[EL]
Although your post seems very logical, intuitive and rather obvious in
conclusion, I have never read such analysis any where in one and the
same page to focus on this issue.

Opticians would use different words such as "monochromatic" to
describe single-frequency-waves that may interfere to give finite
interference patterns.
An oscilloscope is a machine founded on the "time" domain, tracing
amplitude variations over a steady time scale we call the time-base.
An interferometer would be founded on the "space" domain from that
perspective, such that what is being displayed is the instant
"projected-steady-space-scale"; tracing amplitude variations too, but
in this case it is the spatial distribution rather than the temporal
distribution that matters.

Waves from a hot filament definitely interfere all the time, but what
is it that we should expect from countless patterns overlaid randomly,
both in time and space!
If we use a painting roller (with a pattern engraved on its surface)
in all directions on a wall and use it repeatedly, we shall end up
painting the wall completely.

The hot filament certainly emits an undetermined number of "mixed"
monochromatic rays.
We can verify that by using "monochromatic-color-filters" and
polarization filters.

In other words, we can "extract" pure-frequency-rays from that
"noise".

As for destructive-constructive interferences, they do not occur from
rays emitted from a single source hot filament because of the nature
of the spatial propagation over time.
Multiple sources and/ or bounced rays are the ones that interfere and
superimpose when they "collide" eventually.

A single photon may not interfere with itself due to the unique
spatial-temporal characteristics of each and every photon.
Photons of the same ray too are in different loci at an instance of
time or at the same locus but at different time-interval-slots.

All manipulations begin with beams (bundles of rays); we may split
them, change their path, delay them, polarize them and make them react
at a predetermined point in space-time.

In LASER devices, it is not true that a single photon could
"stimulate" a complete beam.
There is always an excitation source of high energy photons that
determines the energy of the beam. It is the resonant emission that
"magnifies" the constructive interference, producing an extremely high
amplitude waves-beam.

The law of energy conservation demands that we understand that there
can be no successive stimulation in a chain reaction scenario. The
mirrored surface and the partially mirrored surfaces are responsible
for the accumulation of energy at first, until a steady state is
reached, where the incoming number of photons (from source) equals the
outgoing number of photons. The lasing material is responsible for the
monochromaticity and polarization of the resulting beam. The frequency
of the waves is a LASER beam is the "lower-energy-bound" corresponding
frequency.

EL




John Kennaugh wrote in message o.uk...
How does light cause interference phenomena? Silly question. Light is a
wave when you add waves together they interfere. Ripple tanks and all
that. I don't think it is as simple as that. As an electronics engineer
I am familiar with different sorts of signal and the sinusoidal wave
shape, similar to what we see in a wave tank plays an important part.

However there is another signal familiar to electronics engineers and
that is a noise signal. Pure noise has a random structure. One can still
talk about intensity (power level) but this is a mean level, the
amplitude varies randomly and the interval between zero crossing points
is also random so concepts such as phase do not have meaning. It does
not have a frequency but it may have a bandwidth. If you put it through
a band pass filter then when looked at in the frequency domain the power
is evenly distributed across the band. If you look in the time domain
the upper frequency cut off will limit the maximum rate of change in
amplitude.

I am under the impression that very early radio transmitters, which used
sparks were in effect EM noise generators. It was not until the
thermionic valve oscillator was invented that sinusoidal EM radio waves
could be generated. My understanding is that something like a filament
light bulb generates light 'noise' and that it was not until the
invention of the laser that light joined radio in producing a pure
waveform.

If I have that right, and it seems logical that a filament should
produce light via a random process then it is the em equivalent of band
limited noise. What we see is the upper frequency limited by the
temperature of the filament and the lower frequency by the limitation of
our eyes in the infra red but no regular 'waves' in the wave tank sense.
The problem is that if you take a noise signal. Split it into two. Delay
one path and mix with the undelayed path then there seems no reason why
'a bit of it delayed' should have any fixed relationship with the 'bit
which is not delayed' - essential to give a static interference pattern.

Scott Murray suggested (from theory underlying the laser) that when a
photon is randomly emitted it acts like a catalyst and stimulates the
emission of a mass of other photons. This would suggest that light from
a filament is not in fact random but in short bursts of coherent light
with no fixed phase relationship between one bust and the next but a
fixed relationship within the burst.

If one takes the sinusoidal wave tank model then one expects to get
constructive interference if the path difference is n wavelengths (where
n is an integer) and destructive interference if the path difference is
(n+0.5)wavelengths. If my suggestion is correct there would be a maximum
value for n beyond which this would not be true. i.e. when n is
sufficiently large that you start mixing different uncoherent bursts
together. This in turn would give a rough idea of how long a burst
lasts.

I may of course in my ignorance be describing well know phenomena.
Comments please.

EL writes
[EL]
Although your post seems very logical, intuitive and rather obvious in
conclusion, I have never read such analysis any where in one and the
same page to focus on this issue.

Opticians would use different words such as "monochromatic" to
describe single-frequency-waves that may interfere to give finite
interference patterns.

I think perhaps we should define a little closer here what we mean. In
electronics noise is a useful signal for measurement purposes. 'White
noise', if analysed in the frequency domain, contains all frequencies at
equal power levels. If you filter it using a narrow band filter then you
will produce a narrow band of frequencies. If it is in the audio band it
will sound like a single frequency. If the bandwidth is carefully chosen
it will sound very like a violin playing that note. Why a violin and not
a flute? A flute produces a near perfect sine wave. Essentially when a
violin is bowed the string tries to resonate at a single frequency but
the bow keeps re-stimulating it so that it is like a sine wave
constantly being restarted at different phases which gives its
characteristic sound. One would still describe a violin as playing a
particular note, a particular frequency.

The term 'monochromatic' simply means single colour this may or may not
be 'phase coherent' in the sense a sine wave is. If you pass white light
from a filament lamp (or the sun) through a red filter it reduces the
bandwidth of the light it does not make it phase coherent. Even if you
use a monochromatic light source such as a sodium lamp# the process for
producing the light is still essentially random but because it involves
a particular transition of energy states produces single energy photons
= single frequency it is not the same as phase coherent light from a
laser. To produce interference lines you surely need phase regularity.

# I know sodium has two lines close together not one but I can't think
of a source which is truly 'single line'.

An oscilloscope is a machine founded on the "time" domain, tracing
amplitude variations over a steady time scale we call the time-base.

Correct and if you trigger a scope on a sinewave you get a stable
picture. If you put the same signal into the other scope input and
select 'add' it will add the two together and produce a sine wave twice
the amplitude. If you delay one input relative to the other the
amplitude will reduce until one is delayed by half a period compared to
the other when the amplitude becomes zero. A useful way of demonstrating
interference.

If you try it with a noise source then there is no stable picture just a
blur. No matter what the delay you put in the average amplitude is the
same (actually = sqr(2) x what it was with one signal).

An interferometer would be founded on the "space" domain from that
perspective, such that what is being displayed is the instant
"projected-steady-space-scale"; tracing amplitude variations too, but
in this case it is the spatial distribution rather than the temporal
distribution that matters.

Delaying one by a specific distance is related to delaying a specific
time by c.

Waves from a hot filament definitely interfere all the time, but what
is it that we should expect from countless patterns overlaid randomly,
both in time and space!
If we use a painting roller (with a pattern engraved on its surface)
in all directions on a wall and use it repeatedly, we shall end up
painting the wall completely.
## See comment below.

The hot filament certainly emits an undetermined number of "mixed"
monochromatic rays.

Again from electronics there is a difference between a comb spectrum and
a noise spectrum. On a spectrum analyser the first will appear as a
series of peaks close together each being an individual frequency the
second has no definite peaks just one broad peak corresponding to the
noise bandwidth.

We can verify that by using "monochromatic-color-filters" and
polarization filters.

In other words, we can "extract" pure-frequency-rays from that
"noise".

I have explained above that narrowing the frequency range does produce
pure frequency in the sense of regular phase. If you do a Newton's ring
experiment with white light you still get rings but the rings are
coloured. ## I am not sure in what you write above whether you are
implying that you don't. If you view such coloured rings with a
monochromatic filter I believe you will see the same result as you would
had you used monochromatic light i.e. put the same filter in front of
the light source.

As for destructive-constructive interferences, they do not occur from
rays emitted from a single source hot filament because of the nature
of the spatial propagation over time.
Multiple sources and/ or bounced rays are the ones that interfere and
superimpose when they "collide" eventually.

A single photon may not interfere with itself due to the unique
spatial-temporal characteristics of each and every photon.
Photons of the same ray too are in different loci at an instance of
time or at the same locus but at different time-interval-slots.

All manipulations begin with beams (bundles of rays); we may split
them, change their path, delay them, polarize them and make them react
at a predetermined point in space-time.


In LASER devices, it is not true that a single photon could
"stimulate" a complete beam.

I never said it did.

There is always an excitation source of high energy photons that
determines the energy of the beam. It is the resonant emission that
"magnifies" the constructive interference, producing an extremely high
amplitude waves-beam.

The law of energy conservation demands that we understand that there
can be no successive stimulation in a chain reaction scenario.

It is surprising that we still rely on cloud chambers to view the
progress of particles and never assume that the particle looses energy
when it initiates the formation of a bubble.

The suggestion was that a number of atoms have reached an energy state
where an electron could drop from one energy level to a lower energy
level and produce a photon, all energy involved coming from the atom
itself. That the first photon to do so spontaneously, may trigger others
which are 'ready' in an 'excited state' - whatever the terminology might
be - none of the photons so produced taking any energy from the
initiating photon. That a burst of photons so produced ends up a burst
of phase regular 'wave' - what you call 'pure-frequency-rays'.

The
mirrored surface and the partially mirrored surfaces are responsible
for the accumulation of energy at first, until a steady state is
reached, where the incoming number of photons (from source) equals the
outgoing number of photons. The lasing material is responsible for the
monochromaticity and polarization of the resulting beam. The frequency
of the waves is a LASER beam is the "lower-energy-bound" corresponding
frequency.

If you say so!


John Kennaugh
wrote

How does light cause interference phenomena? Silly question. Light is a
wave when you add waves together they interfere. Ripple tanks and all
that. I don't think it is as simple as that. As an electronics engineer
I am familiar with different sorts of signal and the sinusoidal wave
shape, similar to what we see in a wave tank plays an important part.

However there is another signal familiar to electronics engineers and
that is a noise signal. Pure noise has a random structure. One can still
talk about intensity (power level) but this is a mean level, the
amplitude varies randomly and the interval between zero crossing points
is also random so concepts such as phase do not have meaning. It does
not have a frequency but it may have a bandwidth. If you put it through
a band pass filter then when looked at in the frequency domain the power
is evenly distributed across the band. If you look in the time domain
the upper frequency cut off will limit the maximum rate of change in
amplitude.

I am under the impression that very early radio transmitters, which used
sparks were in effect EM noise generators. It was not until the
thermionic valve oscillator was invented that sinusoidal EM radio waves
could be generated. My understanding is that something like a filament
light bulb generates light 'noise' and that it was not until the
invention of the laser that light joined radio in producing a pure
waveform.

If I have that right, and it seems logical that a filament should
produce light via a random process then it is the em equivalent of band
limited noise. What we see is the upper frequency limited by the
temperature of the filament and the lower frequency by the limitation of
our eyes in the infra red but no regular 'waves' in the wave tank sense.
The problem is that if you take a noise signal. Split it into two. Delay
one path and mix with the undelayed path then there seems no reason why
'a bit of it delayed' should have any fixed relationship with the 'bit
which is not delayed' - essential to give a static interference pattern.

Scott Murray suggested (from theory underlying the laser) that when a
photon is randomly emitted it acts like a catalyst and stimulates the
emission of a mass of other photons. This would suggest that light from
a filament is not in fact random but in short bursts of coherent light
with no fixed phase relationship between one bust and the next but a
fixed relationship within the burst.

If one takes the sinusoidal wave tank model then one expects to get
constructive interference if the path difference is n wavelengths (where
n is an integer) and destructive interference if the path difference is
(n+0.5)wavelengths. If my suggestion is correct there would be a maximum
value for n beyond which this would not be true. i.e. when n is
sufficiently large that you start mixing different uncoherent bursts
together. This in turn would give a rough idea of how long a burst
lasts.

I may of course in my ignorance be describing well know phenomena.
Comments please.

--
John Kennaugh
to email convert the number from hex to decimal

[EL]
Greetings John, please realize that I am in agreement with you
generally, and that that was further confirmed by reading your reply.
See below for some comments.


John Kennaugh wrote in message o.uk...
EL writes
[EL]
Although your post seems very logical, intuitive and rather obvious in
conclusion, I have never read such analysis any where in one and the
same page to focus on this issue.

Opticians would use different words such as "monochromatic" to
describe single-frequency-waves that may interfere to give finite
interference patterns.

I think perhaps we should define a little closer here what we mean. In
electronics noise is a useful signal for measurement purposes. 'White
noise', if analysed in the frequency domain, contains all frequencies at
equal power levels. If you filter it using a narrow band filter then you
will produce a narrow band of frequencies. If it is in the audio band it
will sound like a single frequency. If the bandwidth is carefully chosen
it will sound very like a violin playing that note. Why a violin and not
a flute? A flute produces a near perfect sine wave. Essentially when a
violin is bowed the string tries to resonate at a single frequency but
the bow keeps re-stimulating it so that it is like a sine wave
constantly being restarted at different phases which gives its
characteristic sound. One would still describe a violin as playing a
particular note, a particular frequency.

The term 'monochromatic' simply means single colour this may or may not
be 'phase coherent' in the sense a sine wave is. If you pass white light
from a filament lamp (or the sun) through a red filter it reduces the
bandwidth of the light it does not make it phase coherent. Even if you
use a monochromatic light source such as a sodium lamp# the process for
producing the light is still essentially random but because it involves
a particular transition of energy states produces single energy photons
= single frequency it is not the same as phase coherent light from a
laser. To produce interference lines you surely need phase regularity.

# I know sodium has two lines close together not one but I can't think
of a source which is truly 'single line'.

An oscilloscope is a machine founded on the "time" domain, tracing
amplitude variations over a steady time scale we call the time-base.

Correct and if you trigger a scope on a sinewave you get a stable
picture. If you put the same signal into the other scope input and
select 'add' it will add the two together and produce a sine wave twice
the amplitude. If you delay one input relative to the other the
amplitude will reduce until one is delayed by half a period compared to
the other when the amplitude becomes zero. A useful way of demonstrating
interference.

If you try it with a noise source then there is no stable picture just a
blur. No matter what the delay you put in the average amplitude is the
same (actually = sqr(2) x what it was with one signal).

An interferometer would be founded on the "space" domain from that
perspective, such that what is being displayed is the instant
"projected-steady-space-scale"; tracing amplitude variations too, but
in this case it is the spatial distribution rather than the temporal
distribution that matters.

Delaying one by a specific distance is related to delaying a specific
time by c.

[EL]
Yes, but the difference between the two cases is not in the fact that
the wave-phase-spatial-location is being shifted when time is shifted
but that the wave-phase-shift (when comparing the two interacting
waves) is rather in a plane perpendicular to the direction of the wave
in the case of "interference patterns" rather than being in the same
plane of the wave direction-line in the case of
electronic-amplitude-modulation.
That is why phase-coherence of monochromatic waves is not a sufficient
prerequisite for the production of interference patterns but the
spatial polarity of the phase-shift too is essential, which may be
observed as a polarization of the pattern. Of course non-polarized
waves will also show a polarity in the interference pattern
corresponding to the direction of the phase shift.




Waves from a hot filament definitely interfere all the time, but what
is it that we should expect from countless patterns overlaid randomly,
both in time and space!
If we use a painting roller (with a pattern engraved on its surface)
in all directions on a wall and use it repeatedly, we shall end up
painting the wall completely.
## See comment below.

The hot filament certainly emits an undetermined number of "mixed"
monochromatic rays.

Again from electronics there is a difference between a comb spectrum and
a noise spectrum. On a spectrum analyser the first will appear as a
series of peaks close together each being an individual frequency the
second has no definite peaks just one broad peak corresponding to the
noise bandwidth.

We can verify that by using "monochromatic-color-filters" and
polarization filters.

In other words, we can "extract" pure-frequency-rays from that
"noise".

I have explained above that narrowing the frequency range does produce
pure frequency in the sense of regular phase. If you do a Newton's ring
experiment with white light you still get rings but the rings are
coloured. ## I am not sure in what you write above whether you are
implying that you don't. If you view such coloured rings with a
monochromatic filter I believe you will see the same result as you would
had you used monochromatic light i.e. put the same filter in front of
the light source.

As for destructive-constructive interferences, they do not occur from
rays emitted from a single source hot filament because of the nature
of the spatial propagation over time.
Multiple sources and/ or bounced rays are the ones that interfere and
superimpose when they "collide" eventually.

A single photon may not interfere with itself due to the unique
spatial-temporal characteristics of each and every photon.
Photons of the same ray too are in different loci at an instance of
time or at the same locus but at different time-interval-slots.

All manipulations begin with beams (bundles of rays); we may split
them, change their path, delay them, polarize them and make them react
at a predetermined point in space-time.


In LASER devices, it is not true that a single photon could
"stimulate" a complete beam.

I never said it did.

[EL]
Yes, I apologize if I sounded like I implied that you did.
There are two major sequences of events in the production of LASER
beams.
The first sequence involves the source high-energy-photons bouncing
randomly in planes inclined on the direction of LASER beam
target-direction. This sequence of events includes the atomic energy
level excitation of valence-electrons preparing them for action
(warming-up), where the bouncing is mainly off the cylindrical
half-mirrored surface of the LASER-tube.
The second sequence of events involves the full-mirrored back and the
half-mirrored front of the tube on which (emphasizing)
resultant-emission-photons bounce to-and-fro until phase coherence
randomly happens along the lattice, at which point a single photon
literally "collects" all the energy contained in its path (from all
the pre-excited-electrons) before exiting the LASER tube.
I only cared to clear a common confusion concerning the difference
between a wrong "stimulated-excitation" and a correct Stimulated-
Emission.

The total energy of the beam is governed by the efficiency of the
energy conversion applied to the excitation of the core by the source
photons.
Stimulated emission precisely means that the returning of excited
electrons to their ground states is stimulated by the event of the
being amplified photon entering the spatial domain of the excited
electron such that the being emitted photon is coherently in-phase
with stimulating photon hence "adding-up" amplitudes constructively.
This sequence of events where an amplified-stimulating-photon could
pick-up more energy along its path (which is parallel to the axis of
the LASER tube) may happen many times before the ray exits the lattice
and onto the absorbing target.



There is always an excitation source of high energy photons that
determines the energy of the beam. It is the resonant emission that
"magnifies" the constructive interference, producing an extremely high
amplitude waves-beam.

The law of energy conservation demands that we understand that there
can be no successive stimulation in a chain reaction scenario.

It is surprising that we still rely on cloud chambers to view the
progress of particles and never assume that the particle looses energy
when it initiates the formation of a bubble.

The suggestion was that a number of atoms have reached an energy state
where an electron could drop from one energy level to a lower energy
level and produce a photon, all energy involved coming from the atom
itself. That the first photon to do so spontaneously, may trigger others
which are 'ready' in an 'excited state' - whatever the terminology might
be - none of the photons so produced taking any energy from the
initiating photon. That a burst of photons so produced ends up a burst
of phase regular 'wave' - what you call 'pure-frequency-rays'.

The
mirrored surface and the partially mirrored surfaces are responsible
for the accumulation of energy at first, until a steady state is
reached, where the incoming number of photons (from source) equals the
outgoing number of photons. The lasing material is responsible for the
monochromaticity and polarization of the resulting beam. The frequency
of the waves in a LASER beam is the "lower-energy-bound" corresponding
frequency.

If you say so!

[EL]
Assuming that I know what I am talking about.
smile

EL




John Kennaugh
wrote

How does light cause interference phenomena? Silly question. Light is a
wave when you add waves together they interfere. Ripple tanks and all
that. I don't think it is as simple as that. As an electronics engineer
I am familiar with different sorts of signal and the sinusoidal wave
shape, similar to what we see in a wave tank plays an important part.

However there is another signal familiar to electronics engineers and
that is a noise signal. Pure noise has a random structure. One can still
talk about intensity (power level) but this is a mean level, the
amplitude varies randomly and the interval between zero crossing points
is also random so concepts such as phase do not have meaning. It does
not have a frequency but it may have a bandwidth. If you put it through
a band pass filter then when looked at in the frequency domain the power
is evenly distributed across the band. If you look in the time domain
the upper frequency cut off will limit the maximum rate of change in
amplitude.

I am under the impression that very early radio transmitters, which used
sparks were in effect EM noise generators. It was not until the
thermionic valve oscillator was invented that sinusoidal EM radio waves
could be generated. My understanding is that something like a filament
light bulb generates light 'noise' and that it was not until the
invention of the laser that light joined radio in producing a pure
waveform.

If I have that right, and it seems logical that a filament should
produce light via a random process then it is the em equivalent of band
limited noise. What we see is the upper frequency limited by the
temperature of the filament and the lower frequency by the limitation of
our eyes in the infra red but no regular 'waves' in the wave tank sense.
The problem is that if you take a noise signal. Split it into two. Delay
one path and mix with the undelayed path then there seems no reason why
'a bit of it delayed' should have any fixed relationship with the 'bit
which is not delayed' - essential to give a static interference pattern.

Scott Murray suggested (from theory underlying the laser) that when a
photon is randomly emitted it acts like a catalyst and stimulates the
emission of a mass of other photons. This would suggest that light from
a filament is not in fact random but in short bursts of coherent light
with no fixed phase relationship between one bust and the next but a
fixed relationship within the burst.

If one takes the sinusoidal wave tank model then one expects to get
constructive interference if the path difference is n wavelengths (where
n is an integer) and destructive interference if the path difference is
(n+0.5)wavelengths. If my suggestion is correct there would be a maximum
value for n beyond which this would not be true. i.e. when n is
sufficiently large that you start mixing different uncoherent bursts
together. This in turn would give a rough idea of how long a burst
lasts.

I may of course in my ignorance be describing well know phenomena.
Comments please.