time issue raised earlier, too. Remember that the limit of information propagation is c, the speed of light in a vacuum, and that the speed of light in air is somewhat lower (0.9997c). Metamaterials may have a speed of light faster than that of air, so you can get information ahead of the in-air wavefront, but given the small percentage differences in velocity, you could only do this over a long distance.
The better solution (for the OPs posited projection) is to engineer better screens, using metamaterials that have high reflection over a narrow incident/wider viewing angle, and zero reflectivity for other incident light. Work has begun on metamaterial traps that have zero incident reflectivity (for microwaves currently) that direct all incident radiation into the center. This will be ideal for energy capture (ie solar) but super-black materials are a possibility over a wide range of angles.