 |
Description  |
|
|
The present invention pertains to a coherent radiation discrimination
system. More particularly, it relates to a system capable of detecting and
indicating the presence of coherent light such as that produced by a
laser.
With increased usage of lasers, it has become desirable for a number of
different reasons to be able to detect and indicate the presence of laser
radiation in a given place. For example, considerable concern has been
expressed in the literature with respect to the damage which may occur to
the eyesight of a person who inadvertently looks into a laser beam. In a
factory environment, a number of lasers may be operating throughout the
facility and, as a practical matter, it may be difficult to limit access
to the facility only to those thoroughly cognizant of the danger and
knowledgeable as to the location of the lasers. Different ones of the
lasers may be in use at any given time and certain of them may be operated
only occasionally so that not all personnel are made constantly aware of
their existence. The hazard is analogous to that in using X-rays.
Of course, various danger alarms or signals may be paralleled with the
primary switch of each laser, but even these are subject to being
disconnected in connection with routine maintenance, movement of the
lasers to different positions, and so forth. Accordingly, it is desirable
to have an indication system completely independent of the laser itself
and of its control apparatus.
Moreover, presently available lasers operate at a number of different and
comparatively widely-separated wavelengths. Typical examples are those
operating at 0.6328, 0.848, 1.06, 1.08 and 1.15 microns. Further
complicating the situation is the fact that some lasers, such as those
operating at the last three wavelengths mentioned, produce light only in
the infrared spectrum which is not visible to the human eye. This renders
it necessary to be able to detect laser light over a wide spectrum and in
both the visible and invisible portions thereof.
An individual who has complete control over the laser in question may
employ any of several known techniques to provide a warning in a given
area that the laser is operating. As one example, it is known to the art
to derive a sample from the laser beam against which protection is desired
and feed it to a photo-detector to which the beam from a second laser also
is applied. The latter laser in this case may be essentially completely
shielded so as to avoid any problem with its radiation. The detector in
effect heterodynes the two light beams and develops an output signal equal
to any difference in frequency or phase between the respective beams. This
output signal then can be utilized for an indicator or alarm. However,
such a system is of no avail when the laser against which protection is
needed is operated by a third party. With increasing power available from
lasers, it is possible to cause a laser beam to travel substantial
distances as from one portion of a factory to another or even possibly
into a neighboring building. Thus, to provide complete protection in all
of these different situations, it is necessary that the detection system
to be employed be completely independent from the laser protected against,
be capable of detecting laser light at a variety of wavelengths and
directions, be sufficiently sensitive to detect even very small amounts of
laser light power and be capable of distinguishing from a wide variety of
background radiation.
It accordingly is a general object of the present invention to provide a
radiation discrimination system which avoids all of the problems noted
while achieving all of the desired ends mentioned.
A more specific object of the present invention is to provide a system for
detecting coherent radiation and for indicating its presence only on the
basis of its coherency.
A further object of the present invention is to provide a system capable of
achieving the foregoing aims and objectives and yet which is comparatively
simple of manufacture and operation.
A radiation discrimination system in accordance with the present invention
includes a spaced pair of parallel partially reflective mirrors which
together define an interference filter for radiation of predetermined
wavelength and direction dependent upon the spacing between the mirrors.
Coupled to at least one of the mirrors is a driving source for varying the
spacing between the mirrors and correspondingly varying the transmission
wavelength of the filter for a given incidence angle. Disposed to one side
of the interference filter is a detector responsive to radiation
transmitted by the filter for developing a detection signal. Finally, the
system includes means responsive to the detection signal and a signal
representative of the action of the driving source for developing an
indication effect only upon transmission through the interference filter
of coherent radiation.
The features of the present invention which are believed to be novel are
set forth with particularity in the appended claims. The invention,
together with further objects and advantages thereof, may best be
understood by reference to the following description taken in connection
with the accompanying drawings, in the several figures of which like
reference numerals identify like elements, and in which:
FIG. 1 is a schematic diagram of a simplified embodiment of a light
discrimination system;
FIG. 2 is a cross-sectional side-elevational view of an intereference
filter which may be employed in the system of FIG. 1;
FIG. 3 is a schematic diagram of one complete embodiment of the present
invention; and
FIG. 4 is a schematic diagram of an embodiment alternative to that of FIG.
3.
In FIG. 1, an interference filter 10 is composed of a pair of
partially-reflective parallel mirrors 11 and 12 spaced apart by a distance
d and for purposes of explanation symmetrical about an axis of normal
incidence 13. Mirrors 11, 12 are movable relatively toward or away from
one another while maintaining parallelism. Disposed outwardly from mirror
11 is a detector 14 capable of developing an electrical signal in response
to light from a spot 15 transmitted through a section of interference
filter 10 having a diameter D. Spot 15 represents an area of light
reflected or scattered from a diffuse surface such as the interior wall
surface of a building. The light scattered from spot 15 is coherent,
having been produced by a laser beam.
As is well known, an interference filter usually is constructed of mirrors
having an extremely high reflectivity. Light is transmitted through the
filter only when the spacing between the mirrors and the index of
refraction of any material therebetween is such that the effective optical
path length through the filter is one-half the wavelength of the light or
an integral multiple thereof. Such a conventional interference filter has
response transmission peaks which are extremely narrow in terms of
wavelength. However, as used in the systems herein to be described,
mirrors 11, 12 preferably have a much lower reflectivity than typical as a
result of which the transmission response peaks are substantially
broadened. With the individual mirror reflectance of the order of 40
percent, for example, a plot of the optical transmission power against
wavelength for a fixed spacing d, or against the spacing d for a fixed
wavelength, reveals a somewhat sinusoidally shaped curve varying between
approximately 100 percent transmission and a percentage transmission the
same as would be obtained when no interference effects are present so that
the mirrors merely attenuate the light.
When coherent light from spot 15 falls upon interference filter 10, light
transmitted through the filter appears in a pattern composed of a central
disc surrounded by a plurality of rings of successively increasing
diameter. As discussed in Fundamentals of Optics by Jenkins and White,
Third Edition, McGraw-Hill Book Co., Inc. (1957) p. 277, the points of
maximum transmission in the ring pattern are consistent with the
expression:
m.lambda. = 2d cos .theta., (1)
where m is the order of the ring, .lambda. is the wavelength of the light
in microns, and .theta. is the angle between axis 13 and the direction of
the incident radiation. A typical response pattern of the rings is
illustrated by dashed undulating curve to the right of mirror 11 in FIG.
1. To illustrate the relationships involved, with mirrors 11 and 12 having
equal reflectances of one-half, assume they are separated by a distance d
of 100 wavelengths of the incident radiation which has a wavelength of 1
micron. Thus, for 100 wavelengths the mirror separation is about 4 mils.
From equation (1) it is determined that on axis 13 (.theta. =0) the
central disk is of the 200th order. It may be noted that the order m goes
to smaller integers in the direction lateral to axis 13; that is, the next
bright ring in the output pattern corresponds to an m of 199. This
indicates that the diameter of the central disk subtends an angle from the
source such that cos .theta. = 199/200.
With the radiation from spot 15 traveling a distance R to the interference
filter, the ratio R/D represents the f number of the detector in object
space. This value is also equal to 1/d.theta. . Thus, for a detector
diameter D of 2 cm and a distance R of 10 feet or 300 cm, the f number is
the comparatively large value of 150. The first derivative of equation (1)
is:
.delta.m= (- 2d/.lambda. sin .theta..delta..theta. (2)
For detection of a sector of a single band by the detector placed off axis
13, .delta.m= -1. From equation (2), it is found that the maximum angle
for which this illustrative device is effective is in terms of a value for
.theta. in the order of .+-.48.degree.. Even though this is a relatively
large field of view, it will be seen hereafter that the approach is
capable of enabling a still much larger field of view. The figures given
heretofore are used only for convenience of basic illustration.
The ring patterns are developed in the output of the interference filter
only if the light is coherent at a wavelength in a transmissive region of
the filter response curve. Consequently, the filter inherently
discriminates against ordinary incoherent background light. In order to
detect the presence of coherent light in the field of view, the mirror
separation d is varied by at least one-half the wavelength of light
throughout the spectral region under observation. In practice, the actual
physical distance between the mirrors need not be changed. Instead, the
spacing may be effectively varied by altering the refractive index of a
material between the mirror surfaces. It is known that this can be
accomplished by forming the mirrors on opposing surfaces of an
electro-optic dielectric material across which an electric field is
applied. The index of refraction in such a material varies as the
magnitude of the field is changed. Other materials respond in a similar
manner to heat. It is also possible to instantaneously change the
thickness of the material by launching acoustic waves into it.
As embodied herein, the spacing between mirrors 11 and 12 is varied
electromagnetically. In FIG. 2, an electromagnetic driving apparatus
includes a core of U-shaped cross-section having a coaxial inner and outer
leg portions 17 and 18. The free end of outer leg 18 is slightly elevated
over the free end of inner leg 17 between which photodetector 14 is
disposed. Clamped by a spacer ring 19 to the outer core leg end is an
annular disk 20 of flat, thin tool steel such as that from which saw
blades are made. The inner peripheral edge of disk 20 is concentric with
and approximately oriented above the inner leg 17. Annulus 20 is radially
serrated from its inner perimeter so as to create a plurality of
electromagnetic armatures in the form of individual strips. Nested within
legs 17 and 18 is a coil 22. Mirror 11 is affixed to disk 20 in a position
covering the opening therein. Secured on the top side of ring 19 are a
plurality of adjustable spacers 24 on the upper side of which is secured a
clamping ring 25 the inner circumference of which carries mirror 12.
Spacers 24 are split horizontally to a substantial depth and include a
tapered hole bottomed by a threaded section which receives a threaded slug
27. The depth of penetration of the slugs is adjustable for the purpose of
aligning mirror 12 parallel with mirror 11. In practice it has been found
that three adjustable spacers distributed circumferentially around the
assembly 120.degree. apart perform satisfactorily.
In operation, energization of coil 22 flexes disk 20 to move mirror 11
relative to mirror 12 while maintaining parallelism between the mirrors.
Moreover, the primary energizing signal is alternating, but a
direct-current bias also is caused to flow in coil 22 so that the
alternating flux magnitude in the gaps between the core and annulus 20
varies about a point on a generally linear portion of the magnetic
characteristic. This also avoids possible ambiguity in polarization
between different coils when more than one is used as in push-pull
arrangements.
Referring again to FIG. 1, detection of the rings produced by a source of
coherent light is accomplished by varying the separation between mirrors
11, 12 by a minimum of one-half wavelength. In one system which has been
operated, this variation constituted a linear motion several wavelengths
long. For a constant linear motion of mirror 11, the frequency observed in
the signal developed by detector 14 is determined by the wavelength
.lambda. and the angle of incidence .theta.. Consequently, in that system
it is necessary to monitor a bandwidth somewhat over 1 octave wide,
although the necessary instantaneous bandwidth is extremely low and
typically of the order of only a few cycles per second. The detector
current exhibits a signal-to-noise ratio in excess of ten when observing a
spot ten feet away produced by a 0.6328 micron He-Ne laser having an
output power of less than 1 milliwatt.
It is preferred, however, to change the mirror spacing d sinusoidally. In
this manner, when spot 15 is produced by light of 1 micron wavelength
(again a convenient value for illustration), a total excursion of 5
wavelengths results in the production of a 10 cycle-per-second signal by
detector 14. In consequence of the sinusoidal motion, the detector
produces a frequency modulated signal having side-bands which are
multiples of the mechanical driving frequency up to a limit which is
determined by the modulation index. It can be shown that the amplitide of
these sidebands is given by the appropriate Bessel function and, depending
upon such parameters as wavelengths, angle of incidence, and amplitude of
vibration, there are different conditions when particular sidebands can
and cannot exist. Therefore, if the signal processing circuitry is of the
simplest variety which merely selects a single one of the sidebands, it is
possible for the apparatus to have blind spots. There are, however, pairs
of sidebands which cannot go to zero simultaneously. One such pair is
composed of the second and third harmonics of the fundamental frequency.
Accordingly, it is contemplated to take advantage of the existence of the
detected f.m. signal spectrum for the purpose of producing an indication
signal while at the same time taking steps to insure against the existence
of blind spots. Moreover, the systems to be described also take advantage
of the fact that the particular sidebands existing in the signal are a
function of the source which drives mirror 11 to vary the mirror
separation. As a result, those sidebands are independent of both the
wavelength of the incident radiation and of its angle of origin. A change
in the angle of origin does alter the modulation index and the
distribution of the sidebands with increased higher order sidebands
occurring upon movement of the origin away from axis 13.
One such approach is illustrated in FIG. 3 wherein interference filter 10,
detector 14 and driving coil 22 are represented schematically. Coil 22 is
driven by signals from an oscillator 29 which also feeds a frequency
doubler 30 and a frequency tripler 31. A first highly selective amplifier
32 is tuned to the second harmonic of the signal from oscillator 29 and
feeds an amplified signal to a synchronous detector 33. Similarly, another
sharply selective amplifier 34 responds to the third harmonic in the
signal from detector 14 and feeds an amplified signal to a synchronous
detector 35. The output signals from detectors 33 and 35 are fed in
parallel to an indicator 36 which may be simply an alarm of either audible
or visual nature. The alarm may be located at a remote position, if
desired. Doubler 30 and tripler 31 each include adjustable phase delay
networks so that the respective signals they feed to the detectors are in
phase with the respective carrier signal harmonics derived by amplifiers
32 and 34.
Except for the purpose of eliminating possible blind spots as mentioned
above, the basic principles are illustrated by considering the action only
of amplifier 32, double 30 and synchronous detector 33. The system
including the other synchronous detector 35 works the same way except for
operation on the third harmonic; indicator 36 responds to an output signal
from either synchronous detector. In one system using this approach, the
nominal spacing between the elements of interference filter 10 is
approximately 0.004" and oscillator 29 varies that spacing at a rate of
approximately 300 cycles. Amplifier 32 thus is sharply selective of the
second harmonic 600 cps signal. Correspondingly, doubler 30 develops a
signal also of 600 cps and the two 600 cps signals are then fed into
synchronous detector 33 in phase. The output signal from the synchronous
detector exhibits an extremely high signal-to-noise ratio, with a very
narrow bandwidth, the bandwith used in the particular system being in the
order of 1 cycle. Comparable performance is obtained at several other
harmonics and the system readily detects a laser beam spot produced by a
laser output of 1 milliwatt at a range of 20 feet onto a scattering
surface. Practical mechanical design limits this particular system to a
field of view angle of about .+-.30.degree., although a .+-.45.degree.
field of view at such a range is well within capability.
With respect to the light viewed by detector 14, it is to be observed that
in the presence of coherent radiation a complex spectrum of harmonics of
the modulation frequency appears in the detector output signal. At the
same time, there are no such harmonics unless coherent radiation is
present in the field of view and the optical passband of the interference
filter. Consequently, incoherent background radiation is discriminated
against by the technique. As described, the driving signal applied to the
mirrors is the fundamental reference signal and this permits independent
synchronous detection to be employed for any of the harmonics of the
driving frequency. This makes available arbitrarily small post-detection
bandwith.
It is desired that the peak-to-valley intensity of transmission difference
of the interference filter be at least 50% so that each of the two mirrors
should have an individual reflectance of approximately 40%; it is
preferred that they maintain at least 30% transmission over a full octave
in wavelength. In one successful application, the mirrors were fabricated
of rutile on a quartz substrate, although the transmission bandwidth is
slightly limited. A mirror material providing at least 30% reflectivity
over a very large transmission range is the nickel alloy known as Inconel.
It is of course preferred that the photosensitive element of detector 14
have adequate wavelength response over the entire range, such as from 0.6
to 1.4 microns at room temperature, a sufficiently large sensitive area to
be consistent with a large field of view, and adequate response time.
While a silicon-junction photodetector has provided satisfactory response
up to about 1.08 microns, the presently preferred detector utilizes lead
sulfide as the active element and exhibits a room temperature response
which extends well beyond 2 microns. Moreover, the lead sulfide
photodetector has substantial detectivity for modulation components well
beyond 1500 cycles per second, consistent with the driving-frequency
harmonics involved.
The choice of driving frequency influences the sensitivity of the system.
The driving signal frequency is limited at the low end by the desire to
avoid 1/f noise which originates in the photodetector, amplifier and
signal detector circuits. It is limited on the high end by the necessity
for an adequate detectivity at the frequency corresponding to the highest
harmonic of the modulation frequency being detected. Also, operation at
harmonics of the local power line frequency is to be avoided.
A different approach to the signal processing and additional features are
included in the alternative and preferred embodiment of FIG. 4 wherein
interference filter 10, detector 14 and driving coil 22 for tuning the
interference filter are indicated schematically the same as in FIG. 3. At
the outset, the FIG. 4 system represents an improvement by virtue of the
inclusion of a second photodetector element 40 identical with the element
in detector 14 except that it is exposed only to the background radiation
commonly falling on both detectors 14 and 40. The two detectors 14, 40 are
coupled in push-pull across the split primary winding 41 of a transformer
42 the secondary winding 43 of which is coupled to the input terminals of
a differential amplifier 44. Consequently, in the absence of a coherent
light ring selected by interference filter 10, detectors 14 and 40 produce
only identical background signals which are cancelled by the action of
transformer 42 and differential amplifier 44.
The signal output from amplifier 44 is passed through a sharply selective
filter 45 in this case peaked at 210 cps and having a passband of only
about 6 cps. The signal output from filter 45 is further amplified in an
amplifier 46 from which it is fed to a synchronous detector 47. The
driving signals for the interference filter are developed by an oscillator
48 which in this case generates a driving signal of 105 cps which is fed
through a highly selective filter 49 and raised in level by an amplifier
50. From amplifier 50, the driving signal is fed through a selected one of
two resistors 51, 52 by way of a switch 53 to the input winding 54 of a
transformer 55. From the secondary winding 56 of the latter the driving
signal is fed through the one of another pair of resistors 57 and 58
selected by a switch 59. From the selected one of the latter resistors,
the signal is fed by way of a DC potential source 60 to coil 22. Source 60
is for the purpose of applying the static bias to coil 22 as discussed
above with respect to FIG. 2.
A second 105 cycle-per-second signal from oscillator 48 is fed through a
frequency doubler 61 and a filter 62, sharply selective of the resultant
210 cps signal, to a trigger generator 63 by way of a phase shifter 64. In
this case, generator 63 is of the well-known Schmidt variety, producing
one pulse in phase correspondence with each cycle of the reference
oscillation.
As in the embodiment of FIG. 3, the resultant demodulation component from
detector 47 is fed to an indicator 66, in this instance by way of an
amplifier 65 and an extremely low-frequency low-pass filter 67 composed of
a series resistor 68 and a shunt capacitor 69. Filter 67 has a time
constant of about 10 seconds so as to have a bandwidth of about 1/30th of
a cycle per second. It will be observed that the noise bandwith therefore
is restricted to the small value of 1/15 of a cycle per second.
In practice, and in view of theoretical reasons to be discussed, it has
been found that the wavelength range may be extended still further by
sequentially switching between two different amplitudes of the driving
signal that varies the spacing of the interferometer mirrors 11, 12; that
is, the range of wavelengths scanned by filter 10 is periodically varied.
At the same time, a similar shift in the average spacing of mirrors 11 and
12 is incorporated into the system to shift the position of nulls in the
sensitivity pattern of the output of the interference filter and thus to
avoid the previously-mentioned blind spots; this is achieved by
periodically changing the static bias applied to coil 22. With this latter
approach, observation at only one harmonic is necessary. To these ends,
the system of FIG. 4 further includes a timer 70 which operates switches
53 and 59. While a variety of timing mechanisms may be used for this
purpose, it is convenient simply to employ a rotating cam shaft with the
cam surfaces for the respective armatures of switches 53 and 59 correlated
so that each of the four possible switch combinations is sequentially
chosen. Such sequencing enables maximization of the amplitude of the
observed harmonic throughout a very wide wavelength range while at the
same time protecting against blind spots. In a typical embodiment, the
duration of each sequence is approximately 45 seconds so that a full scan
takes three minutes.
In each of the systems discussed, the basic approach is that of using a
photodetector to examine a region of the interference filter where a
fringe ring occurs upon the presence of coherent light having a frequency
appropriate to be passed by the filter and then tuning the filter so as
effectively to scan a wide range of possible wavelengths. The
characteristic ring pattern appears and hence an output indication is
given from the system only when coherent radiation within the designed
wavelength range is present. While particular emphasis has been given
herein to the detection of laser radiation, it is to be noted that the
systems are likewise applicable in other environments. For example,
certain stars emit particular characteristic lines in their spectrum and
the principles disclosed are applicable therefore to star acquisition and
tracking systems.
To understand more clearly the design parameters of the present system, it
may be helpful to examine in more detail the transmission characteristic
of the interference filter itself. The percentage transmittance T as a
function of plate separation of a Fabry-Perot type interferometer as here
employed with medium to low surface reflectance can be expressed:
##EQU1##
where T.sub.o is the mean interferometer transmittance averaged over one
cycle of the quantity cos (4.tau..delta. cos .theta./.lambda.), .alpha. is
the peak to zero excursion of transmittance as the value .delta. cos
.theta. increases by 1/8 of a wavelength, .delta. is the interferometer
plate separation measured normal to the reflecting surfaces, .lambda. is
the wavelength of the transmitted radiation and .theta. is the angle of
incidence of the radiation on the interferometer. Equation (3) assumes
that the plates are parallel and that the incident radiation is
monochromatic and collimated. Also, the higher order cosine terms are
neglected since these become significant only for higher surface
reflectances than those of the embodied apparatus. The lower reflectance
values actually utilized maximize the magnitude of .alpha..
When .delta. is given a time-varying modulation and the driving signal is
at cos .omega..sub.c t, the transmission T can be expressed in the form:
##EQU2##
in which .DELTA. is the mean normal interferometer spacing, .delta..sub.o
is the peak-to-zero spacing modulation measured normal to the mirrors and
.omega..sub.c is the modulation frequency. Equation (4) can be expanded
to:
##EQU3##
In turn, equation (5) can be further expanded to:
##EQU4##
In equation (6), .phi..sub.o = 4.pi. cos .theta..DELTA./.lambda. minus
the interferometer phase thickness and .theta..sub.o is the modulation
index which is equal to the value 4.pi.cos .theta..delta..sub.o /.lambda.
minus the peak-to-zero phase modulation. With the system operating only on
the second harmonic of .omega..sub.c as in FIG. 4, for example, the
effective transmittance can be written in the form
T= T.sub.o .alpha. cos .phi..sub.o 2 J.sub.2 (.theta..sub.o) cos
2.omega..sub.c t (7)
Examination of equation (7), and as verified by actual operation, reveals
that it is the cos .phi..sub.o term which results in sensitivity nulls in
the field of view and this is the reason for either shifting the dc bias
as in FIG. 4 or also looking at the third harmonic signal as in FIG. 3.
The .theta..sub.o term determines the effective wavelength range. Again,
it is to extend this range that the sequential change between resistors
51, 52 to in turn change the amplitude of the driving signal is
incorporated into the FIG. 4 system.
Thus, the disclosed apparatus affords unique and useful information
concerning the presence of absence of light anywhere within a wide
spectral range by observing the particular characteristics of coherence.
Its most widespread application, then, is to detect the presence within a
given area of laser radiation. The disclosed systems are equally suitable
for the detection of either visible or invisible radiation.
While particular embodiments of the invention have been shown and
described, it will be obvious to those skilled in the art that changes and
modifications may be made without departing from the invention in its
broader aspects, and, therefore, the aim in the appended claims is to
cover all such changes and modifications as fall within the true spirit
and scope of the invention.
* * * * *
|
|
 |
|
 |
Description |
|
