Device that blocks waves of one phase, but transmits wave of opposite phase?

[franco1991]

Is there any process or device that blocks EM waves of one phase (by any means, absorption, reflection, refraction, etc) but allows the transmission of EM waves of an alternate or opposite phase? Such that only waves of a certain phase can transmitted?

Alternatively, are there any devices/processed wherein the index of refraction is dependent on phase, such that waves of one phase will travel in one direction, and waves of an opposite phase will travel in a different direction (they will be refracted at different angles)?

 

[mathman]

My guess - highly unlikely. Phase is simply a relationship depending on an origin, which has no particular physical meaning.

 

[vb2341]

Well, I'm not sure this answers your question properly, you may want to expand on the background idea of this question a bit more, but something to look into might be lasers and coherence. I'm under the impression that you're trying to figure out something to do with interference, though I'm not really sure.

But like mathman said, phase doesn't really mean anything more than where the wave started.

 

[Neandethal00]

Is there any process or device that blocks EM waves of one phase (by any means, absorption, reflection, refraction, etc) but allows the transmission of EM waves of an alternate or opposite phase? Such that only waves of a certain phase can transmitted?

Alternatively, are there any devices/processed wherein the index of refraction is dependent on phase, such that waves of one phase will travel in one direction, and waves of an opposite phase will travel in a different direction (they will be refracted at different angles)?

If complexity is not an issue, it is possible to design a device which will pass waves of one phase only. But the waves must arrive one at a time. It would be more difficult if two or more waves arrive at the detector at the same time.

There are many electronic circuits that can detect 'zero crossing' of waves.

In the alternative case if the phase difference is caused by frequency differences, a prism should separate them.

 

[cjameshuff]

Phase isn't an absolute, inherent property of a wave, it is a description of how much it lags or leads some arbitrary reference. A receiver half a wavelength further from a transmitter than a reference receiver will see a signal 180 degrees out of phase relative to that received by the reference receiver, but both will receive the exact same signal, just at slightly different times.

You could probably produce something with non-linear optics that has phase-dependent behavior, but you would need to provide a reference signal to measure the phase against. For a somewhat related process, look into holography. A laser beam is split using some method that preserves phase, part of it being used to illuminate a scene, and the remainder as a reference to allow the phase of the reflected light to be recorded on film as an interference pattern.

 

[skeptic2]

Is there any process or device that blocks EM waves of one phase (by any means, absorption, reflection, refraction, etc) but allows the transmission of EM waves of an alternate or opposite phase? Such that only waves of a certain phase can transmitted?

Phase relative to what? How are you measuring phase?

 

[franco1991]

If complexity is not an issue, it is possible to design a device which will pass waves of one phase only. But the waves must arrive one at a time. It would be more difficult if two or more waves arrive at the detector at the same time.

How would this be done? Does such a device/process have a name, or it is hypothetical? And would the device work for low frequencies, or can high-frequency waves be applied as well?

Also, could you post a link explaining the circuits able to detect zero-crossing of waves, and likewise would such circuits work for high frequency waves, or can the period not be faster than the processing speed of the circuit?

 

[tasp77]

Not sure I am following the OP, but for a device sorting out light by phase, moving the device 1/2 of a wavelength (not much!) will have it deleting the 'wrong' photons. And moving the device half a wavelength could result from a slight change in temperature, acoustic vibration, seismic effects, or almost just looking at it funny.

 

[Neandethal00]

How would this be done? Does such a device/process have a name, or it is hypothetical? And would the device work for low frequencies, or can high-frequency waves be applied as well?

Also, could you post a link explaining the circuits able to detect zero-crossing of waves, and likewise would such circuits work for high frequency waves, or can the period not be faster than the processing speed of the circuit?

A zero-crossing circuit actually produces a square wave, high to low and low to high transitions take place at the zero crossing of the input sine wave or a noisy sine wave. You'll still need some type of microcontroller to read times when the transitions ocurred. You can build the zero-crossing circuit using any simple analog OP-Amp or comparator chip. Highest frequency I have seen such a circuit can handle is 140KHz.

I tried to attach a jpg but doesn't seem to work. OK, it may be working

 

[sophiecentaur]

I don't think this is the sort of answer that was expected in the original question and relies on the interference process but:

You could have an oscillator operating at the same frequency as the wave in question. If you adjusted its phase and amplitude then it could could be made to cancel the wave. This could only work for a wave along a transmission line or in one particular direction, using an antenna / loudspeaker (depending on the nature of the wave in question). The cancellation would not work where the 'geometry' was wrong for cancellation. If the phase of the arriving wave were to change then the cancellation would not work and you could end up with enhancement (constructive interference). Interestingly, the power that was being blocked from one direction would be re-directed into other directions rather than being absorbed.
This system is an 'active' one, relying on synchronisation (phase locking) of a local oscillator with the wave that you require to eliminate. There is not a passive way of achieving it, though, because phase is relative and needs to be specified with some Origin.

 

[cjameshuff]

A single zero cross detection circuit won't give you phase...as has already been stated, phase is relative. Two zero cross detectors and a reference signal will give you phase, though, as the difference in zero crossings. You will need to treat the rising and falling edges differently to handle the full range of possible relative phases.

Another approach to phase measurement is to multiply the input signal with a reference signal, and filter the result. Multiplying two sine waves in phase gives a sine wave of double the frequency, half the amplitude, and an offset that makes all parts of the waveform positive. As the sine waves move toward being 180 degrees out of phase, that offset shifts to make the output waveform entirely negative:
http://www.wolframalpha.com/input/?i=plot+sin(x),sin(x),sin(x)*sin(x)
http://www.wolframalpha.com/input/?i=plot+sin(x),sin(x+pi/2),sin(x)*sin(x+pi/2)
http://www.wolframalpha.com/input/?i=plot+sin(x),sin(x+pi),sin(x)*sin(x+pi)

If you low-pass filter the product of the two signals, the result depends on their relative phase, being zero when they are 90 degrees out of phase, 0.5 when in phase, and -0.5 when 180 degrees out of phase (for input signals with amplitudes of 1...units depending on what exactly you're working with).

But that's a fair ways from blocking a particular phase of EM radiation while passing another. I think the hologram approach is the one to take here, you need a reference to produce an interference pattern with, and a material that'll react to that interference pattern.

 

[sophiecentaur]

A single zero cross detection circuit won't give you phase...as has already been stated, phase is relative. Two zero cross detectors and a reference signal will give you phase, though, as the difference in zero crossings. You will need to treat the rising and falling edges differently to handle the full range of possible relative phases.

Another approach to phase measurement is to multiply the input signal with a reference signal, and filter the result. Multiplying two sine waves in phase gives a sine wave of double the frequency, half the amplitude, and an offset that makes all parts of the waveform positive. As the sine waves move toward being 180 degrees out of phase, that offset shifts to make the output waveform entirely negative:
http://www.wolframalpha.com/input/?i=plot+sin(x),sin(x),sin(x)*sin(x)
http://www.wolframalpha.com/input/?i=plot+sin(x),sin(x+pi/2),sin(x)*sin(x+pi/2)
http://www.wolframalpha.com/input/?i=plot+sin(x),sin(x+pi),sin(x)*sin(x+pi)

If you low-pass filter the product of the two signals, the result depends on their relative phase, being zero when they are 90 degrees out of phase, 0.5 when in phase, and -0.5 when 180 degrees out of phase (for input signals with amplitudes of 1...units depending on what exactly you're working with).

But that's a fair ways from blocking a particular phase of EM radiation while passing another. I think the hologram approach is the one to take here, you need a reference to produce an interference pattern with, and a material that'll react to that interference pattern.

There are many ways of determining the phase difference between two signals. The best method will depend on the frequency and nature of the wave (sound / light / RF).
You can use an appropriate method to phase lock a local oscillator which can then be used to 'null' out / cancel a signal from a source. If someone disturbs the phase / frequency of this signal, your Nulling device needs to chase this phase change until it achieves cancellation again. "The hologram" approach" is an example of this general principle of applying an interfering wave to cancel out another wave. But you always need a phase locked wave in order to achieve this cancellation - in the case of a hologram, this is obtained by splitting the source to provide a reference beam. This is only possible under some circumstances - not in the case of a radio wave, received from an unknown source.

Sound cancelling headphones achieve this sort of thing by using feedback to cancel interfering signals at a point right by your ear canal. But it's not quite what the OP was about, I think.

 

[cjameshuff]

If you just want to cancel out an electronic signal, you just need an inverting amplifier, and arrangements to match delays through the amplifier and outside it. Or if you know the wavelength, a partially silvered mirror in front of a fully-silvered mirror, with a half-wavelength separation. This gives a mirror with a "notch" in the spectrum of wavelengths it'll reflect. This doesn't allow selectivity between different phases, though...I don't see any way to achieve this without a reference to measure phase against.

It should also be asked if the original poster might be confusing phase with polarization. The two are quite different, but have some similarities that might cause confusion. There's quite a few materials and structures that'll separate EM radiation of different polarizations.

 

[sophiecentaur]

It should also be asked if the original poster might be confusing phase with polarization. The two are quite different, but have some similarities that might cause confusion. There's quite a few materials and structures that'll separate EM radiation of different polarizations.

I did wonder about that, myself. Perhaps he'll let us know. Then stand back for another barrage!