Technical field
[0001] The subject of this invention is a device generating periodic unsteady pressure and/or
flow processes in the fluid that passes through it. It can find application in all
those areas of technology that work with flows of fluids (in particular gases), for
example in aeronautics and building turbomachines - such as steam, gas, or wind turbines.
Background art
[0002] Fluidic oscillators generating periodic unsteady pressure and/or flow processes in
the fluid passing through it. Typically, the fluidic oscillator is formed by a system
of mutually interconnected cavities in which the overall hydrodynamic instability
gives rise to self-excited oscillation. These oscillator have been nowadays developed
for a number of uses, of which it is particularly proper to mention the control of
fluid flow past bodies. The generated oscillating fluid flow from the oscillator output
terminal acts on the external flow past the body and influences such phenomena as
the flow separation from the surface of the body. In this application, the flow from
the oscillator cavity, usually located inside the body, is lead into an actuator nozzle
positioned by its exit in such locations on the body surface in which the boundary
layer formed on the body is particularly sensitive to acting disturbances. As a result,
it is possible by relatively small input power to influence the boundary layer so
as to suppress its tendency towards separation from the body - the tendency that is
often found when the angle of attack of the body relative to the direction of the
outer flow reaches a certain limit value. Of course, the fluid passing through the
oscillator may be, with advantage, the same fluid as the one that flows past the body.
On the other hand, it is also possible, using the fluidic oscillator, to produce a
premature separation at small attack angles. It is therefore possible the remove the
operating limits that are otherwise caused by the behaviour of boundary layer. For
example, a wind turbine blade may be kept operating in a regime where otherwise the
boundary layer separation would stall it. Thus the flow control with the oscillatory
control flow leads to a wider freedom for design and operation of fluid handling machinery
- in the above example the wind turbine may be used in a wider range of wind velocities.
An important fact that has been demonstrated by a number of investigators is the pulsating
jet issuing from the actuator nozzle acts on the boundary layer more effectively than
if it were a steady, non-oscillatory jet.
[0003] Known versions of fluidic oscillator applicable to the above mentioned purposes contain
a jet-type bistable fluidic amplifier and at least one channel, called feedback channel,
that is connected to the control nozzle(s) of the amplifier and carry into it a feedback
signal. The bistable amplifiers are usually based on applying the Coanda effect of
fluid jet attachment to a solid wall. The knowledge about these amplifiers is not
widespread, nevertheless there is sufficient information about them - their working
principles, layouts, and functioning - in available literature. For example, they
are the subject of Section 4.4. named "
Switching valves based on the Coanda effect" in the monograph " Pressure-Driven Microfluidics",
published by ArtechHouse, Inc., Norwood, MA, U.S.A.
[0004] Current versions of bistable jet-type fluidic amplifiers are almost invariably of
planar layout. This means the cavities forming the oscillator are made in a planar
plate by removal of material - for example by a photochemical method (etching according
to a photographically transferred mask) - into the same depth. The cavities made this
way are then closed by being covered with planar cover-plates.
[0005] The fluid in which the oscillation is generated is brought into the supply nozzle
of the amplifier. From the exit of the supply nozzle it issues as a jet. To obtain
the bistability, on both sides of the said supply nozzle are positioned attachment
walls, symmetrically facing one another. The jet has therefore an equal opportunity
to attach by means of the Coanda effect of jet attachment to either one of them. The
attachment walls are inclined so that the attachment causes a change the jet flow
direction. One attachment wall guides the attached jet to the first output terminal
- while the other attachment wall would lead it to the second output terminal. In
the use of the oscillator for the above mentioned control of an outer fluid flow,
both output terminals are connected to actuator nozzles. Of course in its other uses
the oscillator co-operates with other connected devices instead of the attachment
nozzles. The flow inside the amplifier is controlled by control nozzles. These are
positioned so that they oppose each other, with their exits positioned in the locations
between the attachment walls and the exit of the supply nozzle. By a flow from the
suitable control nozzle it is thus possible to switch the jet from one attachment
wall to the opposite one and thus, as a result, from one output terminal into the
other one. Fluidic oscillators known so far use for generating the oscillation a feedback
connected to the control nozzles. Two basic types of the feedback are used, both invented
by R. W. Warren in the U.S.A. The classical version, known from the
US Patent 3,158,166, "Negative feedback oscillator", filed 7
th Aug. 1962, is characterised by two feedback loops. Each of them guides, separately
on the opposite sides of the amplifier, the feedback flow from the output terminal
of the amplifier to the same-side control nozzle. A simpler version of the feedback
is characterised by a single feedback channel connecting the two control nozzles.
This layout was first used by
E.C. Spyropoulos and published a conference contribution on this subject on page 27
in Proceedings of the Fluid Amplification Symposium, organised in 1964 by Harry Diamond
Laboratories in Washington D.C. A related oscillator with a single feedback channel is the subject of the
US Patent Nr. 4 231 519 Bauer P.: "Fluidic Oscillator with Resonant Inertance and Dynamic Compliance Circuit", filed
in 1980.
[0006] An important property of both versions of the so far known oscillators is the frequency
of generated oscillation being practically directly proportional to the intensity
(i.e. flow rate) of the fluid through the oscillator. In many applications, in particular
in the above mentioned use to the control of boundary layer separation, there is a
demand for high oscillation frequency. This requirement leads not only to high velocity
flows inside the actual oscillator, but also to high flow velocity in the connected
loads - such as the above mentioned actuator nozzles. This mutual dependence of the
velocities is due to the oscillator and its load being mutually bound by matching
conditions. A detailed description of these conditions is available in the paper "
Fluidic control of reactor flow - pressure drop matching" published in Chemical Engineering
Research and Design, Vol. 87 , p. 817, 2009. This paper actually solves the matching problem for the particular situation in
which the load connected to the output terminal of a fluidic amplifier is a chemical
reactor. Nevertheless the character of the load is unimportant from the point of view
of the matching and conditions analogous to those described there apply to any load,
including the actuator nozzles.
[0007] The high flow velocities in the actuator nozzles, resultant from the requirement
of high frequency, however, cause troubles. One of the obvious disadvantages are the
high dissipative losses - because the magnitude of energetic losses in flowing fluid
rapidly increases with the flow velocity, very roughly with the second power of the
velocity magnitude. Another problem in many situations is the non-availability of
a suitable high pressure fluid source, the high pressure being necessary for generating
the high velocity. Finally, a very important disadvantage of high-velocity flows in
actuator nozzles is the less effective action on the controlled boundary layer - the
high-velocity jet is said to "pierce" through the boundary layer instead of the desirable
pushing action on it.
Disclosure of the invention
[0008] The disadvantages described above are removed by the device according to the present
invention. It is a fluidic oscillator with a bistable jet-type fluidic amplifier having
cavities that include including the supply nozzle, first control nozzle, second control
nozzle, a cavity with the first attachment wall and the second attachment wall, the
first output exit and second output exit, where on one side of the supply nozzle is
located the first control nozzle while on the opposite side of the supply nozzle there
is a second control nozzle and the first control nozzle as well as the second control
nozzle are located with their mouths against each other, and downstream from the first
control nozzle is positioned the second attachment wall and downstream from this there
is the second output exit, while downstream from the second control nozzle is positioned
the first attachment wall downstream from which there is the first output exit and
the first attachment wall is facing the second attachment wall, characterised by presence
of the resonance channel which is connected to the first control nozzle while its
other, free end is open into space and the second control nozzle open into the same
or a different space.
[0009] The space mentioned above is the atmosphere or a closed cavity, with an advantage
the inner space of a pressure tank. The dimensions of this space are assumed to be
large - so large that the outlets from the fluidic oscillator do not mutually influence
one another. In practice this means that these dimensions ale larger many times (for
example at least twice) than the linear dimensions of the fluidic amplifier. Of course,
if the free end of the resonance channel and the second control nozzle are open into
different spaces, then the size of the spaces is not limited by these conditions.
[0010] Between the entrances into the first outlet terminal and the second outlet terminal
may be positioned variously shaped splitter, usually of the wedge shape.
[0011] The resonance channel is with advantage positioned in the same plate as the cavities
of the jet-type bistable amplifier and is made by removal of plate material into the
same depth as the cavities of the jet-type bistable amplifier. In the next manufacturing
operation after this removal, the cavities are separated from the atmosphere by cover
plate placed on the plate. The cavities may be also made in a thin flat plate by removal
of the plate material along the line that represent the outer circumference of the
cavities. The advantage of this possibility is the necessity to remove a smaller total
amount of the plate material so that the manufacture may be faster and/or less expensive.
In this case, to cover the cavities and separate them from the atmosphere there are
to be two cover plates, one positioned above and the other below the plate in which
the cavities were made. The cover plate may contain a separate part covering the bistable
amplifier and another part covering the resonance channel, or all the cavities may
be covered by a single common cover plate. There are also known methods of manufacturing
the cavities in which the produced oscillator body is made in one piece (it may be,
e.g. the method of the "lost wax") in which case no cover plates are needed.
[0012] The fluidic oscillator may also involve an inlet for another (second) gas. With an
advantage, such a second gas inlet may be made up of at least one inclined nozzle
open by its exit into the resonance channel. The inclined nozzle serves for introduction
of the second gas so that it fills the resonance channel - up to such an extend that
the second gas flow out from the free end of the resonance channel. In that case the
frequency of generated oscillation is determined by the properties of the second gas.
It is obvious the if this change of the properties due to the presence of the second
gas is to have the largest possible effect, the exit of the inclined nozzle into the
resonance channel must be positioned as near as possible to the first control nozzle
of the bistable jet-type amplifier.
[0013] It may be useful for the oscillator to contain also a sensor making possible sensing
and/or measurement of the frequency of the generated oscillation. This is particularly
advantageous is the case of the oscillator having the second gas inlet, because the
frequency of the generated oscillation depends on the character and state of the second
gas and the oscillator according to this invention may then have the useful function
of a meter for measurement of either the composition of binary gas mixture or a detector
of concentration changes. Also, either with the second gas inlet or without it, the
oscillator may assume the role of a temperature meter. The sensor may be placed anywhere
next to the cavities of the fluidic oscillator in positions where the oscillation
causes periodic changes of the flow. Nevertheless, positioning at or near to the output
terminal is particularly advantageous because the changes in the course of each oscillation
period are there the highest and this makes easier meeting or overcoming the sensitivity
limits of the sensor.
[0014] The inlet of the second gas may be, however, connected with the supply nozzle - as
long as there is a supply of another cold gas into the oscillator cavities somewhere
between the sensor and an upstream from it located cold gas inlet. This possibility
is advantageous in those cases where the oscillator is to serve as a temperature meter
- utilising its generated frequency dependence on temperature - with a sensor that
cannot operate at a too high temperature, in particular the temperature of hot gas
supplied into the supply nozzle. By properly positioned inlet of the cold gas the
sensor temperature is decreased to an acceptable level. In an alternative version,
the oscillator may contain two bistable jet-type amplifier connected so that the upstream
bistable jet-type amplifier is used in generation of the oscillation while the downstream
amplifier serves for amplifying the oscillation signal. The sensor is then positioned
in or near the cavities of the downstream bistable jet-type amplifier. In the use
for the temperature measurements, the hot gas, the temperature of which is measured,
is fed into the supply nozzle of the upstream amplifier while the downstream amplifies,
fed with cold gas, decreases the temperature in the vicinity of the sensor.
[0015] The fluidic oscillator may be characterised by the presence of a sensor for evaluation
of the frequency of the generated oscillation and this sensor is positioned inside
the oscillator cavities where a periodic fluid flow takes place, with advantage in
the first output exit or in the second output exit.
[0016] The fluidic oscillator used for control of boundary layer on the surface of a body
exposed to fluid flow has at least one of its output terminals connected to an actuator
nozzle for generation of pulsatory flow-controlling jet.
[0017] The resonance channels may be straight or curved. Its length may be adjustable or
even continuously variable during the oscillator operation. Often requested property
is compactness of the whole oscillator including the resonance channel. To achieve
the compactness, the resonance channel may be meandering or shaped as a spiral. The
curvature radii of these versions should not be small - for example the radius should
be nit smaller than roughly one half of the total length dimension of the bistable
fluidic amplifier, because too small radius would lead to deformation of the spatial
shape of the pressure waves propagating in the channel and this might lead to irregular
oscillator functioning.
[0018] It is important that the fluidic oscillator according to this invention is constituted
of two parts. It is, on one hand, a known fluidic bistable jet-type amplifier, and
on the other hand the resonance channel.
[0019] Bistable jet-type amplifier contains cavities, such as in particular the supply nozzle,
first control nozzle, second control nozzle, the cavity for flow issuing from the
supply nozzle with the first attachment wall and opposite to it positioned second
attachment wall, the first output terminal and the second output terminal.
[0020] The spatial arrangement of the bistable amplifier is such that there is, on one side
of the supply nozzle, the first control nozzle and on the other, opposite side of
the supply nozzle there is the second control nozzle, where the first as well as the
second control nozzles are mutually arranged so that their exits are directed against
each other. Downstream from the first control nozzle is positioned the second attachment
wall and further downstream from it there is the second output terminal. On the other
side, downstream from the first control nozzle there is the first attachment wall
downstream from which there is the first output terminal. The first attachment wall
and the second attachment wall are located opposing each other symmetrically relative
to the axis of the supply nozzle. Positioned between the two entrances into the first
output terminal and the second output terminal may be a variously shaped splitter,
usually shaped more or less as a wedge. The second basic component of the oscillator
according to this invention is a resonance channel, which is connected to the first
control nozzle of the bistable amplifier while its opposite free end is open into
a stagnant space. The second control nozzle of the bistable amplifier is open into
the same or another stagnant space. This stagnant space may be the atmosphere. In
the usual planar version all parts of the amplifier are made for example by removal
of the material on the surface of a flat plate. All the cavities are then separated
from the surrounding atmosphere by a cover plate. The stagnant space may be any other
large space containing a fluid such as air, gas or liquid or a gaseous or liquid mixture.
Operating the fluidic oscillator according to the invention with liquid or liquid
mixtures is possible, but not advantageous because of the small compressibility of
usual liquids which leads to high propagation of pressure waves so that achieving
the usual range of frequencies means having an impractically long resonance channel.
It is, however, possible to use as the working fluid a liquid containing in sufficient
concentration small gas bubbles, the compressibility of which together with high density
of the liquid component of the mixture ensures low velocity of pressure changes propagation
in the resonance channel.
[0021] The material of the plate in which the fluidic oscillator is made is inessential
and the decisive role in its selection have practical manufacturing and operational
aspects. Similarly inessential is the material of the cover plates that close off
the oscillator cavities in the plate, because the reason why the cover plate is used
is only preventing the fluid - in particular gas - from leaving the cavities. The
material of the plates will be chosen in particular by consideration of the manufacturing
technology applied in making the cavities, especially in the used technology is not
a common one - such as the photochemical technology requesting sensitivity to light
(or other forms of electromagnetic radiation) used to for the photographic transfer
of the cavity shapes. The material in this case may also require special properties
needed for efficient removal of the plate material by etching or similar processes.
Nevertheless, these aspects are nowadays known in the field of fluidics and do not
represent any new material properties.
[0022] According to the disclosure of this invention it may be advantageous to have the
fluidic oscillator, with the cavities of the jet-type bistable amplifier made, e.g.,
by a photochemical procedure or by laser cutting on a numerically controlled machine
tool, made as a recession in a plate of the depth everywhere the same so that the
resonance channel is formed simultaneously using the same procedure in the same plate
as the recession of the same depth as are the cavities of the jet-type bistable amplifier.
[0023] From known fluidic oscillators differs the layout according to the present invention
only by the absence of the feedback channel or channels, in place of which the oscillation
generation is maintained by the resonance channel connected to the first control nozzle,
its one free end open into the atmosphere or another large space, while the second
control nozzle is also open into such space.
[0024] The main advantage of the fluidic oscillator according to the invention is the fact
that the frequency of generated oscillation is essentially dependent only on the length
of the resonance channel - and on the speed of propagation of pressure wave in the
resonance channel. It does not therefore depend on the magnitude of the fluid flow
rate passing though the oscillator. It is thus possible to achieve rather high frequency
of the generated oscillation without the necessity to work with high velocities of
the flow through the oscillator. This means it is possible to avoid high energetic
losses associated with high flow velocities. The new principle also avoids the unwanted
frequency variations caused in the so far known oscillators by accidental causes as
is the variations of the pressure of the supplied gas.
[0025] As was already mentioned above, the oscillator according to this invention makes
possible its being used to measurement of fluid properties of fluids - especially
gases - brought into the resonance channel. The advantage brought by this possibility
is the fact that the change of the properties - such as change in temperature or chemical
composition - causes the change of the frequency of generated oscillation. Frequency
is easily and exactly measured by a relatively simple and inexpensive sensor. The
output values coded in the oscillation frequency may be easily and with advantage
converted into a digital signal, suitable particularly for subsequent computer processing
of the acquired information.
Overview of the figures in drawings
[0026] The accompanying five pictures show five different examples of the fluidic oscillators
according to this invention.
[0027] In Figs. 1, 2, and 3 is presented a fluidic oscillator designed for delivering air
flow into actuator nozzles that suppress boundary layer separation on the wind turbine
blades. There is, in Fig. 1, a perspective view of the cavities of the oscillator
before their being covered by a cover plate. The next Fig. 2 is a scale drawing of
the cavities of an oscillator that was used in feasibility studies in aerodynamic
laboratory. The geometry of the amplifier has been known before and was already used
in several other oscillators described in available literature. Because of the drawing
to scale and of given most important basic dimensions, the shape of the cavities presented
in Fig. 2 is fully determined. With an oscillator of this geometry was experimentally
investigated the dependence of the frequency of generated oscillation on the length
of a straight resonance channel which is shown in Fig. 3. It is, of course, possible
to expect that with a different geometry the dependence would be somewhat different
or the oscillator would function in a different frequency range than shown in Fig.
3. Nevertheless, the general character of the dependence is likely to be more or less
similar.
[0028] In Fig. 4 is shown a layout for an application requiring very low frequency of generated
oscillation while retaining the requirement of compactness.
[0029] Fig. 5 present the oscillator according to this invention used for digital measurement
of high temperature gas.
[0030] Fig. 6 describes a version of the oscillator making possible adjustment of the oscillation
frequency and also perform a frequency modulation of the generated fluidic signal.
[0031] The last Fig. 7 then presents an alternative layout of the input part of the digital
gas temperature meter.
Examples
Example 1
[0032] In the example of the oscillator presented in Figs. 1 to 3, it is made of plastic.
It is one of a large number of relatively small oscillators positioned side by side
inside a wind turbine blade. Each of these small oscillators is connected by its output
terminals to two actuator nozzles from a row of such nozzles with its exits opened
at the blade surface. The oscillatory fluid flow from these nozzles achieves a control
of the character of the air flow past the blade adapting the functioning of the wind
turbine to the instantaneous wind velocity without the need of complicated and expensive
mechanical turning of the blade. Immediately responding and fast change of the character
of the flow past the blade may be also used for suppressing the changes of the aerodynamic
conditions arising when the rotating blade moves past the mast of the turbine. It
is a relatively small change in the conditions, nevertheless it is associated with
a change of the acting forces and because it arises periodically at each full circle
of the turbine shaft rotation. This may lead, as all periodic loading do, to fatigue
type collapse of the blade. Suppressing this effect by the action of the flow from
the actuator nozzles may thus bring substantial advantages concerning the dimensioning
and hence also price of the wind power station.
[0033] The oscillator in this example contains on one hand the bistable fluidic jet-type
amplifier
10 and, on the other hand, the resonance channel
1. The cavities of the oscillator are made by photoetching as a recession of the same
depth everywhere, formed in the plastic material plate. The cavities made this way
are then closed - so that the air from them cannot escape - by a flat plane cover
plate
100. Only a part of the cover plate
100 is seen in Fig. 1. It is a part covering the end of the resonance channel
1. Nevertheless, the same uniform covering of the plastic plate with the cavities is
made everywhere over the whole plate surface.
[0034] The working fluid is here air supplied under pressure into the supply nozzle
16. On one side of the exit from this supply nozzle
16 there is the first control nozzle
11, while on the opposite side is the second control nozzle
12. Next to the exit from the first control nozzle
11 there is the second attachment wall
14 - and symmetrically on the opposite side there is, next to the exit from the second
control nozzle
12, is positioned the first attachment wall
13. Further downstream in the direction of the flow from the supply nozzle
16 exit there is the first output terminal
17 and similarly downstream from the second attachment wall
14 there is the second output terminal
18. Between the two inlets into the output terminals
17,
18 is positioned a wedge-shaped splitter
6, in upstream part of which, opposite to the supply nozzle
16, the experience has shown it is advisable to make at the wedge tip a small trough-shaped
recession between two cusps, as it is particularly well apparent in the following
Fig. 2.
[0035] Fig. 2 provides a general information about a specific layout of the cavities, as
tested on a model. It is the layout in which all cavities were etched to the depth
2 mm (millimetres). The width of the supply nozzle
16 is also 2 mm. Both control nozzles with the mutually opposed exits, the first control
nozzle
11 and the second control nozzle
12 are of 1.4 mm exit widths. The small trough-shaped recession between two cusps in
the splitter 6 has radius 2.85 mm and the apex of this round recession is at the distance
13.78 mm from both control nozzles. Both attachment walls
13,
14 are planar and positioned symmetrically relative to the axis of the supply nozzle
16, from which each attachment wall is declined by the angular distance 20° so that
the angle between both attachment walls is 40°.
[0036] As seen in Fig. 2, the first output terminal
17 is connected to the first actuator nozzle
40 with its exit on the blade surface
20 in a blade of a wind turbine, while the second output terminal
18 is in a similar manner connected to the second actuator nozzle
50. The second control nozzle
12 is open into the free space, which in this case is the atmosphere. To the oppositely
located first control nozzle
11 is connected the resonance channel
1. Together with the cavities of the bistable fluidic amplifier
10 the straight resonance channel
1 is made in the same plastic plate. It is made by photoetching to the same depth and
as a result it is of rectangular cross section. It length is L. The free end
101 of the resonance channel
1, which is meant to be the end opposite to the one connected to the first control
nozzle
11, is open into the free space of atmosphere.
[0037] The oscillator made according to Figs. 1 and 2, after connection of its supply nozzle
16 to a compressed air source, starts self-excited oscillation in which the air jet
leaving the exit of the supply nozzle
16 is in an alternating manner switched from the first attachment wall
13 to the second attachment wall
14, back and forth. This switching is a consequence of pressure waves which travel through
the resonance channel
1, are reflected form its free end
101 and enter into the first control nozzle
11.
[0038] At designing of this oscillator it was requested that the outflow pulsation from
the actuator nozzles
40,
50 should be at frequency 300 Hz. At the same time, available for supplying the oscillators
were sources with relatively low air pressure. They did not suffice for achieving
the oscillation at that frequency level in the traditional oscillator layouts with
the feedback flow channels. On the contrary, in the layout of the oscillator according
to this invention, with the resonance channel
1, it was possible to obtain the requested frequency values even at rather low supply
pressure levels, because the frequency is determined by the length L of the resonance
channel
1, the supply pressure levels being almost irrelevant (considering the sources available
in a wind turbine). The requested frequency value 300 Hz was in this case obtained
with the resonance channel
1 length L equal to 140 mm, which is a size that made possible stowing the oscillators
inside the turbine blade without any problems.
[0039] In principle, the same design of the fluidic oscillator, with the actuator nozzles
40,
50 connected to it, may be also used for different applications than just the boundary
layer control. For example, the oscillation may be used for agitation of air in vessels
or spaces in which the solid particles carried by air would separate and sediment.
The agitation prevents such sedimentation. The actuator nozzles
40,
50 may be also positioned against a hot surface (e.g. a surface of highly loaded electronic
components) that should be cooled. The oscillating output flow from actuator nozzles
40,
50 cools more effectively than it is possible to obtain with steady, non-pulsating air
flow. Also the experience with drying wet surfaces has shown that with the pulsation
the removal of water vapour is mire intensive. The oscillation may also speed up or
intensify the process of mixing various fluids, such as, e.g., the reagents entering
into chemical reactions. Because some chemical reactions can be more rapid if the
reagents pulsate in small chemical reactors, it is possible to replace the actuator
nozzles
40,
50 at the output terminals
17,
18 of the oscillator directly by connected small chemical reactors. For the case in
which it is useful to know the oscillation frequency may be to the oscillator connected
- or positioned anywhere in side the oscillator - a sensor
3. A practical location for placing the sensor
3 is the first output terminal
17 or the second output terminal
18.
Example 2
[0040] In originally tested models of the oscillator according to this invention had always
a straight resonance channel
1. If there is a requirement of low frequency of generated oscillation, the corresponding
length L of the straight resonance channel
1 is really large and this leads to problems. One of them is the difficulty with stowing
the long straight resonance channel
1 in the available stowage space. Other problem is the manufacturing difficulty: the
overall dimensions of the oscillator may exceed the working length of manufacturing
facilities. Fig. 4 presents an example of the fluidic oscillator according to this
invention that was required to oscillate at a rather low frequency - and at the same
time was to be manufactured by laser cutting on a numerically controlled machine tool
with relatively small dimensions of working table.
[0041] Even in this example the working fluid was air supplied under pressure into the supply
nozzle
16. As is the case also in the other examples, on one side of the supply nozzle
16 exit is the first control nozzle
11 while on the opposite side is the second control nozzle
12. Further downstream in the direction of the flow from the supply nozzle
16 there is on one side the first attachment wall
13 leading to the first exit terminal
17 while on the opposite side there is the second attachment wall
14 and the second output terminal
18. These components are all made in a constant-thickness metal plate as cavities closed
form the top as well from the bottom by flap metal cover plates
100, not shown in this illustration. They are similar to the cover plate
100 shown in Fig. 1, which closed the oscillator cavities only from the top side since
in that case the cavities were of only a certain depth while here in Fig. 4 the metal
is removed through the full thickness of the plate. As in the precious example, the
resonance channel
1 is also in this second example connected to the first control nozzle
11 while its free end
101 is open to atmosphere - as is also the second control nozzle
12.
[0042] A special feature of this embodiment example is achievement of the desirable compactness
in spite of the relatively large length L of the resonance channel
1, which is here not straight but curved. There were fears, when this version was designed,
that the curvature would deform the fronts of the propagating pressure waves during
their passage through the resonance channel
1. The deformation could lead, in some extreme operational regimes, to irregular character
of the generated oscillation. Fortunately, experimental verifications proved that
if the curvature radii of the resonance channel
1 axis are not smaller than the characteristic dimensions of the bistable fluidic amplifier
10 the curvature does not have an adverse influence of the oscillator functioning. In
this particular embodiment, the resonance channel
1 of the overall length 317 mm could be placed together with the fluidic amplifier
10 in a metal plate of dimensions 97 mm x 160 mm, while the curvature radius of the
resonance channel
1 was mere 35 mm, which is only 3.5-times the width of the resonance channel
1 and 0.58-times the overall length of the fluidic amplifier.
[0043] The operation of this example of the fluidic oscillator was the same as in the previous
example and does not therefore need a detailed explanation. The oscillator was also
operated with actuator nozzles
40,
50. These were placed at a larger distance from the plate with the bistable amplifier
10 and the generated oscillations were led to them by metal tubes not shown in the illustration.
Example 3
[0044] Because the frequency of oscillation generated by the oscillator according to this
invention depends on the propagation velocity of pressure waves in the resonance channel
1 and this velocity depends on the temperature of the fluid - the effect particularly
strong if the fluid is a gas, it is possible to use the oscillator with advantage
in the role of temperature sensor. An example is the sensor for measurement of temperature
of combustion products - with the frequency-modulated output signal being a particularly
attractive feature. If it is made from a temperature-resistant material (in the case
discussed here the material is molybdenum alloy - but also suitable may be ceramic
materials) which has no other role than to keep its structural integrity, the sensor
may operate without limitations at temperatures so high that usual temperature sensors
would withstand them only for seconds of not less.
[0045] Such temperature sensor is presented in Fig. 5. Its bistable fluidic amplifier
10 is with respect to its shape and size identical to the example discussed above in
association with Fig. 1. The difference is on one hand in the cavities being in this
latter case the cavities are made in a small plate of molybdenum alloy, and on the
other hand is the presence of the additional second gas inlet
2. This has here the form of three inclined nozzles
22 issuing into the resonance channel
1. In addition, to obtain proper pressure distributions, there is the first fluidic
resistor
4 connected to the first output terminal
14. Similarly, there the second fluidic resistor
5 connected to the second output terminal
18. Also, a sensor
3 sensing air flow velocity is positioned in the first output terminal
17. In principle, the sensor may be placed anywhere in the oscillator cavities where
there are periodic changes in the flow. The positioning in one of the output terminals
17,
18 is particularly advantageous because the changes there are very large.
[0046] Also this oscillator after connection of the supply nozzle
16 to an air source enters the regime of self-excited oscillation. The frequency of
the oscillation depends on the speed with which the pressure waves move between the
both ends of resonance channel
1. The combustion products the temperature of which is to be measured, are introduced
into the second gas inlet
2. Their flow is directed by the inclined nozzles
22 so that they pass through the resonance channel
1 and leave it through its free end
101 into the atmosphere. This means the fill the path of the pressure waves. As in other
gases, the propagation velocity of the waves depends on the combustion product temperature
and as a result of this dependence the time taken by the wave to pass through the
resonance channel
1 length L. This means a change in temperature results in a change of the frequency
of generated oscillation. The sensor
3 should not be exposed tot he very high temperature of the combustion products and
this is ensured in this embodiment example by the fact that past the sensor
3 flows cool air supplied under pressure into the supply nozzle
16. To ensure this cooling activity is the task of the fluidic resistors
4,
5 through which the cool air leaves into the atmosphere. They ensure the cavities of
the bistable jet-type amplifier
10 are filled with air at a higher pressure which prevent the combustion products from
entering there.
Example 4
[0047] The fluidic oscillator according to this invention may be also used for measurement
of other gas properties as long as these properties influence the pressure wave propagation
velocity. As another example is here mentioned measurement of composition of binary
gas mixtures-the components of this mixture having a different propagation velocity.
The layout of this example is fully identical to the above discussed Example 1. The
difference is the presence of the fluidic resistors
4,
5 replacing the actuator nozzles
40,
50 in the same manner as was mentioned in the Example 3 shown in Fig. 5. Of course,
the difference is so to say a terminological one, because nozzles - such as the actuator
nozzle
40,
50 - are actually a common form of fluidic resistors. Also identically as in the previous
Example 3 shown in Fig. 3, the oscillator now contains a simple sensor reacting by
a change in its electric output signal to a change of the instantaneous flow rate
in the first output terminal
17. The investigated binary mixture - in this particular example it was a mixture of
hydrogen and carbon monoxide, cannot have (in contrast to the danger associated with
high temperature) an adverse effect on the sensor
3 so that the mixture is supplied directly into the supply nozzle
16. The fluidic resistors
4,
5 are adjusted so that the pressure drop on them will secure in the cavities of the
bistable jet-type amplifier
10 a pressure higher than atmospheric. This pressure difference will preclude any possible
mixing of the investigated gas mixture with atmospheric air, which may be in some
regimes sucked into the oscillator cavities and thus change the properties if the
investigated mixture. It is possible to adjust the fluidic resistors
4,
5 for keeping the pressure of the gas mixture in the cavities so high that some of
the investigated gas mixture will leave through the resonance channel
1 into the atmosphere through the free end
101. In the case of the investigated gas mixture being poisonous, explosive or otherwise
dangerous, the resonance channel
1 will lead it not into atmosphere but into a closed tank or vessel, from where it
will be led further into a neutralisation facility.
Example 5
[0048] The oscillator in its basic embodiment as shown in Fig. 1 generates oscillation of
practically constant frequency. The control of the action of the actuator nozzles
40,
50 connected to the oscillator output terminals
17,
18 is then only two-positional: the driving gas (air in the case of Fig. 1) is either
brought into the oscillator or not. Thus the control of the actuator nozzles
40,
50 is either acting - at a given frequency - or not. There are, however, applications
of the fluidic oscillators where it is advisable or even requested to adjust the control
action or perhaps even vary it continuously. In particular, it may be requested to
adjust the frequency of generated oscillation, e.g. according to the local conditions
- or the task may be to vary continuously the frequency during the oscillator operation.
The frequency change may be dependent on some input signal fed into the oscillator
so that the generated oscillation is frequency-modulated. The Example 5 as shown in
Fig. 6 represents a possible arrangement of the variable-frequency oscillator. It
uses the fact of the frequency being dependent on the length L of the resonance channel
1. The frequency variation is in this case achieved by varying this length.
[0049] The jet-type bistable fluidic amplifier
10 itself as shown in Fig. 6 is essentially identical with the layout presented in Fig.
2. The working fluid is air supplied under pressure into the supply nozzle
16. On one side of the exit from this nozzle is the first control nozzle
11 and on the other, opposite side is also here the second control nozzle
12. Next to the exit of the first control nozzle
11 there is the second attachment wall
14 - and symmetrically on the other side is next to the exit of the second control nozzle
12 is located the first attachment wall
13. Further downstream along the flow from the supply nozzle
16 there are the first output terminal
17 and the second output terminal 18. Between them there is - as was already in Fig.
2 - a wedge-shaped splitter
6, again with the groove between the two cusps. The first output terminal leads to
the first actuator nozzle
40 while the second actuator nozzle
50 is connected with the second output terminal 18. Both output terminals
17,
18 are here longer than was the case in Fig. 2 above, but this is not of any significant
consequence for the oscillator operation. The second control nozzle
12 is on its inlet side open into the reference space which in this case is the atmosphere.
To the first control nozzle
11 on the opposite side is connected the resonance channel
1, which in this example is not straight but, in a certain kinship to the example presented
in Fig. 4, its axis is curved. Also here the free end
101 of the resonance channel
1 is open to the reference space - the atmosphere.
[0050] To make possible the changes of the resonance channel
1 length, this channel is led in an arc formed in the slider
120. This is so shaped that it may be, by an action on the eyes
125, moved in the direction of the arrows S without the resonance channel
1 being discontinued. This is achieved by the fixed part of the resonance channel
1 being made in a body with two excrescences
122. This fit into the corresponding recessions in the slider
120. The wall of the excrescences
122 are very thin so that the cross sections in the resonance channel 1 in these locations
do not vary substantially along the channel length if the slider is shifted at larger
distances from the body with the fixed channel part. In adjustable version of this
tuneable oscillator the changes in the position of the slider
120 are made, e.g., by turning a screw and then fixing the slider
120 in its new position. In the version with modulation of the carrier oscillation frequency
the slider
120 is moved by an electromechanical transducer, for example based on the piezoelectric
or electrodynamic (vice-coil) principle.
Example 6
[0051] In a similar manner as in the previous illustration, also the last Fig. 7 shows the
oscillator according to this invention used as a sensor for digital measurement of
high temperature of a gas (or gas mixtures). The difference is in the second gas inlet
2 being not led by the inclined nozzle(s) into the resonance channel 1, but is led
into the supply nozzle
16. The inclined nozzle are here not needed. The geometry of the cavities immediately
downstream from the supply nozzle
16 exit is identical to the already above described geometry examples. There is again
one side of the exit from the supply nozzle
16 the first control nozzle
11 and on the other, opposite side is the second control nozzle
12. Next to the exit of the first control nozzle
11 there is the second attachment wall
14 and symmetrically on the other side is next to the exit of the second control nozzle
12 is located the first attachment wall
13. Further downstream along the flow from the supply nozzle
16 there are the first output terminal
17 and the second output terminal
18. Between them there is the wedge-shaped splitter
6, again with the groove between the two cusps. The first output terminal leads to
the first actuator nozzle
40 while the second actuator nozzle
50 is connected to the second output terminal
18. The resonance channel 1 is also here connected to the first control nozzle
11 while its free end
101 is open into atmosphere.
[0052] The difference is in this last example in there being not only one fluidic amplifier:
the first output exit terminal
17 is here led onto the second control nozzle of the second amplifier
212 and symmetrically the second output terminal
18 is connected to the first control nozzle of second amplifier
211. The supply nozzle of the second amplifier
216 is connected to a cold air source
202. Opposite to the supply nozzle of the second amplifier 216 there are also two outlets,
the first output terminal of the second amplifier
217 and second output terminal of the second amplifier
218. Into the first output terminal of the second amplifier
217 is placed a sensor
3 reacting to the local pressure variations. Both output terminals of the second amplifier
217,
218 lead further downstream into the common outlet
220.
[0053] There were two problems solved in designing this input part of a digital gas temperature
meter. Firstly, the temperature in the second gas inlet
2 is here so high that if the sensor
3 were to be chosen so as to withstand it permanently, its price would be inconveniently
high. It would be probably necessary to complicate the sensor by an integral cooler.
The second problem was in this particular case a too low available pressure in the
second gas inlet
2. With this low supply pressure the amplitude of the generated oscillation would be
small so that the sensor
3 would have to be very sensitive and, again of significantly higher price. Both problems
are solves simultaneously by the addition of the second fluidic amplifier. Through
the second amplifier flows - and thus is led to the sensor
3 - the cold air through the cold air inlet
202. An the same time the amplification property of the second fluidic amplifier is used
to amplify the generated oscillation. The sensor
3 thus does not need to be very sensitive and may be significantly cheaper. Problems
such as the imperfect linearity that may be encountered with fluidic amplifiers are
of no consequences here since all what has to be asked for is just transfer of the
frequency, which is not influenced by possibly imperfect properties of the second
amplifier. There are no fluidic resistors in this example connected to the output
terminals
17,
18, because their respective roles are taken over by the control nozzles of the second
amplifier
211,
212. The pressure from on them secures the flow of the hot gas through the resonance
channel
1.
Industrial applicability
[0054] Fluidic oscillators in general replace currently the so far usually used mechanical
pulsators serving to generate periodic oscillation in fluids, for example for the
purposes of increasing the intensity of transport processes such as heat transfer
(cooling or heating) or mass transfer. Firs of all, however, the subject of the present
invention is expected to be used for control of boundary layer separation from the
surface of bodies exposed to fluid flows and transition in turbulence. This way it
is possible to achieve increased effectiveness of aircraft lifting surfaces but also,
when used in gas, steam, and wind turbines, increased efficiency of such turbomachines.
[0055] The above embodiments of the invention as well as the appended claims and figures
show multiple characterizing features of the invention in specific combinations. The
skilled person will easily be able to consider further combinations or sub-combinations
of these features in order to adapt the invention as defined in the claims to his
specific needs.