U.S. patent number 3,668,756 [Application Number 04/817,431] was granted by the patent office on 1972-06-13 for method for making fluid channels.
This patent grant is currently assigned to M. V. Bekaert S. A.. Invention is credited to Andre A. Wieme.
United States Patent |
3,668,756 |
Wieme |
June 13, 1972 |
METHOD FOR MAKING FLUID CHANNELS
Abstract
A method of making flow channels in fluid control devices
comprising coining into the surface of the device a length of hard
drawn wire having an appropriate form, and removing the length of
wire from the channel thus formed.
Inventors: |
Wieme; Andre A. (Zwevegem,
BE) |
Assignee: |
M. V. Bekaert S. A. (Zwevegem,
BE)
|
Family
ID: |
8649265 |
Appl.
No.: |
04/817,431 |
Filed: |
April 18, 1969 |
Foreign Application Priority Data
Current U.S.
Class: |
29/890.09;
29/424; 29/432; 29/604; 29/846 |
Current CPC
Class: |
F15C
5/00 (20130101); Y10T 29/49833 (20150115); Y10T
29/49069 (20150115); Y10T 29/49155 (20150115); Y10T
29/49812 (20150115); Y10T 29/494 (20150115) |
Current International
Class: |
F15C
5/00 (20060101); B21d 053/00 (); B21k 029/00 ();
B23p 015/26 () |
Field of
Search: |
;29/157,424,604,625,627,432 |
References Cited
[Referenced By]
U.S. Patent Documents
Primary Examiner: Campbell; John F.
Assistant Examiner: Rooney; Donald P.
Claims
What I claim is:
1. A method for making a fluid control device having a tubular
fluid passageway therein comprising:
a. providing a plate member,
b. coining into the surface of the plate member a length of hard
drawn wire having an appropriate form,
c. removing the length of wire so as to leave a channel therein
generally conforming to the form of the wire,
d. applying to the plate member a cover member of sufficient
dimension to cover the channel,
e. thereby forming a tubular fluid passageway in the fluid control
device, and including the step of
f. providing means in said device for providing fluid access to
said passageway from an external source.
2. A method for making a fluid control device having a tubular
fluid passageway therein comprising:
a. providing a plate member,
b. bending a length of hard drawn wire in the form of the
passageway,
c. positioning said length of hard drawn wire on the surface of the
plate member in coincidence with the place where the passageway is
to be formed,
d. coining said length of hard drawn wire into said surface,
e. removing said length of hard drawn wire,
f. applying to the plate member a cover member of sufficient
dimension to cover the passageway,
g. thereby forming a tubular fluid passageway in the fluid control
device, and including the step of
h. providing means in the device for providing fluid access to said
passageway from an external source.
3. A method for making a fluid control device having a tubular
fluid passageway therein comprising:
a. providing a plate member,
b. providing a press matrix,
c. coining a length of hard drawn wire having an appropriate form
partially into the surface of the press matrix so as to provide a
projection on the surface of the press matrix in the form of the
passageway,
d. pressing together the surface of the press matrix and the
surface of the plate member,
e. removing the press matrix and the wire so as to leave a channel
in the surface of the plate member generally conforming to the form
of the wire,
f. applying to the plate member a cover member of sufficient
dimension to cover the channel,
g. thereby forming a tubular fluid passageway in the fluid control
device, and
h. providing means in the device for providing fluid access to the
passageway from an external source.
4. A method for making a fluid control device having a tubular
fluid passageway therein comprising:
a. providing a pair of plate members,
b. positioning a length of hard drawn wire of an appropriate form
between the two plate members,
c. pressing the pair of plate members together in a registering
position so as to coin the wire into the surface of at least one of
the members,
d. separating the plate members,
e. removing the wire so as to leave a channel in the surface of at
least one of the plate members,
f. replacing the plate members together in the registering
position,
g. thereby forming a tubular fluid passageway in the fluid control
device, and
h. providing means in the device for providing fluid access to the
passageway from an external source.
5. A method as in claim 1 and wherein:
a. the plate member is aluminum, and
b. the hard drawn wire is a steel wire.
6. A method as in claim 2 and wherein:
a. the plate member is aluminum, and
b. the hard drawn wire is a steel wire.
7. A method as in claim 3 and wherein:
a. the plate member is aluminum, and
b. the hard drawn wire is a steel wire.
Description
DESCRIPTION OF THE INVENTION
The present invention relates to fluid control and logic devices,
and has particular relation to a method for making a unit of the
sandwich-type structure comprising a channel configuration at the
contact surface between two bodies. More particularly, the present
invention resides in the fact that the flow channels are obtained
by coining in the contact surface of at least one of the bodies a
wire which has been prepared by the wire-drawing process and which
has been bent in appropriate form.
Fluid control and logic systems are more and more used, instead of
electronic systems, for the control and command of fluid operated
machines, in order to avoid the need of signal transformers, such
as electric or electromagnetic relay control valves, which must
translate the electronic output command signal to a fluid pressure
signal for the operation of the machine. The complete machine,
control and logic system included, can then be supplied with fluid
pressure power only and does not need any additional electrical
power supply. Moreover, fluid elements have an almost infinite
lifetime and do not break down when wrongly interconnected.
Logic elements perform the elementary logic functions
(AND-function, OR-function, bistable memory function) in an
interaction chamber according to, for example, either the principle
of the exchange of kinetic energy between two or more fluid jets or
the principle of the boundary layer effect. Amplification can be
achieved for example, by a laminar high energy flow which can be
made turbulent by a low energy impinging jet in an interaction
chamber. Owing to the fact that the fluid interaction chambers
perform the same elementary logic functions as the electronic logic
components, the circuit diagrams of fluid logic and control systems
are analogous to the electrical circuit diagrams, and not only
fluid interaction chambers were to be developed, but it was also
necessary to provide for the interconnection between the different
interaction chambers by means of a number of channels through which
the fluid can pass from one chamber to the other.
It is known, as disclosed, for example, in U.S. Pat. Nos. 3,207,168
3,016,066 and 3,024,805, that fluid channel configuration,
including interaction chambers and interconnection channels can be
formed by a plurality of flat plates, either two or three plates,
the plates being sandwiched together and sealed fluid-tight one to
the other by adhesives, machine screws, clamps or other suitable
means. In the case of the two-plate sandwich structure, one plate
is molded, etched or cut in order to contain the channel
configuration on its surface, and this plate is covered by another
flat plate, so that the flow in the unit is confined by the plates,
and that the channel configuration is realized at the contact
surface between the two plates. In this way, it is possible to
realize at least a part of a circuit in a bidimentional
arrangement.
In electrical systems, the interconnections between the electronic
logic components are made by metallic conductors, either wired or
printed, of practical zero resistance, and eventually in series
with resistances of which the resistance value can be held between
very narrow limits if necessary. In fluid logic systems however of
the sandwich-type structure, the interconnection channels present
more than one problem. Firstly, an interconnection channel always
acts partly like a fluid integrator, as a result of the volume of
the channel itself. Consequently, high frequency signals are
strongly attenuated, sharp impulses are not transmitted, and the
possible operating speed of the system is reduced. Secondly, an
interconnection channel has always a flow resistance which cannot
be neglected by the designer of the circuit. It is mostly desired
to reduce that resistance to a minimum. Thirdly, even when the
presence of a resistance is accepted or needed in the design, it is
absolutely necessary that the predicted channel resistance could be
realized without difficulties and within narrow value limits, and
in a repetitive manner adaptable for mass production. And forthly,
the "wiring" of flow channels can be realized as grooves in the
surface of a flat plate, but the present methods for cutting,
molding or etching the grooves are not very flexible. Those methods
are only more economical with respect to the wiring with rubber
tubes between separate interaction elements, when great quantities
of the same circuit must be made.
It is an object of the present invention to provide a method for
making an interconnection fluid channel configuration at the
contact surface between two plates of a sandwich-type structure,
which method can give a solution to one or more of the above
mentioned problems.
Other objects, features and advantages will become apparent upon
consideration of the following description and the accompanying
drawings in which:
FIG. 1 is a perspective view of the contact surface of a plate of a
fluid device having a plurality of fluid channels,
FIG. 2 is a view of the same surface comprising a number of hard
drawn wires ready to be coined in the surface,
FIG. 3 is a press matrix obtained by coining hard drawn wires for a
part of their thickness in the surface of the matrix.
Referring to the drawings, there is illustrated in FIG. 1 an
embodiment of the invention, given by way of example. The device
comprises a first plate 1 with a configuration of channels, such as
channel 2, formed in its surface. A second flat plate (not shown)
is laid upon the first plate and clamped, sealed or otherwise
fastened by screws, clamps or adhesives to this plate. The
connection between the plates should be made fluid-tight, so that
the fluid flow in the resulting unit is confined by the plates, and
that the fluid is only enabled to flow through the defined
openings, passages and cavities between the two plates. The
channels 2 serve to interconnect the different fluid interaction
chambers of a pure fluid logic system, such as chamber 3 or 4. It
is clear however that the channels can serve for interconnection of
fluid logic devices with moving parts, analog elements instead of
digital devices, and other fluid control systems in which
interconnection is needed between different function devices. The
channels serve partly also to connect the functional devices, such
as the pure fluid logic interaction chambers of this example, to
the input and output orifices 5.
The interaction chambers may have been formed in the plate 1
itself, such like chamber 3, but may also be a part of a separate
fluid flow device, such like chamber 4, which is connected to the
channel configuration in any suitable manner by which the fluid
flow is enabled to pass from the separate device to the channel
configuration of the plate. In this example, the separate device is
formed by a plate 6, which comprises an interaction chamber 4 and
the necessary input, output and exhaust channels. This plate 6 is
fixed on the back sides of plate 1 in the same way as the cover
plate (not shown) of plate 1, in order to define a fluid-tight
configuration of passages and cavities which is connected to the
configuration on the top side of plate 1 by means of holes 8 in
this plate.
The configuration between plate 1 and its cover plate may only be a
part of a complete system, where several analogous plates are, for
example, stacked on top of each other. The plate 1 may only serve
to interconnect several functional standard plates to each other,
and consequently contain the interconnection "wiring," but, as will
be explained later, can also contain the necessary resistances,
capacitances and even passive or active elements. The orifices to
the other devices of the system can be made in the plane of the
configuration, such as orifice 5, or perpendicular to that plane,
such as orifice 8.
The fluid interaction chambers 3 may have been made in the plate 1
by any suitable method, such like cutting, etching or molding,
before, after or during the operation by which the channels are
formed, the invention being confined as to the method by which the
channels are obtained. Those channels are obtained by coining
pieces of hard drawn wire into the surface of the plate.
When pressing a body of sufficient hardness into a plate of soft
material, the body penetrates into the plate, pushing away the soft
material by plastic deformation. Heating is not necessary, but the
material of the plate must be much softer than the material of the
wire. Otherwise the wire is deformed during the coining operation,
the channel cannot be made in the desired shape, and the wire can
no longer be used for a subsequent operation. The hardness of the
wire in relation to the hardness of the plate depends on the
desired degree of accuracy. By a hard steel wire is consequently
meant a wire the hardness of which is sufficiently high in relation
to the hardness of the plate in order to reach the desired degree
of accuracy. Aluminum can be used successfully in connection with
hard drawn steel wire ends. Aluminum has a Brinell BHN hardness of
50 to 100 which corresponds to a tensile force of maximum 45
kilograms per square millimeter, and a steel wire of carbon content
of, for example, 0.85 percent, drawn to 20 percent of its original
section gives excellent results. But it is clear that other
combinations of plate material and drawn wire material can easily
be found. As plate material may also be used soft iron, tin,
copper, stainless steel or plastic material, such like polyethylene
- polyvinyl chloride - polypropylene - ABS. Aluminum however has
the advantage to be a light-weight, not easily deformable material
and soft enough to be used with steel wires. Drawn wires of any
material, hardness and diameter are available on the market in
great quantities. Especially steel wires can be found in any
combination of diameter and hardness from 0.05 mm on with
tolerances of 0.005 mm, and with a tensile force of 300 kilograms
per square millimeter being also a value of evaluation of the
hardness, which can be obtained by a judicious choice of carbon
content, percentage of reduction of the cross-sectional area during
the drawing operation and eventually by a subsequent process of oil
hardening. These processes of wire drawing are sufficiently known
in the literature on wire drawing, such as the book Stahdraht of A.
Pomp, and the products are widely on the market in any combination
of diameter and hardness.
The resulting coined channels will be very smooth owing to the fact
that the body impressed in the plate surface is a drawn wire, which
is itself very smooth owing to the drawing process in a wire die.
The smoothness can be increased when the reduction of the wire
cross-sectional area per drawing die is diminished, and also when
at least the last reduction steps are obtained by a wet drawing
process as explained for example in Trefilage de l'Acier of M.
Bonzel. Such smooth shining steel wires are also obtainable in the
market in any combination of diameter and hardness. When aluminum
plates are used with hard steel wire, it has also been observed
that the grain structure is refined in the regions undergoing the
plastic deformation during the coining process. Consequently, very
smooth channel surfaces are obtained.
Smooth channel surfaces are advantageous because of the resulting
small flow resistance. For the same desired or maximum allowed
resistance value it is possible to design channels of smaller
sections, so that the channel volume is diminished and consequently
the working speed of the system can be increased. Channels with a
high l/d ratio (l= length and d = diameter) have also a higher
pressure limit under which limit a laminar flow can be secured.
Laminar flow in the channels is desirable because it preserves the
signal strength and decreases internal heating. As a result, the
devices with narrow-smooth channels can work at higher signal
pressures without entering in turbulence.
In order to make very narrow channels by coining, it is not only
necessary to have smooth hard cylindrical bodies to be coined into
a soft plate, but those bodies must be made within very narrow
tolerances in order to be able to make the devices in a repetitive
manner and to meet the specifications of the devices. For a
relative tolerance value of for example 10 percent, the region of
acceptable diameter values decreases in proportion to the diameter,
and for very narrow channels, the possibility of shaping
cylindrical bodies within narrow tolerances becomes critical. As
has been pointed out hereinabove, smooth hard drawn wires are
available in the market in any combination of diameter and hardness
and with tolerances up to 0.005 mm.
There are two kinds of flow resistances. Firstly, there are
channels in which the flow resistance is realized by narrowings and
sharp turnings. These resistance-devices are not very linear, that
is to say the pressure difference output and input is not a linear
function of the fluid flow, but rather a quadratic function. There
are also channels in which the flow resistance is realized by
porous plugs. With very narrow smooth channels, it is, however,
possible to realize in any easy way, calibrated flow resistances
within narrow tolerance limits and which are acceptably linear. The
flow can be held laminar for high pressure values and the
resistance is realized by the flow resistance of the walls. An
example of the obtained results will be given hereunder.
Resistance-elements are sometimes needed in a flow circuit for
purposes of feedback, time delaying with a resistance-capacitance
element, such as resistance 9 and capacitance 10 or resistance 11
and capacitance 12 in the example of FIG. 1. In this example three
forms of execution of narrow channel resistances are shown.
Resistance 13 is a simple straight channel resistance. Resistance 9
is made by two parallel straight channels. The l/d-ratio of each
channel is greater than the ratio of a single channel with the same
total resistance value. In that way higher pressures can be used
with less possibility of turbulence. Consequently, for low
resistance values it can be preferable to use several parallel
channels, which may not necessarily have the same cross-section.
One or more narrow channels can be used for adjustment. Large
resistance values can be obtained with a channel such like 11.
Owing to the large l/d-ratio, the laminar flow is not disturbed in
the bendings of the channel and cannot turn into turbulent flow. As
a result, a large laminar channel of high resistance value can be
formed in a small area of the surface of the plate.
Plate 1 also comprises, as a way of example, two passive
AND-circuits, of which one has been indicated by the numeral 14.
This circuits comprise two inlet channels 15 and 16 for the input
signal flow jets and two output channels 17 and 18, which are in
the prolongation of channels 15 and 16 respectively and which lead
to the exhaust orifices 22 and 23. At the intersection of input and
output channels, there is a flow jet interaction chamber 3. An
output-signal channel 20 starts from this chamber to another
functional element. In operation, when only one of the input
channels 15 or 16 deliver a flow jet, this jet flows directly from
the input channel in the opposed output channel to the
corresponding exhaust-orifice and no output signal appears in
channel 20. When however both input channels deliver a flow jet,
those flow jets inpinge on each other and deviate the other jet in
this way so that only one resulting jet is formed which is directed
to output channel 20. It is clear that it is necessary that input
and opposed output channel be accurately in the prolongation of
each other. This is possible when one single straight wire is used
for coining the input and output channel together.
FIG. 2 illustrates a way for coining the wire into the surface of
plate 1. The same plate has been shown, without cover plate and
without the active elements fixed on the back, ready to undergo the
last operation of coining wires 24, 25, 26 and 27 into the surface.
The other orifices, cavities and channels are supposed to be made
in a preceding operation. Four pieces of hard wire are cut at
length, properly bent and laid on the surface in coincidence with
the place where the channel is to be coined. Subsequently, a plate
of hard steel 28, shown in dotted lines in FIG. 2, is pressed
against plate 1, in the direction of the arrow, and the wires enter
into the material of plate 1. In order to prevent any movement of
the wire before it penetrates in the plate, it will be sometimes
necessary to fix the wires to the surface, for example, by gluing,
or as shown with wire 24, by bending both ends of the wire into the
orifices 29 and 30 in which the corresponding channel will debouch.
This gives a smooth transition between channel and orifice. It has
been shown especially that the input and output channels 15 and 17
of the passive AND-element 14 of FIG. 1 are obtained by coining of
one single straight wire 26. As for the hardness of plate 28, there
is no problem when it is at least as hard as the wire, although
this is not necessary. It must be hard enough in proportion to the
hardness of the plate 1, in order that the wires be coined
completely in plate 1, and only for a negligible part in plate
28.
The illustrated method of laying the wires separately on the
surface and coining the unit, is not the only one which can give
satisfying results. The wires may be firstly bent in proper form
and laid upon a hard flat steel surface in the same configuration
of the channels. Subsequently a plate of soft iron is pressed upon
the wires, which penetrate into this soft iron plate. But the plate
is not pressed until it comes in contact with the hard steel
surface. The wires are only partly impressed in the soft iron.
After that, the wires are glued in the channels which have been
formed in the soft iron plate, and a press matrix is obtained in
that way, as shown on FIG. 3. The plate has then on its surface a
number of projecting ribs 31. Such a press matrix plate can then be
turned upside down and pressed upon for example an aluminum plate
or a plate of plastic material. Wires of different thickness will
have to be impressed in the soft iron plate at different
moments.
Returning to FIG. 2, when the plate 28 is not of hard material, but
of the same material as plate 1, then the wire is partly impressed
in plate 1 and partly in plate 28, when both plates are pressed
together. The wire is removed afterwards and the plates brought
together in the same relative position as they are when pressed
together. Then a canalisation is formed of exactly the shape of the
wires. Reference holes can be drilled when both plates are pressed
with the wires between them. At that moment, the contact surfaces
of both plates adapt to each other by the influence of the
pressure, and flow leakage through the area between the plates can
be reduced to a minimum. In this way, two plates of aluminum can be
pressed together with hard steel wires between them, and channels
of very accurate dimensions can be formed. This method is
particularly interesting for making resistance devices. A
resistance has been made in this way, by pressing two flat aluminum
plates together, with an oil-tempered steel wire (0.65 percent
carbon, diameter 0.5 mm, length 42 mm) between them. The resistance
value varied from 22.7 gr-sec/cm.sup.5 .+-. 5% to 27.80
gr-sec/cm.sup.5 .+-. 5% for a pressure difference between input and
output varying from 600 gr/cm.sup.2 to 1,000 gr/cm.sup.2.
When using the first method (pressing with a hard plate which is
not the cover plate) the coining may give irregularities on the
surface of the soft plate, in the vicinity of the coined channels
and the material tends to be pushed upwards. These irregularities
can be lapped away in order to secure fluid tightness. As the
lapping operation can be controlled very accurately, and the
lapping does not affect the internal surface of the channel, this
operation does not present any problem.
Using drawn wires as press matrices is an extremely simple, cheap
and flexible method for making wiring boards, analogous to printed
circuit boards in electronics. Standard functional elements can be
made in great number, but generally wiring boards are more adapted
to specific applications and rarely must be made in great
quantities which could justify to develop etching masks, moulds
etc. No special tools are needed for using this coining method. The
wires may be bent by hand. For this reason, this method is very
well adapted for small quantity wiring boards.
It will be clear that the drawn wire must not have necessarily a
circular section. Hard steel wires with square sections for example
can also be found in the market. The method is not only adapted for
coining into flat surfaces or plates. Press matrices can be made in
cylindrical form, with coined wires therein, to be used as rotating
press matrices.
While the invention has been described in its preferred
embodiments, it is understood that those embodiments and the words
used have only been given as an example, rather than as a
limitation, and that changes within the purview of the appended
claims may be made without departing from the true scope and spirit
of the invention.
* * * * *