U.S. patent number 3,568,692 [Application Number 04/708,456] was granted by the patent office on 1971-03-09 for optical machining process.
This patent grant is currently assigned to Bowles Engineering Corporation. Invention is credited to George D. Hinschelwood, Eric E. Metzger.
United States Patent |
3,568,692 |
Metzger , et al. |
March 9, 1971 |
OPTICAL MACHINING PROCESS
Abstract
A fluid amplifier includes a rigid base, a layer of
light-absorptive adhesive overlying a surface of the base and a
layer of light-polymerizable plastic bonded to the base by the
adhesive; the light-polymerizable material being solid in the
unpolymerized state and in the polymerized state having a plurality
of channels and nozzles formed therein to define a fluidic element.
A cover plate seals the channels and nozzles to complete the
element.
Inventors: |
Metzger; Eric E. (Silver
Spring, MD), Hinschelwood; George D. (Washington, DC) |
Assignee: |
Bowles Engineering Corporation
(Silver Spring, MD)
|
Family
ID: |
24845856 |
Appl.
No.: |
04/708,456 |
Filed: |
November 27, 1967 |
Related U.S. Patent Documents
|
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
Issue Date |
|
|
219168 |
Aug 24, 1962 |
|
|
|
|
Current U.S.
Class: |
137/827;
137/833 |
Current CPC
Class: |
F15C
5/00 (20130101); Y10T 137/2224 (20150401); Y10T
137/2191 (20150401) |
Current International
Class: |
F15C
5/00 (20060101); F15c 005/00 () |
Field of
Search: |
;137/81.5
;204/159.23 |
References Cited
[Referenced By]
U.S. Patent Documents
Primary Examiner: Scott; Samuel
Parent Case Text
This application is a division of our application, Ser. No.
219,168, filed Aug. 24, 1962, and entitled "Optical Machining
Process."
Claims
We claim:
1. A pure fluid amplifier or logic element comprising a metallic
backing plate, a lower uniform layer of light-absorptive material
covering a surface of said backing plate, an upper uniform layer of
light polymerized material overlying said lower layer, and having a
plurality of interrelated channels and nozzles formed therein and
extending therethrough and to said lower layers, the walls defining
said channels and nozzles forming an angle relative to the plane of
said layers of not more than 5.degree..
2. The combination according to claim 1 wherein said angle is not
more than 2.degree..
3. A pure fluid amplifier or the like comprising a body of
light-polymerized material having appropriate channels and nozzles
formed therein said amplifier having been formed with a fluid
resistor located in one of said channels and comprising a plurality
of cylindrical closely spaced stalks of said light-polymerized
material, said plurality of stalks extending across said one of
said channels from the bottom to the top thereof, the walls
defining said channels, nozzles and stalks forming an angle
relative to the plane of said layers of not more than
2.degree..
4. A pure fluid amplifier or logic element comprising a unitary
structure including a metallic backing plate, a lower uniform layer
of light-absorptive material covering a surface of said backing
plate, an upper uniform layer of light polymerized material
overlying said lower layer and having a plurality of interrelated
channels and nozzles formed therein and extending therethrough and
to said lower layer, and a cover plate sealed to the surface of
said upper layer remote from said backing plate.
5. The combination according to claim 11 further comprising a
plurality of threaded apertures extending through said metallic
backing plate into registry with preselected regions of said
channels, and fluid flow connectors threaded into said
apertures.
6. The combination according to claim 4 wherein said
light-polymerized material is a solid in the unpolymerized
state.
7. A pure fluid amplifier or logic element comprising a unitary
structure having a rigid backing plate, a lower uniform layer of
light absorptive material covering and secured to a surface of said
backing plate, an upper uniform layer of light polymerized material
overlying and secured to said lower layer and having a plurality of
interrelated channels and nozzles formed therein and extending
therethrough to said lower uniform layer, and a cover plate sealed
to the surface of said upper layer remote from said rigid backing
plate, a plurality of apertures extending through said rigid
backing plate into registry with preselected regions of said
channels and fluid flow connectors secured in said apertures.
Description
The present invention relates to fluid amplifier devices and more
particularly to methods of making fluid amplifiers and the articles
resulting from such method.
The term "fluid amplifier" as used herein refers to a recent
development in fluid systems in which amplification of one or more
of the parameters of a flowing stream can be effected in an
apparatus employing no moving parts. A typical example of such a
device is an apparatus having fluid supplied to a power nozzle
which issues a stream of fluid toward the apex of a divider located
downstream from the nozzle. Control nozzles are disposed on
opposite sides of the stream closely adjacent to the power nozzle
and upon the issuance of fluid streams from the control nozzles,
the main stream is deflected from its center position causing more
(or all) of the fluid to flow along one side of the divider than
along the other side of the divider. The energy, pressure, or mass
flow supplied to the control nozzle is less than the change in the
corresponding parameter in the side of the divider to which the
main stream is deflected. In consequence, the apparatus provides a
gain of output signal over input signal and may be classified as an
amplifier.
In order to provide significant amplification in such a system, the
region of interaction between the main stream and the control
stream or streams is normally confined between top and bottom
plates so that the main stream is confined, at least in the region
of interaction, to its plane of deflection. As such the main stream
appears as a deflectable divider passing through the interaction
region and when a control stream impinges thereupon, it cannot flow
around or through the main stream and therefore deflects the main
stream.
An amplifier of this type may function in numerous ways and for a
more detailed description of the various forms which such elements
may take, references is made to Reilly U.S. Pat. No. 3,030,979 for
Induction Fluid Amplifier, issued Apr. 24, l962, Wadey U.S. Pat.
No. 3,005,533, for Fluid Keyboard Using Jet-Pipe Valves, issued
Oct. 24, 1961, Wadey U.S. Pat. No. 3,034,628, for Pneumatic
Keyboard, issued May 15, 1962, U.S. Pats. No. 3,024,805 for
Negative Feedback Fluid Amplifier by B. M. Horton issued March 13,
1962; No. 3,016,066 for Fluid Oscillator by R. W. Warren issued
Jan. 9, 1962; No. 3,001,698 for Fluid Pulse Converter by R. W.
Warren issued Sept. 26, 1961; No. 3,004,547 for Bounded Jet Fluid
Amplifiers by H. Hurvitz issued Oct. 17, 1961; No. 3,001,539 for
Suction Amplifier by H. Hurvitz issued Sept. 26, 1961.
Briefly summarizing the various types of devices which may be
realized by the apparatus described above, the units are capable of
operation as analogue amplifiers per se, or amplifiers with
positive or negative feedback, bistable devices, oscillators and
may be incorporated in systems for approximating many of the
functions now performed substantially only by electronic circuits.
Thus, analogue amplifiers may be cascaded to provide high-gain
units or may employ varying amounts of feedback to provide units of
high stability and low noise or may employ various passive elements
in the feedback loops to provide narrow-band or wide-band
amplifiers or may employ positive feedback with a loop gain of less
than one, to provide high-gain amplifiers. The bistable elements on
the other hand may be combined with fluid logic elements to provide
pulse counters, shift registers and other logical gating and
control circuitry.
It is apparent from the above that if these fluid elements are to
be able to perform the various functions intended, they must have
long term stability, which means low drift in the absence of or in
the presence of a sustained input signal, must be relatively noise
free, must not generate spurious signals or pulses particularly
when employed in pulse logic and the engineer who wishes to employ
such devices in a system must be able to provide a design which
when reduced to a physical device performs in the manner
anticipated. This, of course, is analogous to design in electronic
circuits where an engineer selects a particular tube and then
designs a circuit including the values of resistors, capacitors,
inductors or other elements necessary for utilization with the tube
to effect the desired function. It is necessary for the
manufacturers of the tubes and resistors and other passive or
active elements to be able to produce these elements on a mass
basis while maintaining their desired functional
characteristics.
The fluid amplifier must also be capable of performance in the
manner for which it is designed. However, there are a number of
critical parameters in such systems which have rendered the
fabrication of the apparatus to a particular design difficult and
expensive when the prior art techniques are employed. One of the
difficulties which is encountered in the mass fabrication of fluid
amplifiers to specific tolerances is the maintenance of laminar
flow in the system. Turbulence in any fluid system is a random
phenomenon which becomes lesser or greater on a completely random
basis. Turbulence may introduce sufficient noise to obliterate or
completely mask a signal in an analogue system after several stages
of amplification or may alternatively or concurrently introduce
long term instability in the beam, which is reflected as a drift in
the absence of a signal or in the presence of a long term signal.
In order to minimize turbulence in fluid systems, and to maintain
substantially laminar flow throughout such a device, or a system in
which the device is included, there must be no abrupt
discontinuities in the wall and top and bottom plates of the
apparatus and the side and end walls must be extremely smooth.
Since the passages in devices of this type are long relative to the
cross-sectional dimensions of the device even minor imperfections
in the surface of the walls or grainness therein may introduce
sufficient turbulence to have serious consequences on the operation
of such a device.
Another important design parameter of such systems relates to the
dimensions of the various channels forming the fluid amplifier. The
sizes of the various nozzles are quite critical. In a specific
system, the relative size of the main nozzle as opposed to the
control nozzle or nozzles determines the gain of the system.
Therefore, when designing a system with a specific gain it is
necessary in fabricating such a device to have the nozzle
cross-sectional areas to conform closely to the design size. Also
the termination of a nozzle at the interaction region is a critical
parameter since an imperfection in the geometry in this area may
produce undue spreading of the beam which will prevent complete
switching of the beam in a bistable device or digital circuitry or
produce a sluggish or improper response of the amplifier when
employed in analogue systems. A further consideration relative to
size of channels is the slope of the sidewalls. In most instances,
it is desirable to have substantially vertical walls with no more
than a 5.degree. convergence toward the center of the channel and
preferably a convergence of 2.degree. to 3.degree. for each wall.
Greater variations in tolerances produce devices which do not
conform to their design performance with sufficient accuracy for
wide commercial applicability.
Design of fluid equivalents of electrical capacitors and inductors,
which are known as inertances and capacitances, are relatively
simple in that the inertance constitutes a channel which is long
relative to its cross-sectional dimensions and a capacitance is
merely an enlarged region in a channel. The realization, however,
of a fluid resistor, in systems of this type where no moving parts
are employed, is more difficult and in the past resort has been had
to porous plugs which are cut to size to fit in a channel and
having a length which, taken in consideration with the porosity of
the plug, provides the necessary fluid resistance. Such a technique
for providing fluid resistors is not always satisfactory since
material of the proper porosity when taken in consideration with
the length of the channel available for insertion of the porous
material may not provide the actual resistance desired.
Several techniques for fabricating fluid amplifiers and systems
have been previously developed but are expensive and time consuming
and in certain instances do not provide the characteristics
required for low-noise, longterm stability, etc. as set forth
above. One such technique is conventional machining in metal or
plastic which provides excellent results in larger units but is,
necessarily, time consuming and expensive. On the other hand when
employing metal machining on small units where orifices may be only
a few mils wide and deep, the devices are often extremely difficult
if not impossible to fabricate.
Another technique previously employed for fabricating such devices
is a photoglass etching technique in which a photo resist is
applied to a glass plate and the negative of the fluid amplifier
optically projected onto the photo resist. Thereafter acid is
applied to the plate and the glass is etched where the photoresist
has not been struck by light and is therefore not developed. This
technique is also quite slow and has the disadvantage that the
channels often have sloping sides due to under cutting of the glass
behind the developed photoresist. Also graininess may result from
such a technique unless extreme care is exercised and even then it
cannot always be eliminated. Such graininess tends, as indicated
above, to introduce turbulence into the system and severely limits
the ability to cascade fluid amplifiers to form large operational
units.
The difficulty of providing fluid resistors particularly in small
units has been indicated previously in the above discussion.
It is therefore a broad object of the present invention to provide
a method of fabricating fluid amplifiers or pure fluid systems
which is quick, economical, may be practices by unskilled personnel
and which provides a finished unit conforming closely to the
performance desired by the designer.
It is another object of the present invention to provide a method
for fabricating pure fluid elements which have substantially
vertical and extremely smooth sidewalls, and have channels of
uniform depth and which method is equally applicable to small as
well as medium size units.
It is another object of the present invention to provide a method
of fabricating pure fluid elements in which fluid resistors may be
fabricated as a part of the fabrication process for the basic
element and in which the porosity of the element may be selected
from a wide range of resistance values.
In accordance with the present invention it has been discovered
that fluid amplifiers having all of the desirable characteristics
set forth above may be made by a process including a
photopolymerization step. More specifically a photopolymerizable
plastic or substance, which forms a uniform layer on a base
material, has a photographic negative or positive, depending upon
the type of device to be fabricated, laid over and in contact with
its exposed surface. The areas which are to be ultimately removed
from the polymerizable material appear on the negative as
completely opaque regions whereas the areas which are to remain are
represented on the negative by substantially completely transparent
regions. An appropriate source of light provides a bundle of
substantially parallel rays which are perpendicular to the surface
of the plastic. The light source is selected to produce
polymerization of the plastic upon which the rays are incident.
After complete exposure is effected the plastic plate is washed in
a solution in which the unpolymerized material is soluble. This
causes the unpolymerized material to be removed and the plastic
remaining forms the body in which the channels of the amplifier are
cut. Thereafter a cover plate is bonded to or otherwise maintained
in fluid sealing relationship with the exposed surface of the
plastic thereby providing independent channels through the plastic
which form the pure fluid amplifier. Materials suitable for such a
process are disclosed in the U.S. Pat. to Plambeck No. 2,760,863
issued Aug. 28, 1956, among others.
The process described above has a number of important advantages
over the prior art processes which advantages however establish
several critical parameters in the system to obtain the full
benefit of the process. Unlike a photoetch technique employed on
glass, the solvents employed in the present process do not attack
the polymerized plastic and therefore undercutting of the pattern
is not a serious problem as it is where a material is used that
attacks the basic constituent of the plate. In consequence, the
sides of the channels formed in the plate maintain an angle as
determined substantially by the angle of the rays from the light
source relative to the surface of the plate and the degree of
parallelism of the rays. The design of the optical system employed
in this process therefore is relatively critical. It has been found
that the rays of the source should have a divergence angle of no
greater than 5.degree. and no less than 1.degree.. The tolerance in
sidewall slope provided by such a source of light falls within the
design tolerances of systems of this type.
Another critical factor in determining the degree of divergence of
light rays permissible in a system of this type is determined by
the depth of penetration of the light relative to the width of the
channels. This problem becomes particularly critical when forming
nozzles and fluid resistors as integral parts of the design. In
accordance with a further feature of the present invention, the
fluid resistors are formed by a dot screen process. In utilizing
the dot screen process for fabricating fluid resistors, the degree
of porosity in a particular region is determined and this FIG. is
converted to a FIG. representing the diameter of and spacing
between the plastic stalks extending over a predetermined length of
a passage. The channels between the stalks may be staggered to
provide the desired degree of resistance over as short a length as
possible. The dot screen is formed directly on the photographic
negative on which the remainder of the amplifier is laid out. As
indicated above the spacing between such elements, that is, between
the stalks, is ordinarily quite small and in order to insure
penetration of the light the bottom of the plastic layer the light
rays must be substantially parallel; that is parallel to within a
5.degree. divergence angle. It is of course necessary in such a
system as this for the plastic to be substantially transparent to
the activating light to provide the requisite degree of
polymerization at the base of the stalk. Such materials are
disclosed in the aforesaid patent. These materials should also be
relatively thermally inactive at temperatures to which they are
raised by the impinging light so as to prevent polymerization in
regions adjacent those through which the light passes.
Another consideration concerning the light properties of the unit
is that the base material should be substantially light-absorptive
so that that portion of the light passing completely through the
plastic is absorbed rather than scattered back into the plastic. If
the base material is metallic, as is preferable in the present
invention, then a binder for binding the plastic to the base should
be light absorptive to prevent such back reflection which would
result in spreading of the light and therefore produce
polymerization in regions where it is not desired.
A source of light for such a system is the sun, which is directed
to the surface of the plastic through a long tube having
nonreflective inner walls. That portion of the light from the sun
which is not parallel as a result of light scattering in the
earth's atmosphere, is removed by the light-absorptive inner
surface of the tube and the light emanating from the far end of the
tube at the surface of the plastic is composed substantially
completely of parallel rays.
Another light source which lends itself more readily to commercial
production is an adaptation of the optics employed in the so called
Schlierin system of photography which provides substantially
parallel bundles of light having a relatively large cross-sectional
area. The light source of the Schlierin system may comprise
numerous devices such as an optical maser (laser), an arc source,
etc. The laser has certain distinct advantages in the system of the
present invention, relating primarily to the high intensity of the
light provided by such a source. In the Schlierin system, the light
which is masked to provide a very small spot is located at the
focal point of a mirror. The light from the source is expanded to
the size of the mirror and reflected therefrom to provide a
parallel bundle of rays, for instance, 6 inches in diameter. Due to
the high intensity of the laser source, the intensity of light
across the parallel bundle of rays is still sufficiently intense to
produce rapid polymerization of the plastic. As indicated above
there is a limit to the degree of heating which the plastic can
withstand before it undergoes a certain degree of polymerization
due to thermal effects. By employing the laser to produce a high
intensity, short burst of light the time of exposure and therefore
the heating effect is minimized. If the plastic is polymerizable in
the presence of ultraviolet light, which is the type of plastic set
forth in the aforesaid patent, then, since at this writing lasers
which directly produce ultraviolet light are not available, the
laser may be employed to excite a quartz crystal which doubles the
frequency of the red light from the laser, in this case a ruby
laser, to provide the requisite ultra violet light.
It is therefore another object of the present invention to provide
a method and process for making pure fluid amplifiers employing a
photopolymer process.
It is another object of the present invention to provide a process
for making pure fluid systems having substantially vertical and
extremely smooth sidewalls and bottom walls which process employs
photopolymerization of a plastic material.
It is yet another object of the present invention to provide a
method for making pure fluid systems which method includes
illuminating a photopolymerizable material through a negative of
the apparatus to be formed in which the light source provides a
bundle of substantially parallel light rays of the proper wave
length to effect photopolymerization.
It is another object of the present invention to provide a method
for making pure fluid amplifiers having fluid resistors formed
integrally therewith which process employs a photopolymerizable
material illuminated by a substantially parallel bundle of rays
through a photo negative of the apparatus to be formed; the photo
negative providing appropriately spaced opaque or light
transmittive regions of a density such as to provide the desired
fluid resistor.
The above and still further objects, features and advantages of the
present invention will become apparent upon consideration of the
following detailed description of one specific embodiment thereof,
especially when taken in conjunction with the accompanying
drawings, wherein:
FIG. 1 is a plane view of a pure fluid amplifier to be fabricated
by the method of the present invention;
FIG. 2 is a cross view taken along line 2-2 of FIG. 1;
FIG. 3 illustrates a first step in the method of the present
invention;
FIG. 4 illustrates the produce produced by the process of which
FIG. 3 illustrates one step;
FIG. 5 illustrates the method of providing communication with the
various channels of apparatus illustrated in FIG. 4;
FIG. 6 is a schematic diagram of an optical system which may be
employed in the process of the invention;
FIG. 7 is a cross-sectional view of a fluid resistor produced by
the process of the invention;
FIG. 8 illustrates the light irradiation pattern produced by a
divergent beam; and
FIG. 9 illustrates a fluid resistor produced by employing a beam of
light having a divergence as illustrated in FIG. 8.
Referring specifically to FIGS. 1 and 2 of the accompanying
drawings there is illustrated a pure fluid amplifier of the type
with which the present invention is concerned. The amplifier is
formed as channels in a plate 1. The amplifier which is generally
designated by the reference numeral 2 comprises a power nozzle 3
adapted to be connected to a source of fluid under pressure (not
illustrated). The nozzle 3 terminates in an orifice 4 for issuing a
stream of fluid directed toward an apex 6 of a divider structure 7.
The divider 7, as illustrated in FIG. 1, is symmetrical with
respect to the center line of the orifice 4 and its apex 6 lies
along this center line at a predetermined distance downstream from
the orifice 4. The orifice 4 of the nozzle 3 is formed in an end
wall 8 of an interaction region 9 further bounded by sidewalls 11
and 12 which are nominally parallel to the left and right sidewalls
respectively of the divider 7. The amplifier 2 is provided with
control nozzles 13 and 14 having orifices 16 and 17 respectively
extending through sidewalls 11 and 12. The orifices 16 and 17 are
defined on their lower side, as viewed in FIG. 1, by the end wall 8
and therefore the stream issued thereby are perpendicular to the
undeflected position of the stream issued by the orifice 4.
The sidewall 11 and left sidewall of the divider 7 form a first
output passage 18 and the sidewall 12 and right sidewall of the
divider 7 form an output passage 19. For purposes of description
only a fluid resistor 21 is formed in the outlet passage 19 and
though formerly constituted by a plug or porous material, in
accordance with the present invention the resistor 21 constitutes a
group of generally cylindrical, upright stalks of the material
comprising the plate 1. A cover plate 22 is provided to seal the
top of the plate 1 so as to isolate the various passages in the
plate 1 except to the extent that they are interconnected due to
the intersections thereof in the plate 1.
In the type of structure illustrated in FIGS. 1 and 2, the channels
are formed by removing material from the plate 1. Appropriate
connections may be made either through the plate 1 or cover plate
22 to the source of pressurized fluid and to the various loads
connected to the outlet passages 18 and 19.
In operation of the apparatus illustrated in FIGS. 1 and 2, fluid
under pressure is supplied to the main or power nozzle 3 and issues
from the orifice 4 in a wall defined stream which when undeflected
divides equally at the apex 6 between the outlet passages 18 and
19. By determining the difference in a given flow parameter in the
channels 18 and 19 one may derive, for instance, an output signal
which in the case cited above would be a zero indicating a null
condition. The fluid may be air or water or combinations thereof or
any other fluid material and the parameters which may be measured
are for instance mass flow, energy or pressure depending upon the
dimensions of the various passages in the system. In the particular
system illustrated the apparatus would measure changes in mass
flow.
An input signal is applied to the system of FIG. 1 by applying
fluid to the nozzle 14 which issues a stream from the nozzle 17
intersecting the path of flow of the main power stream issuing from
the nozzle 4. Due to momentum interchange between the two streams,
the power stream is deflected to the left by an amount
proportionate to the momentum of the stream issuing from the
orifice 17 and in consequence the mass flow in the outlet passage
18 is increased while the mass flow in the outlet passage 19 is
decreased. This differential effect may be measured as indicated
above by a differential detector to provide an indication of the
degree of deflection of the stream which in itself is an indication
of the momentum of the stream issuing from the nozzle 17. As the
flow from the orifice 17 is increased or decreased, the degree of
deflection of the main stream increases or decreases therewith,
changing the proportions of the stream entering the channels 18 and
19 as a function of the input signal. The main stream may have a
greater momentum than the momentum of the stream issuing from the
nozzle 16 or 17 or both and therefore the amount of fluid switched
to one or the other of the outlet passages is greater than the
amount of fluid issuing from the orifice 17. In effect then,
amplification of the signal applied to the nozzle 14 is effected
and a pure fluid amplifier is provided.
The same effect is accomplished if fluid is applied to the nozzle
13 except for the fact that the main power stream is deflected to
the right in response to application of the fluid to the nozzle 13
rather than to the left as when fluid is applied to the nozzle 14.
Fluid may be applied simultaneously to both of the control nozzles
13 and 14 in which case the stream issuing from the nozzle 3 is
deflected in accordance with the differences between the parameters
of the two control streams. Due to the amplification function of
the apparatus the output signal developed is an amplified function
of this difference.
Where a device such as that illustrated in FIG. 1 is to be employed
as an amplifier it must meet all of the rigid requirements of any
analogue amplification system. Specifically the output signal must
have a high signal-to-noise ratio; it must be a predetermined
function of the input signal; it must have stability in the absence
of an input signal or in the presence of a constant longterm signal
and all of the parameters must be reproducible during successive
periods of operation. Also as previously indicated it is essential
that a device once designed perform, when fabricated, as originally
intended.
Discussing for the moment this latter feature, in a specific
example of an amplifier, the orifices 4, 16 and 17 may have a width
equal to 20 mils; and the depths of the channels (orifices) may be
one-tenth of an inch. Such a system would be normally employed for
pressure gain, particularly in digital systems. If a wall converges
inwardly at 1.degree., at a depth of one-tenth of an inch the
channel reduced in width on one side by 1.75 mils; a total of
31/2mils for both walls. This degree of convergence is permissible
since under these circumstances the variation in nozzle width falls
within the normal design tolerances of the system with the top of
the nozzle being 20 mils wide and the bottom of the nozzle being
161/2mils wide. However, in the same system if the angle of
convergence of the sidewalls is 5.degree. then the bottom of the
channel is only 31/2mils wide. Such a construction falls completely
outside of the permissible tolerances in the system and the device
is unacceptable. On the other hand if the apparatus is intended to
be employed, for instance, as an energy or mass flow amplifier, the
widths of the various nozzle are increased to a greater extent that
the depth of the channels and the 5.degree. angle of convergence
provides an acceptable design parameter. It is apparent therefore
that in order to provide devices by mass production processes, this
angle of convergence must be maintained within very close
tolerances.
As to the signal-to-noise ratio of the amplifier, this factor is
basically a function of turbulence in the system.
In order to provide a low-noise signal, it is necessary to
establish substantially laminar flow throughout the device since
turbulence is a random effect having extensive regions of
unpredictable eddies along the boundary layers of the various flow
regions. These eddies are completely unpredictable as to continuity
of a particular eddy and formations of new ones or sizes of the
various eddies. It has been found that if care is not taken to
insure substantially smooth walls in a device such as illustrated
in FIG. 1, cascading of three stages produces a noise signal which
completely overrides the signal information. Thus it is absolutely
essential in such a system to obtain smoothness of walls to be able
to cascade these elements into useful systems.
Another difficulty encountered in such systems is the longterm
stability thereof. in certain regions of the apparatus if the wall
has large discontinuities as opposed to grainness of the walls,
large vortices may form in these regions which expand and contract
and result in longterm drifts in the no-signal or constant-signal
output flows from the apparatus. Such discontinuities have been
found to occur primarily where machining techniques are employed to
form a master whereas the turbulent flow producing eddies normally
result where acid etching techniques are employed.
Referring now specifically to FIG. 3 of the accompanying there is
illustrated a first step in the preparation of a fluid amplifier in
accordance with the process of the present invention. A light
polymerizable plastic 26 is previously mounted on a base member 27
which in this particular example illustrated is metallic. Under
these circumstances a layer 28, employed to bind the plastic layer
26 to the base layer 27 which is substantially opaque or
light-absorptive so as to prevent scattering of light from the
metal into the polymerizable material. A photographic transparency
30 is disposed on top of the exposed layer of plastic 26 and has
formed thereon areas which are opaque or light where ever the
material from the plastic layer 26 is to be removed. The
transparency 30 also has transparent areas where it is desired to
have the light pass into the layer 26 to produce polymerization at
which locations the material of the layer 26 will not be
subsequently removed.
The light for polymerizing the plastic is derived from a source
which provides parallel rays 29 by means to be described in greater
detail subsequently. The parallel rays produce substantially
vertical sidewalls of the various channels. After a predetermined
time for exposure, which is calculated in accordance with the
intensity of the light and the light energy required to produce
sufficient polymerization of the plastic 26, the light is turned
off and the negative is removed. Thereafter the laminated plate
comprising the layers 26, 27 and 28 is washed in a bath of a
suitable solvent for the unpolymerized material to produce the
member as illustrated in FIG. 4. These portions of the layer 26,
where light did not impinge thereupon, are removed down to the base
or the binder layer 28. Thus the layer of binder material 28 forms
the bottom of the channels whereas the polymerized material of the
layer 26 forms the sides of the channel.
A finished fluid amplifier element is then produced by employing a
further layer 31 of suitable material to seal the upper surface of
layer 26 permitting communication between the passages only a
result of interconnections thereof. The various connections to the
appropriate channels may be made by drilling holes through the base
plate 27 and layer 28, such as to provide an aperture 32 in
communication with, for instance, a channel 33 formed in the basic
element. Relating such a connection to the apparatus of FIG. 1 the
aperture 32 may constitute the supply passage to the nozzle 3 of
the fluid amplifier. The aperture 32 will normally be tapped and
provided with a fitting 34 adapted to accept a connection with a
member or a pipe or similar device which is returned to the supply
source.
The layer 31 may constitute a plastic, such as a celluloid layer,
which is bonded by a suitable bonder to the upper surface of the
plastic layer 26. Addition techniques for providing a sealing
relationship may be to employ a celluloid layer which is clamped by
means of screws extending between the metal plate 27 and a further
metal or plastic plate disposed over the layer 31. At times it is
convenient to employ both techniques wherein a binder is placed
between the layer 26 and the film 31 and then a further metal or
plastic plate is laid over the plastic plate 31 and clamped by
means of bolting with screws to the plate 27.
Referring now to FIG. 6 of the accompanying drawings there is
illustrated a light source which is suitable for use with the
apparatus of the present invention. The source utilizes a point
source and a folded optical system so as to reduce the overall
length of the apparatus. In this particular example a pair of
electrodes 36 are utilized to establish a high-intensity arc which
forms source of light and the light generated thereby passes
through a pin hole in an optical mask 37. The light from arc is
directed through the mask 37 to a mirror 38 which directs the beam
to a concave mirror 39. The mirror 39 in a particular example has a
focal length of 80 inches and is a 6-inch-diameter mirror. The
mirror 38 provides a virtual image 40 of the small hole in the mask
37 at a distance of 80 inches from the surface of the concave
mirror 39. The mirror 39 is set at such an angle relative to the
angle which the rays are directed thereto by the mirror 38 so as to
cause the beams to be directed along an axis 41, which is
horizontal in the illustration of FIG. 6. The divergence of the
beam passing through the hole in the mask 37 and reflected from the
mirror 38 is such that the beam assumes a diameter slightly smaller
than the diameter of the mirror 39 when it impinges thereupon.
Since the virtual image of the aperture in the mask 37 is at the
focal point of the mirror 39, the rays reflected therefrom
constitute a substantially parallel bundle of rays which is
suitable for use as the bundle of rays 29 illustrated in FIG.
3.
It should be noted that by moving the slit 37 relative to the
mirror 38 the virtual image 40 of the aperture in the mask 37 can
be shifted relative to the focal point of the mirror 39 thereby
providing an accurate control on the degree of divergence or
convergence of the beam reflected from the mirror 39. Thus, if
large elements are being fabricated and a 5.degree. divergence
angle is permissible, the slit 37 may be shifted to provide such a
degree of divergence. On the other hand, if small units are being
fabricated and the degree of divergence should be no greater than
1.degree. then the slit 37 may again be shifted so as to shift the
virtual image 40 relative to the focal length of the mirror 39 to
provide the desired degree of divergence. It is to be understood
that the light source constituting an arc source in FIG. 6 may be
replaced with a laser or other suitable high intensity source, the
arc source being illustrated merely for purposes of simplicity of
explanation. In fact, it is preferred to employ a laser in the
apparatus in view of the high-intensity light which may be obtained
from such a source.
In the process described in FIGS. 3 through 5 it is assumed that
the element in FIG. 4 is to constitute the fluid amplifier and
therefore the transparency 30 of FIG. 3 constitutes a photographic
negative of the apparatus to be fabricated. If it is desired to
employ the device produced in FIG. 4 as a master for making many
fluid elements of the same configuration; that is, to employ the
element of FIG. 4 as a mold, then the photographic transparency 30
of FIG. 3 would constitute a positive so that the element produced,
as illustrated in FIG. 4, may be employed as a mold of the devices
to be eventually produced.
As previously indicated, it may be desired to produce fluid
resistors by the techniques of the present invention. In such case,
a cross section of the fluid resistor produced by a dot screen
process would be as illustrated in FIG. 7. Where it is desired to
form the resistor directly in a passage of an element, such as that
illustrated in FIG. 4, the area of the photographic transparency in
the region in which the resistor is to be formed is rendered opaque
except for a plurality of small transparent circles having the
proper diameters and spacing to produce a resistor with the desired
characteristics. The resistor is made up of a plurality of
elongated solid cylinders or stalks 42 of the material of the layer
26, the sidewalls of the cylinders being generally parallel to the
sidewalls of the passage in which the resistor is formed. The
distance between adjacent sides of the stalks 42 is determined by
the fluid resistance and is normally relatively small; in many
instances being of the same order of magnitude as the width of the
orifice of the power or control nozzle.
It can be seen that in an arrangement such as illustrated in FIG.
7, if the angle of divergence of the light beam employed to
polymerize the plastic is not carefully controlled, complete
penetration of the light to the layer 28 cannot be effected.
Referring specifically to FIG. 8 which illustrates the process for
making a resistor, if the angle of divergence of the beam is too
great relative to the depth of the device, which divergence angle
is taken to be about 5.degree. in FIG. 8, then the light passing
through adjacent transparent dots in the negative almost come
together at the binder layer 28 of the blank. The effect of the use
of a light source is illustrated in FIG. 9 where it becomes
apparent that the width of the channels provided between the base
of the stalks 42 are considerably less than they should be and in
consequence the fluid resistance is of far greater magnitude than
that for which it was designed. Thus, it becomes apparent,
particularly when forming fluid resistors and nozzles in small
elements, that the optical system provide closely controllable beam
divergence angles. It should be pointed out that where the blank
formed as illustrated in FIG. 4 is to be employed as a mold that
some convergence of the sidewalls is required in order to expedite
withdrawal of the molded article from the mold. Due to the size of
the passages between the stalks 42, difficulty is experienced in
producing a sufficient flow of solvent particularly at the base of
the stalks to remove the unpolymerized material from
therebetween.
In order to overcome the difficulty in certain instances where the
passages between the stalks 42 are quite small, it is proposed in
accordance with the invention to immerse the plate, during the
washing operation, in a bath having an ultrasonic vibrator disposed
therein. An ultrasonic vibrator sets up vibrations of sufficient
amplitude and intensity to wash away layer after layer of the
unpolymerized material until all material has been removed down to
the binder layer 28.
It has been assumed previously that the entire surface, as
illustrated in FIG. 3, is to be irradiated at one time. It is
possible however to employ light beam sweeping techniques so as to
utilize a higher intensity beam which produces less thermal
effects. Explaining this statement, it is obvious that if the
energy in a source of light is spread throughout a beam having a
diameter of 6 inches, the energy at any point across that beam must
be considerably less than a beam having the same energy
concentrated and a diameter of one-fourth inch since in the latter
case the entire energy in the beam is concentrated in a very small
area. Therefore, it is necessary when employing the large diameter
beam to utilize irradiation times considerably greater than if the
beam were highly concentrated. Since the heating effect in the
blank is a function of time as well as energy of the beam, the
heating effect is greater under a lower intensity beam applied for
a long period of time then if a very high energy beam is employed
in a pulse type of operation. Therefore, in certain instances, it
may be desired to use the light for instance from a ruby laser or
from a quartz crystal energized from a ruby laser so as to produce
ultraviolet light, and scan the surface of the structure, as
illustrated in FIG. 3, relatively rapidly. The heating effects
under such circumstances are quite low and the thermal effects
become inconsequential.
It is apparent from the above discussion that the fluid amplifiers
of the type illustrated in FIGS. 1 and 2 may be readily prepared by
a process employing a lightly polymerizable material. As previously
indicated, with such a process, the slope of the sidewalls of the
apparatus may be carefully controlled. Further the process produces
very smooth sidewalls. In a particular process, a dilute solution
of sodium hydroxide is employed as the wash bath and this material
does not appreciably effect the polymerized element since it does
not constitute a solvent for such material but only for the
unpolymerized material. In consequence one does not obtain the
graininess of the sidewalls that may result from a photoetch
process where the etching is effected by a highly reactive
acid.
* * * * *