Note: Descriptions are shown in the official language in which they were submitted.
.-~LE, P~i Tl; ~ h...-~J-- 219~!583
~ T~ RAl~ LATlON
A FLUIDIC OSCILLATOR AND--A METHOD OF MEASURING A VOLUME-
REIIATED QUANTITY OF FLUID FLOWING THROUGH SUCH A FLUIDIC
OSCILLATOR
The prèsent invention relates to a fluidic
5 oscillator and to a method of measuring a volume-related
guantity of fluid flowing through said fluidic
oscillator .
It is known to use fluidic oscillators for measuring
for a volume-related guantity of fluid such as flow rate
10 if the frequency of the oscillations is measured, or
volume if the number of oscillations is counted. Such a
fluidic oscillator is ~cr~hl~, for e~ample, in French
patent application No. 92 05301 filed by the Applicant,
and based on detecting the freguency of oscillation of a
15 two-~li c~nniql fluid jet in an oscillation chamoer.
The fluid jet is formed as the liguid flow passes
through a slot which opens out into the oscillation
chamber, and it oscillates transversely relative to the
plane of longitudinal ~y Lly of the fluidic oscillator.
20 An obstacle is placed in the oscillation chamber and
possesses a cavity in its front portion, which cavity is
placed on the path of the fluid ~et so that said jet
sweeps the walls of the cavity during oscillation. Flow
rate mea ;UL~ ~ is performed, for example, by detecting
25 the ~et sweeping over the bottom of the cavity as it
oscillates, with the freguency of oscillation of the jet
being proportional to the fluid flow rate.
Also known from patent application GB-A-2 120 384 is
a fluidic oscillator operating on a somewhat different
30 principle since it is a Coanda effect oscillator, but its
end purpose remains measuring a volume-related guantity
of a fluid by detecting the nSr~ t; nn freguency of a
jet of fluid. That fluidic oscillator includes three
obstacles housed in an oscillation chamber, two of the
35 oscillators being situated on opposite sides of the
longitudinal plane of :,y ~ y immediately downstream
from the fluid ~flmicclnn opening into said chamber and
z 2192583
co-operating with the si~e walls of the oscillation
chamber to form two symmetrical ~h~nnPl s, and the third
obstacle is disposed facing the fluid .9fl-"1 CR~ nn opening,
but downstream from the first two obstacles at the sides.
During its sweeping motion, the fluid ~et meets one
of the side obstacles, and attaches thereto, the fluid
flow then moving Ul):.i,L~III and taking the channel formed
between said obstacle and one of the walls of the
oscillation chamber, thereby causing the fluid to
10 circulate again.
When the fluid flow reaches the upstream zone
situated close to the fluid ;~rlml cqi nn opening at which
the base of the fluid ~et is situated, the flow then
causes said ~ et to switch over towards the other side
15 obstacle and the same rhPn~ is reproduced with said
other side obstacle. The fluidic oscillator also
~n,~ Pc two ultrasound tr~n~ Prs disposed on either
side of a fluid ilow position such that ~ltrasound
signals are emitted and received in planes that are
20 substantially transverse to the lons~itudinal direction of
fluid flow.
In the meas~rement method described, one of the
tr~nq~llnPrs emits towards the other tr~nC~lllrPr which is
placed downstream from the first or in the same
25 transverse plane as the first, and the emitted signal is
modulated by the oscillations of the fluid jet in the
oscillation cham]~er so the other transducer receives said
ultrasound signal as modulated in this way.
On the basis of the received signal, it is possible
30 to detect the frequency f of oscillation of the fluid
~et, and to deduce therefrom the flow rate or the volume
of the fluid that has flowed through the fluidic
oscillator .
That Britis~l patent application then ~Yrl~nc that
35 the received signal is ~ ted and transformed into a
pulse signal where each pulse corresponds to a unit
2192583
volume of fluid swept through by the fluid jet during its
oscillation .
That meaYul~ t technique provides the advantage of
achieving good measurement repeatability over the usual
range of flow rates for a fluidic os~ tor, However,
in some cases, it is n~. Pqq~ry to obtain very good
accuracy in flow rate or in volume mea,Ul, t, and
conseguently it can be advantageous to have a fluidic
oscillator whose sensitivity can be easily improved over
its usual range of flow rates.
In addition, it is known that a fluidic osclllator
cannot measure a volume-related quantity of a fluid when
the flow rate of the fluid is so low that it is no longer
possible to detect the freguency of fluid ~et
oscillations.
In domestic installations, it is also known that,
for most of the time, the flow rates of a fluia, e.g.
gas, are very lo~, being typically less -~han 200 liters
per hour (l/h). It is therefore particularly important
to be able to measure flows at such rates as well as
being able to measure the maximum values of flow rate
that can happen oo-f~q;nn~lly~ In addition, it is also
necessary to be able to detect fluid leaks when they
occur and thus it must be pnq%l hl F- to distinguish a
leakage rate from a small flow rate.
The present invention seeks to remedy the drawbacks
of the prior art by proposing a fluidic oscillator
structure that is easily adapted to more accurate
meaYul~ t of a volume-related guantity of a fluid over
the usual range of flow rates for said fluidic
oscillator, should that be n~nPqq;~ry, and which is also
easily adaptable to rr~qllr;n~ a small volume-related
quantity of a fluid, at which fluid jet oscillations
disappear .
The invention also proposes a method of measurement
that is adapted to measuring a volume-related guantity of
a fluid in each of the above-s~r~f;~ cases.
~ 2192583
4
Thus, the present invention provides a fluidie
. oscillator that is sy-mmetrical ~hout a longitudinal plane
of ~y Lly, the oscillator eomprising:
means for generating a two-fl~ ~c1nn:l1 jet of fluid
5 that oseillates transversely relative to said
longitudinal plane of ~y LLY;
two ultrasound tr~nc-lllrPrc3;
and means firstly for generating an ultrasound
signal in the fluid flow travelling from one of said
10 transdueers towards the other, and secondly for receiving
said ultrasound signal as modulated by the oscillations
of the ~ et of f luid; and
means for proeessing the received signal so as to
determine a volume-related quantity ~.r,n.~Prn~ng the fluid
15 that has flowed through said fluidic oseillator,
charaeterized in that the ultrasound tr~ncfll~rPrs are
substantially in ~ , t with the longitudinal plane of
~Y L1 Y,
This novel disposition of the ultrasound transdueers
20 in a fluidic oscillator is highly advantageous firstly
beeause by ~ h~Qc~ ns to plaee said tr~neflll~ prs
5pp-~lf~ 1y in the plane of ~y L~Y or slightly offset
therefrom, it is possible to promote detection at the
oseillation freguency of the ~et or at twiee that
25 frequeney, thereby increasing the sensitivity of said
fluidic osclllator, and secondly, while using the same
tr~ncfl-lr~.Prs positioned in this way and whether operating
at the oscillation frequency or at twice the oscillation
frequeney, it is possible to measure small volume-related
30 quantities of a fluid for which the oseillations of the
fluid ~et are too small for it to be pQCc~h1P to deteet
the frequency thereof.
The fluidie oscillator obtained in this way is
referred to as a ", ' ~nPfl" oscillator and it covers a
35 range of flow ~ra~es that is wider than the usual range of
flow rates for a conventional fluidic oscillator.
21~2~83
To this end, provision is made for the fluidic
oscillator to include:
means for emitting and receiving an ultrasound
signal alternately from each of the ultrasound
5 tr~nqfl~ Prq; and
means responsive to each received ultrasound signal
and tc consecutive pairs of ultrasound signals to
determine a value for a magnitude that is characterlstic
cf the propagation speed of said ultrasound signal as
~fl~ ~ Pfl by the fluid flow, and to deduce therefrom a
volume-related quantity applicable to the fluid that has
flowed through said fluidic oscillator.
The ultrasound transducers are disposed in different
transverse planes, an "upstream" one of said tr~ncfl~ Prs
being ~; qpn5Pfl upstream from the means for generating the
two-fl~ -q1nn~1 let of fluid, the other tr~nqdllrPr being
a "downstream" trPnqfl11nP~.
Given that the fluid jet oscillatio~ ~h~.n~ `nnn iS
observed over a large range of flow rates covering the
higher values cf the flow rate, the range of rates or
which fluid jet oscillations are too weak to be capable
of being detected is relatively small.
Consequently, in this small range, it is possible to
use tr~nqA11rPrs that are highly resonant, and therefore
of relatively simple design.
For exampler the magnitude representative of the
speed of propagation of the ultrasound signal may be the
propagation time of said signal. Alternatively said
magnitude may be the phase of said signal.
In an emoodiment of the inven~ion, the fluidic
oscillator comprises:
means for generating a two-fll- ,q~nn~l cscillating
fluid ~et, which means are formed by a fluid admission
opening of transverse dimension or width d and of height
h;
an oscillation chamber connected at one oi its ends
to said ~luid ;lflm~qq~ nn opening and at its opposite end
~ 2~25g~
.
to a fluid outlet opening, said openings being in
Al; ~, ~ L in gaid longitudinal plane o symmetry; and
at least one obstacle disposed in sald oscillation
chamber between the A~m~ CC~ nn opening and the fluid
5 outlet opening.
According to other characteristics ol~ the fluidic
oscillator:
the upstream ~rAnqfl~lr.Pr is disposed U~L1~:dlll from
the fluid Aflm~ cc1 nn opening;
the obstacle has a front portion in which a cavity
is formed facing the fluid Afl~ cq~ nn opening;
the downstream trAncfl~ .or is secured to the
obstacle; and
the downstream transducer is disposed in the cavity
15 of the obstacle.
According to other characteristics o~ the inventicn:
the oscillator ~n~1~7flPc, upstream from the obstacle,
a passage for the fluid that is defined by two walls that
are perppnfll slll Ar to the longitudinal plane o~ :,y Lly
20 and that are spaced apart by a distance _;
the oscillator ~nrlllflPq, upstream from the fluid
Aflm~ cqi nn opening, a longitudinally-extending channel
iorming at least a portion of the passage for the fluid,
said channel being of substantially ccnstant width d that
25 is perpendicular to the distance h;
the channel pcssesses, at one of its ends, a
"downstream" opening that corresponds to the fluid
A(9Ti qq~ nn opening, and at its opposite end, an "upstream"
opening which, iII a plane parallel to the flow direction
30 of the fluid and perpf~nfl~ l =r to the longitudinal plane
Of ~y ~ Lly, is .iullv~l!,~llt in shape, its width tapering
progressively do~n to the width d;
it ~ n~-l llflPq two fluid inlets disposed symmetrically
abcut the longitudinal plane of ~y LLY and opening out
35 into the passage, upstream from the channel;
2192583
it ~ nrl t~ q two flui-d inlets disposed :,y l.Llcally
about the longitudinal plane of symmetry and openlng out
into the passage, upstream from the channel;
two side passages e~tending in a direction that is
generally parallel to the longitudinal plane of
symmetrical and in particular to the longitudinal
direction of the channel each constitutes a f luid inlet,
each of said E)assages being connected f irstly via one end
to a common first chamber perpl~nrlirlll~r to said plane and
secondly, via an opposite end, to a common second chamber
parallel to said first chamber, said first chamber being
provided with a fluid feed;
an empty space forming another portion of the
passage for the fluid is provided upstream of the channel
and the two fluid inlets open out into said empty space;
the upstream transducer is disposed upstream f rom
the empty space;
in that the ultrasound tr~nqflllrF~rs ~re situated on
the same side in a direction perpf~nril r~ r to the
longitudinal direction of the fluid flow, and contained
in the longitudinal plane of ~_y [,Ly;
in that bot~l ultrasound tr~nC~ rprs are secured to
the same one of the walls flef~nins the passage for the
f luid;
the ultrasound tr~nq~lllrGrs are offset in a direction
perp~n~llr--l~r to the longitudinal direction of fluid flow
and contained in the longitudinal plane of symmetry; and
in that each ultrasound transducer is secured to a
respective one of the walls ~ f;n~n~ the passage for the
fluid.
The invention also provides a method of measuring a
volume-related s~uantity of a fluid ilowing through a
fluidic oscillator in which a jet of fluid oscillates
transversely about a longitudinal plane of symmetry, said
method consisting successively in:
emitting an ultrasound signal into the fluid flow
from an ultrasound tr;~nc~lr-pr;
~ ~ 219,~,583
.
receiving said ultrasound signal as modulated by the
oscillations of the ~et of fluid by uslng another
ultrasound transducer; and
procPcc~ng the received signal so as to determine
said volume-related quantity of the fluid that has flowed
through the oscillator;
the method being characterized in that it consists in
emitting an ultrasound signal in a direction that is
substantially contained in the longitudinal plane of
~y Lly.
Advantageously, by ~lign1nJ the ultrasound
tr~ncA~IrPrs accurately in the longitudinal plane of
~>y l,Ly of the fluidic oscillator and by emitting an
ultrasound signal in said plane, and in the fluid flow
direction, said ultrasound signal which is modulated by
the oscillations of the ~et of fluid and as picked up in
said longitudinal plane of sylrmetry is affected mostly by
the frequency 2f where f represents the ~Erequency of
oscillation ol~ said ~et of fluid. By detecting the
frequency 2f, it is therefore possible to double the
sensitivity of the fluidic oscillator over its usual
range of flow rates, i.e. at flow rates for which the
oscillations of the ~et o~ fluid are detectable. This
frequency of 2f cannot be detected when applying the
tprhn~r~l tP~rh~nfJ of patent application GB-A-2 120 384.
In contrast, when it is desired to use the
combination fluidic oscillator to cover the widest
possible range oi flow rates without seeking to improve
the sensitivity of said oscillator, it is not nPrPQC1~ry
to position the ultrasound tr~ncfll~rprc accurately in the
longitudinal plane of symmetry. The ultrasound
transducers are then substantially aligned with the
longitudinal ~lane of symmetry in such a manner as to
present an inclination of 1 to 2 relative to the
longitudinal direction of said plane of symmetry.
Given triqnsfl- rPrs disposed in this way for measuring
a volume-related quantity of flui~ at a high flow rate,
2~58~
.
g
i.e. when the oscillations of the jet of fluid are strong
- enough to enable the frequency thereof to be detected, it
is possible to emit an ultrasound signal in the f low
direction of the fluid through the fluidic oscillator.
5 The ultrasound signal as modulated by the oscillations of
the jet of fluid and as picked up is then affected mostly
by the freguency of oscillation f of said jet of fluid.
With the transducers in this disposition, an ultrasound
signal i5 preferably emitted in the direction opposite to
lO the fluid flow direction through the fluidic oscillator
for the purpose of improving detection of the oscillation
frequency f over the situation where the ultrasound
signal is L,L~a~c l lng in the same direction as the fluid
flow direction.
By having the tr~nqA~ rs in this advantageous
disposition, it is also possible to measure a volume-
related ~uantity of a fluid at a low flow rate, i.e. when
the oscillations of the jet of fluid arP' too small to be
detectable and the method of the invention then consists
successively in:
emitting an ultrasound signal from one of the
transducers towards the other in a direction that is
substantially contained in the longitudinal plane of
~y I,Ly;
receiving said ultrasound signal whose speed of
propagation has been modified by the flow of the fluid;
de~Prm1n;n~ a first value of a magnitude
characteristic oi` said speed of ~,L~,~a~ ion of the
received ultrasound signal;
repeating the above steps after interchanging the
emitter and receiver functions of the ultrasound
tr~ncfl~P~s and de~Prm~n~n~ a second value for said
magnitude characteristic of the speed of propagation for
another ultrasound signal; and
APfl~ n~ therefrom the measurement of a small
volume-related quantity of the fluid.
2~S83
.
The invention i5 particularly advantageous in the
field of gas metering.
Other characteristics and advantages appear from the
following description given hy way of non-limiting
5 illustrative example and made with re~erence to the
~c~ ylng drawings, in which:
Figure 1 is a dia~L Llc longitudinal section view
on the longitudinal plane of ':jy Lly P of an embodiment
of the fluidic oscillator of the invention
Figure 2 is a dia5~1 Llc view of the Figure 1
fluidic oscillator on plane Pl;
Figure 3 is a fr~ Lclly diayl Llc vlew in
section on A through the embodiment of the fluidic
oscillator as shown in Figure l;
Figure 4 i5 a dia~l Llc view on the plane Pl
showing a variant embodiment of the fluidic oscillator
shown in Figure 2;
Figure 5 shows a first variant embo~Liment of the
disposition of tlle ultrasound transducers shown in
Figure l;
Figure 6 shows a second variant embodiment of the
disposition oi the ultrasound tr~ncu1~ c shown in
Figure l;
Figure 7 shows a third varlant f~nho~l 1 nt of the
disposition of the ultrasound transducers shown in
Figure 1;
Figure 8 shows a fourth variant embodiment of the
disposition of the ultrasound transducers shown in
Figure l;
Figure 9 is a diagrammatic view on the plane Pl of
Figure 1 through a second embodiment of a f luidic
oscillator o the invention;
Figure 10 is a block diagram showing a portion of
the electronic circuit used for measuring the volume of
gas flowing through the fluidic oscillator;
Figure 11 is a detailed view on a larger scale of
the electronics block 100 in Figure 10;
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ .
` 21~258~
.
11
Figure 12 shows the~modulated ultrasound signal as
l;f;~fl by ~mrl;f;Pr 102 in Figure 11;
Figure 13 shows the ultrasound signal of Figure 12
after it has been rectified by circuit 104 in Figure 11;
Figure 14 shows the ultrasound signal of Figure 13
after it has been filtered by electronics block 106 of
~igure 11;
Figure 15 shows the ult~asound signal of Figure 14
after passing through electronic ~mr1;f;r~tion block 108
and through electronic peak detection block 112;
Figure 16 shows the operation performed on a peak by
the peak detection electronics block 112; and
Figure 17 is a calibration curve for the fluldic
oscillator of the invention.
As shown in Figures 1 and 2, and as given overall
ref erence 1, the f luidic oscillator of the inventlon is
aprl; t~.;`hl F', for example, to domestic gas metering, and it
rmq5~qs~oq a longitudinal plane of symme~y P which is
disposed vertically and which corresponds to the plane of
20 ~igure 1.
It should be observed that the fluidic oscillator
may also operate in a position such that its plane P is
horizontal or even in some other position without that
disturbing the measurement of the volume-related quantity
25 of a fluid (e.g. flow rate or volume proper). The fluid
flowing through said fluidic oscillator is a gas, but it
could equally well be a liquid, e. g . water.
The f luidic oscillator shown in Figure 1 has a
vertical gas ~eed 10 that is centered relative to the
30 longitudinal plane of -;y - Lly P and opens out into a
"top" first horizontal chamber 12 which is of large size
and disposed symme~rically about said plane P. The flow
section of said top chamber 12 is rectangular in shape
and parallel to the longitudinal plane of symmetry P, and
35 it sub~ects the flow of gas that enters via the feed 10
to a sudden increase in section, e . g . equal to a f actor
.. _ .... _ .. . . . _ _ _ _ _ _ . . . . _ . . ... .. . ., _ _
21g2~8~
12
of 4, in order to destro~ the turbulent :,~ u~,LuLt: of the
flow by reducing its speed.
The top chamber 12 possesses two opposite "end"
openings 12a and 12b each opening out into a respective
5 vertical side passage 14 or 16 ( as shown in Figures 2 and
3 ) of rectangular flow section identical to the flow
section of said first chamber 1. The two vertical side
passages 14 and 16 are symmetrical to each other about
the longitudinal plane of :,y ~Ly P.
Each side passage 14 (16) ~ r~tes firstly at a
"top" one of its ende 14a (16a) that ro;nr~q with the
corr~qp~-nl~n~ end 12a (12b) of the top first chamber 12,
and secondly at its opposite or "bottom" end 14b (16b)
with a "bottom" second chamber 18 that is identical to
the first chamber 12, as shown in Figure 1. The top and
bottom chambers 12 and 18 are symmetrical to each other
about the plane Pl shown in Figure 1 and they are
parallel to each other, but it would als~ be possible or
the volume of the bottom chamber 18 to be smaller.
Each of the two vertical side passages 14 and 16
constitutes a gas inlet and has a middle side opening
14c, 16c of ~ow section parallel to the longitudinal
plane of symmetry P (Figure 2). The gas inlets 4 and 16
open out via their side openings 14c and 16c into an
empty space 20 situated halfway between the bottom and
top chambers 18 and 12. The empty space 20 which forms
an in~ te chamber of smaller size than the chamber
12, possesses a "transverse" long ~; q;rn which is
perpPn~irul~ to the longitudinal plane of symmetry P,
and it is defined along said ~ q;~ln firstly by an
"upstream" end wall 22 and secondly by a "downstream" end
wall 24. The walls 22 and 24 are spaced apart by a
distance which corresponds to the longitudinal ~1; nc;rn
of the side openings 14c and 16c. A channel 26 in
~ on the longitudinal plane of symmetry P is
provided through the downstream wall 24. This channel 26
is ref erred to as the " main " channel and has a transverse
. , ... ,, . , . , .. , . , . .... .,, . . , . . _ , _ _ ,, ,,, , _ _ _
21~2~83
13
nn or "width" d that is substantially constant
along the entire longitudinal ~ n or "length" of
the channel . The length of the channel is pref erably
greater than lOd in order to obtain good accuracy in
5 measuring the volume-related ~uantity of a gas at low
flow rates, i.e. when the oscillation8 of the ~et of gas
are too weak for their frequency to be detectable. As
shown in Figure 2, the upstream opening 26a of the mzin
channel 26 has a shape that Cullv~ 8 in the plane Pl.
10 Each portion of the upstream opening 26a which is on one
side or the othe~ of the plane P has a convex profile,
e. g . a substantially rounded profile, thereby
contributing to progressively reducing the width of said
opening down to the width d of the main channel 26.
In a variant of the invention shown in Figure 4, the
gas inlets or vertical side passages 14 and 16 (not shown
in this figure) open out via side openings 14c and 16c
into respective "gas inlet" horizontal ~ nnF~l c 15 and
17. The flow section of each horizontal channel tapers
20 progressively. The two gas inlet r~h~nn-~l c 15 and 17 are
symmetrical about the plane P and they meet in a zone 19
situated in said plane P immediately upstream from the
upstream opening 26a of the main channel. The horizontal
~hFInnPl C are defined firgtly by a common "upstream" end
25 wall 21 that pro ~ects in a downstream direction, and
secondly by a common " downstream" end wall 23 . As
described above, a main channel 26 in ~1 i!j -nt on the
longitudinal plane of gymmetry P is provided in the
downstream wall. The gas inlet ~.h~nni~l s 15 and 16 and
30 the main channel 16 thus form a horizontal passage for
the gas which is defined above and below by a top wall
and by a bottom ~all ( not shown in the plane of Figure 4 )
that are spaced apart by a height h.
The fluidic oscillator shown in Figures 1 and 2
35 ;n~ means for generating a two-~ nc;r~n~l jet of
gas that oscilla~es transvergely relative to the
longitudinal plane of gymmetry P . These means are f ormed
21~2583
14
by a gas ~11m~ ec1 on openi~g into an "oscillation" chamber
32, which opening nn1nc1tlpe with the downstream opening
26b of the main channel and is rectangular in shape. The
oscillation chamber 32 has one of its ends connected to
5 the downstream opening 26b of the main channel 26 and has
its opposite end connected to a gas outlet opening 24 of
width greater than d. The gas ~m1sc1nn and outlet
openings 26b and 34 are in ~ L on the plane P.
The fluidic oscillator also includes an obstacle 36
10 of height h disposed in the center of the oscillation
chamber 32 between the gas A~9m1 c.51 nr~ opening 26b and the
gas outlet opening 34. A horizontal passage for the gas
situated upstreanl from the obstacle 36 is formed in part
by the empty space 20 and the channel 26 ana is defined
15 above and below respectively by a "top" wall 28 and by a
"bottom" wall 30 ( Figure 1 ) . These two walls 28 and 30
are separated from each other by a height =. Such an
obstacle 36 has already been described i-~ French patent
application No. 92 05301. The obstacle 36 has a front
20 portion 36a in which a cavity 37 is formed, referred to
as the "central`' cavity, which cavity faces the ~lm1 es~ on
opening 26b of the oscillation chamber 32.
Two secondal~y cavities 38 and 39 are also provlded
in the front portion 36a of the obstacle 36 symmetrically
25 about the plane P. The oscillation chamber 32 possesses
side walls 40 and g2 of a shape that substantially
matches the outside shape of the obstacle 3 6, thereby co-
operating ~ith said obstacle to provide two ~y Ll lcal
secondary r.h;~nn~l e Cl and C2 situated on either side of
30 the longitudinal plane of :,y L~ y P.
The width of the secondary nh~nnPl c C1 and C2 is
substantially ~ull~ L,--t in order to avoid disturbing the
~low of gas. The secondary rh;snnPl e Cl and C2 pass round
the obstacle 36 and meet again downstream therefrom in a
35 zone 44 situated immediately upstream from the outlet
opening 36 of the oscillation chamber 32. This outlet
opening 34 opens out into a vertical passage 36, at a
........ . .... . ... . . .. . . _ ... , , _,, , _
2~258~
point halfway up it, as shown in Figure 1. The vertical
passage 46 is, for example, symmetrical about the
longitudinal plane of symmetry P and at a "top" one of
its ends 46a it has a vertical gas outlet 48 centered
5 relative to said plane P. The configuration described
with reference to Figures 1 to 3 has the advantage of
conferring satisfactory compactness to the fluidic
oscillator.
The v~ 1~ o~ the flow of gas in the fluidic
10 oscillator are now described. A vertical flow of gas is
fed to the fluidic ncn7llRtnr via the vertical feed 10
and penetrates into the top chamber 12 of said fluidic
oscillator, where it splits into two portions. These two
portions of the "main" flow travel hori70ntally through
15 the top cha7nber 12 of the fluidic oscillator in opposite
directions perppnfl~ 7l ;7r to the longitudinal plane of
symmetry P. As shown in Figure 3, each portion of the
flow passes through an end opening 12a (~2b) of the top
chamber 12 of the fluidic oscillator and penetrates into
20 one of the vertical side passages 14 ( 16 ), performing
rotary motion prior to being engulfed in the empty space
20 via one of the side openings 14c (16c).
This configuration is flPsl ~nPfl to enable the gas to
get rid of any polluting particles with which it might be
25 charged ( dust . . ) as it passes through the vertical side
passages 14 and 16 where, under the e~fect of gravity and
of the rotary motion of the flow, such particles are sent
towards the bottom chamber 18 of the fluidic oscillator.
h7hen the two portions of the flow of gas penetrate
30 symmetrically about the plane P into the empty space 20,
they meet on sai~ plane P and are engulfed in the main
channel 26 via its upstream opening 26a. The flow of gas
then travels along the main channel 26 and is trnn~f~ --7
into an oscillating two-flir-nc~nn~l jet at the downstream
35 opening 26b thereof. Within the oscillation chamber 32,
the f low of gas alternates between the channel C1 and the
channel C2 prior to reaching the outlet opening 34 and
2192583
16
then flowing up the vertical passage 46 towards the
vertical gas outlet 48.
Given that the vertical passage 46 extends
vertically to a level below that of the oscillation
5 chamber 32, it too can serve to rid the gas of certain
polluting particles if they have not already been
eliminated. As ~entioned above, the fluidic oscillator
may be placed in some otXer position in which there is no
need to provide a bottom chamber 18 for the purpose of
10 removing dust from the gas.
In accordance with the invention, the fluidic
oscillator has two ultrasound tr~nq~llrprq 52 and 5g that
are substantially in ~ L with its longitudinal
plane of ~y LLY. The advantage of removing the ma~or
15 fraction of polluting particles from the gas is to avoid
contaminating the tr~nq~llrprs~ thereby increaslng their
lifetime .
In the embodiment of the invention ~hown in Figures
1 to 3, the ultrasound tr~nqd~lrprs 52 and 54 are offset
20 angularly by about 1. 5 from the longitudinal plane of
~,y LLY P in order to respond mainly to the frequency of
oscillation f of the jet of gas in the ultrasound signal
as modulated by the oscillations of said ~ et of gas .
This angular offset serves to distinguish the freguency f
25 from the freguency 2f in the modulated ultrasound signal.
If the angular offset exceeds 2, then there is a risk of
the ultrasound signals being multiply reflected in the
main channel 26, thereby degrading the guality of the
signal, and in particular reducing its signal/noise
30 ratio. In the event that it is desired to improve the
sensitivity of the fluidic oscillator over its usual
range of flow ra~es ( e. g. 100 l/h to 6000 lJh ), then, in
order to enhance detection of the freguency 2f, it is
necessary to place the ultrasound transducers very
35 accurately in the longitudinal plane of ~y ~ LLY P and it
is also preferable to emit the ultrasound signal from the
upstream end towards the downstream end.
, . . , ,, . . . . .. , .. ,, ... , .. , ,, _ _ _ _ _ _ _ _ _ _ _ .
~ 2192583
.
17
As shown in Figures 1 and 2, the ultrasound
transducers 52 and 54 are disposed facing each other in
different transverse planes. The term "transverse plane"
is used herein to designate a plane perrPnfl~ rlll Al- to the
longitudinal plane o symmetry P and to the gas f low
dlrection. If the ultrasound tr~ncfl~lrP~s æe placed in
the same transverse plane, as they are in the prior art,
then they are not suitable for measuring a volume-related
(luantity of gas at low flow rates since the emitted
ultrasound signals cannot pic~ up in~ormation rnnrP~n;n~
the speed of f 10W of the gas .
The upstream trAnc~lrpr 52 is situated upstream from
the Aflm~ c~ nn opening 26b, and more precisely upstream
from the empty space 20. As shown in Figures 1 and 2,
the upstream trAncfl~lcp~ 52 is reoeived in the upstream
end wall 22 and is thus protected from the gas flow. The
downstream trAnc;lllrPr 54 is secured to the obstacle 36
and, more precisely, it is disposed in t~e main cavity 37
of said obstacle.
As shown in the variant embodiment of Figure 4, the
upstream trr~sducer 52 is disposed in the middle portion
of the upstream end wall 21, i.e. in its portion closest
to the main channel 26, whereas the downstream tr~ncfl~
54 is secured to the obstacle 36 as described above.
In the embodiment described with reference to
Figures 1 and 2, the upstream and downstream tr~ncfll~rP~s
52 and 54 are situated at the same height relative to the
height _ of the channel 26 of the fluidic oscillator. In
a variant of the invention, the upstream and downstream
ultrasound tr;lncfll-rPrc may also be situated at di~:Eerent
heights relati~e to the height h of the channel 26 of the
fluidic oscilIator, but they must always face each other.
For e~ample, as shown in Figure 5, the difference in
height between t~le upstream and do~Tnstrer m ultrasound
tr~ncfl~lrP~s may be substantially e(aual to h.
In another ~rariant of the invention, as shown in
Figure 6, the upstream and downstream ultrasound
2i92583
.
18
trAn~ rPrs 52 and 54 are likewise situated at different
heights, but the upstream tr~nC~ ~ 52 is mounted at the
bottom of a recess 53 formed in the bottom wall 30 of the
fluidic oscillator, beneath the empty space 20. The
5 downstream tr~nQ~ P~ 54 is mounted in the top of a
recess 54 mounted in the top wall 28 of the fluidic
oscillator substantially over the obstacle 36. The
upstream and downstream tr~nC:~1llr~rs are disposed facing
each other.
In yet anotller variant as shown in Figure 7, the
upstream and downstream ultrasound transducers 52 ana 54
are situated at the same height and they no lon~er face
each other. The tr~nc~ rpr~s are mounted in respective
recesses 53 and 55 both of which are formed in the top
wall 28 of the fluidic oscillator The downstream
tr~nQ~9~ rf~ 54 is situated substantially over the obstacle
36 and the recess 55 in which it is installed does not
extend as far as the channel 26 in order:~ to avoid
disturbing the formation of the ~et of gas. In addition,
the downstream transducer 54 must be based downstream
from the channel 26 so that the ultrasound signals can be
modulated sufficiently by the oscillations of the ~et of
gas. The path followed by the ultrasound signals in the
longitudinal plane of ::;y tLy P i5 thus a V-shaped path.
The variant shown in Figure 8 also serves to obtain
a V-shaped path for the ultrasound signals in the
longitudinal plane of :,y Lly P, but with the ultrasound
tr~nQ~ s being disposed at different heights. The
upstream transducer 52 is mounted in a recess formed in
the end wall 22 so as to faoe the obstade 36. The
downstream t~ilnqr~ ~ is still installed in the same
manner as that described with reference to Figure 7.
It should be observed that by placing the downstream
tr~nQ~ ~ 54 above or below the obstacle 36, the
oscillation of the ~et of gas is disturbed less than when
the transducer is placed in the central cavity of said
~1925~3
.
19
obstacle, thereby improving the 5~uality of the ultrasound
signal modulated by the oscillations of said ~ et of gas .
It would also be possible to tilt the upstream
tr~nq~ -Pr 52 to~ards the bottom wall 30 of the fluidic
5 oseillator.
A seeond ~ml~s~ L of the invention is shown in
part in Figure 9, and referenees to the varigus Pl Ls
of this figure are preceded by the digit 2. The fluidic
oscillator 201 is said to be "in line" since it has a gas
feed 210 and a gas outlet 212 which are both in aliç~nment
in the longitudinal plane of symmetry P, unlike the
embodiment shown in Figures 1 to 7 where the --,v~ t of
the flow of gas is round a loop. The gas feed is
connected to a passage 214 which has its downstream end
opening out into a first chamber 216 that is in ~ t
with said passage on the plane P. The first chamber is
of a shape that f lares in a downstream direction until it
comes level with a transverse plane P2 ~hat is
perp~n~ r to the plane P, after whieh it tapers so as
to communicate with an upstream end 218a of a main
ehannel 218 having the same eharacteristies as the main
ehannel 2 6 shown in Figures 1 to 7 . The f irst ehamber
also ~n~ lcll~q a strP;~ml ;n~ element 220 situated
substanti~Llly in the middle the~eof and in ~ t on
the plane P. This element has a reeess 222 that faees
downstream and that reeeives an "upstream" ultrasound
tr~nc~l17-~r 224, thereby proteeting it from the flow. The
strP~ml; n~rl element 220 may also serve to ealm the gas
flow.
The main channel 218 is aligned on the longitudinal
plane of :~y ~ L~ y P and opens out into a second chamber
26 constituting an oscillatlon ehamber having the same
characteristics as the chamber 32 ~i~ql-r~ h~d above with
reference to Figures 1 to 8. This oscillation chamber
inel~ q an obstacle 228 identical to the obstacle 36
shown in Figures 1 to 8. The obstacle 228 has a central
cavity 230 situated facing the downstream open end 218b
.. _ _ . ,, ... , , . _ _ . . .
~ 2ig2s83
.
- 20
of the main channel 218 and it also has two secondary
eavities 231 ana 232 located on either side of said
eentral eavity 230. A seeond ultrasound tr~nC~ rPr 234
is reeeived in the eentral eavity 230 so that the two
ultrasound tr~nq~ rPrs are su-hstantially in ~ on
the longitudinal plane of ~y tL y P .
There follo~s a description with referenee to
Figures 10 to 16 of the metbod of measuring a volume-
related quantity of a gas such as the volume of gas that
10 flows ~hrough the fluidic oscillator as described above
with reference to Figures 1 to 3.
By way oi' example, the range of gas flow rates to be
measured extends from 5 l/h to 6000 l/h (a domestic gas
meter ) .
An electronics unit 60 is shown diagrammatically in
Figure 10 and it serves firstly to feed the various
functional blocks ~PcrrthP~ below with electricity and
secondly to control the method of measur-ing gas volume.
The electronics ~mit 60 comprises a microcontroller 62
connected to an electricity power supply 64, e. g . a
battery, and to a crystal clock 66 whose freguency is
10 MHz, for example, and which is also powered ~rom the
power supply 64. The microcontroller 62 is also
connected to an F~m; C51 rn block 68 and to a reception
block 70, each of which i5 powered by the power supply.
Each of these blocks comprises, for e~ample, an
operational iqmrlifiPr and a .,ullv~:LLer, spPrif;r;~lly a
digital-to-analog ~:u--v~r~er for the ~miCsirn block 68 and
an analog-to-digital ~_:UIIVt:L l,~' for the reception block
70. The eleetronies unit 60 also ;nrll~9Pc a switehing
cireuit 72 that is powered by the power supply 64 and
that is eonnected firstly to the pm1cc;rn and reeeption
blocks 68 and 70, and seeondly to the two ultrasound
tr~nc~ r~rs 52 and 54.
When the oseillations of the jet of gas in the
oseillation chamber 32 are too weak for it to be possible
to deteet the fregueney thereof, i.e. when the flow rate
~ 2192583
21
of the gas is below a transitioII value which i5 egual to
100 l/h, for example, then the ultrasound tr~nC(ll~r~rs 52
and 54 are used to measure the flow rate and thus the
volume of the gas in the following manner ( low flow rate
conditions ):
the upstream transducer 52 emits an ultrasound
signal towards the downstream tr;~nc~tlrrr 54;
the downstream transducer 54 receives said
ultrasound signal whose speed of propagation c is
modified by the speed of the flow of gas vg (c+vg);
a first value is det~rm~ nP~ for a magnitude that is
characteristic of the propagation speed of the received
ultrasound signal, e. g . its propagation time;
the emitter and receiver functions of the ultrasound
tr~nc~-r.Prc 52 and 54 are interchanged;
the downstream tr~nc~ r~r 54 now emits an ultrasound
signal towards the upstream transducer 52;
the upstream tr~nc~llr~r 52 receives-; this ultrasound
signal that propagates at a speed ( c-vg );
. a second value is detPrm~ nP~ for the propagation
time of the ultrasound signal; and
a mea:,uL. t of the ~as flow rate is deduced
therefrom which, by integration, serves to provide a
mea~ ul~ t of the total volume of gas that has passed
through the fluidic oscillator.
With reference to ~igure 10, a measurement is
triggered as follows: the seguencer ( not shown ) of the
microcontroller activates the ~m~ c5~ nn block 68 to send
an electrical signal to the upstream transducer 52, and
also activates t~le power supply 64 to set the switching
circuit 72 so that the ~m~ c5~ nn block 68 is connected to
the upstream transducer 52 and so that the reception
block 70 is connected to the downstream transducer 54.
The electrical signal excites the upstream transducer 52
which emits an ultrasound signal into the gas in the flow
direction of the gas at a specific instant which is
determined by the clock 66. The signal travels through
. _ . _ . . . .. . .
21g2583
22
the gas at speed c while the gas itself travels at a
speed vg. After a time lapse tl measured by the clock
66, the downstream tr;tnC~lt~rpr 54 receives the ultrasound
signal which appears to have been propagating at the
speed c+vg.
To measure the propayation time tl of the ultrasound
signal, rPfPrPnre may be made to the method described in
European patent application No. 0 426 309. That method
consists successively: in gen-erating and transmitting an
ultrasound signa~ made up of a plurality of cycles or
pulses, ;nr~l(lin~ a phase change within the signal; in
receiving the ultrasound signal; and in detecting the
phase change wit~lin the received signal, such that the
instant that corresponds to said phase change enables the
propagation time tl to be detPrm; nP~ . Everything
reguired for implementing that method o~ measuring
propagation time is described in European patent
application No. 0 426 309, and is theref~tre not described
again herein.
Thereafter, the se~uencer o~ the microcontroller 62
causes the switc~ling circuit 72 ~o swap connections so
that the pm; c5; nn block 68 is now cu~ e~ d to the
downstream tr~tnct~lllrpr 54 while the reception block 70 is
connected to the upstream trrtnctrltlrpr 52. A second
ultrasound sisnal is emitted in li~e manner by the
downstream transducer 54 towards the upstream transducer
52 so a to travel in the opposite direction to the gas
flow direction, and the clock 66 determines the time t2
reguired for said ultrasound signal to propagate, in the
manner described above and in European patent application
No. 0 426 309
Given that the propagation times tl and t2 can be
expressed by the following relat;nncth;r~:
tl = I./ ( c-vg ) t2 = ~/ ( c+vg )
the arithmetic and logic unit (not shown) of the
microcontroller 62 calculates the velocity of the gas vg
by applying the following relationship:
.. .. . . .. ... .. . _ ...
~192~3
23
L
vg [ ]
2 t2 tl
from which a gas flow rate measurement Qm is deduced
5 where:
L
Qm = S - t-- - --]
2 t2 tl
5 being the internal section o~ the channel 26.
The microcontroller 62 compares each measured flow
rate value with the predet~m1n~o~ transition flow rate
value as stored in its memory in order to determine
whether the next mea- uL~ t of flow rate should be
performed using the above method or by detecting the
15 frequency of oscillation of the ~et of gas in the
oscillation chamber 32 of the fluidic oscillation ( high
ilow rate conditions ) . If the measured flow rate value
is below the transition flow rate, then the flow rate of
the gas is measured again after a prede~rml nrrl time
20 interval using the above method.
It should be observed that the fluidic oscillator of
the present invention, referred to as a "combination"
fluidic oscillator, makes it possible to use propagation
time meaYuL Ls on an ultrasound signal in the flow of --
25 gas at flow rate values that are small enough to avoidintroducing errors into the flow rate measurements due to
instabilities in the flow of gas, which instabilities are
generated by the transition from laminar flow to
turbulent flow. The low flow rate mea~u" 1_-. thus have
30 the advantage of being accurate and repeatable. In
addition, given that this technique is used to cover a
relatively narrow range of flow rates, it is pr,Rq1h1~ to
make do with narrow band ultrasound tr~nc~ rp~R that
typically have a resonance frequency of 40 kHz, rather
35 than using transducers that are more sophisticated, more
expensive, and that are resonant at lO0 kHz.
If the measured flow rate has a value that is
greater than the transition flow rate value, then the
21g258~
.
24
oscillations of the jet of gas are strong enough for the
freguency thereof to be detected (high flow rate
conditions). Under such Cil~; ~culCeS, the gequencer of
the microcontroller 62 controls the switching circuit 72
5 so that the pm~ qc1 rn block 68 is connected to the
downstream tr~ncd~r~r 54 and the reception block 70 is
connected to the u~ l L~am tr;~,nRdl r-F~r 52. The sequencer
also causes a switch 74 to operate so that the slgnal
coming from the upstream tr~ncfl~lrPr 52 is now treated by
10 the electronics unit 100 that can be seen on the right of
Figure 10. This unit is described below in greater
detail with reference to Figure ll.
Under high flow rate conditions, the microcontroller
62 causes the pm; qc~ nn block 68 to generate a permanent
15 electrical signal for exciting the downstream ultrasound
tr~ncfl 7r~r 54, e.g. a squarewave signal, at a frequency
~u so that the downstream tr~ncfl ~rPr continuously emits
an ultrasound signal of frequency fu tow~rds the upstream
transducer 54 in a direction that is ~ nrl ~ n~ at about
20 1. 5 to the longitudinal plane o~ :,y ~Ly P. The
ultrasound signa~ received by the upstream tr~nC~llrpr is
a signal of irequency fu modulated by the frequencies f
and 2f that are characteristic of the oscillation
rhPnl -'nn O~ the ~et of gas. By way of example, the
25 frequency fu may be equal to 40 kHz and the amplitude of
the electrical excitation signal may be 20 mV.
The Applicant has been able to observe that by
emitting the ultrasound signal against the flow of gas,
it is pr~Rs~hlP to reduce rl~nR~flPrably the influence of
30 the llydLudy,lamic pregsure of the jet, thereby reducing
the energy of the signal that is received in respect of
the frequency 2f. By way of example, a difference of
lO dB has been observed on the amplitude of the signal
received at the frequency 2f, and that suffices to make
35 it possible to distinguish the frequency f from the
frequency 2f in the modulated signal while using
electronic equipment that is simple, cheap, and consumes
2192583
little energy. The Appl~cant has also observed that by
emitting the ultrasound signal ayainst the flow of gas,
the modulated ul trasound signal presents periodicity in
time, thereby facilitating detection of the frequency f.
Thus, when the upstream LLC11:~dUC~CL 52 receives an
ultrasound signal modulated by the oscillations of the
~et of gas, this signal is initially amplified by a low
noise analog ~mrl; f ; ~r 102 . The analog -~rl; f ; l~r 102 is
a non-inverting :qmrl;f;l~r l~C~nP-l for coupling to the
10 electronic circuit that performs mea:iuL~ ~i under low
flow rate conditions, and it is constituted by an
operational ~mrl; f; er A1 whose non-inverting input is
connected firstly to the modulated signal as received by
the upstream tr~n~ rc~r 52 and secondly to ground via a
15 resistor Rl. The $nverting input of this operational
amplifier A1 is connected firstly to ground via a
resistor RZ and secondly to the output Bo of the amplifier
via another resistor R3. The modulated ~nd amplified
ultrasound signal then has the appearance shown in Figure
20 12. ::
A conventional halfwave rectlfier circuit 104 is
shown in Figure ll and comprises a resistor R~ connected
between the output Bo of the amplifier A1 and the
inverting input B1 of an operational ~m?1~f;~r Az whose
25 non-inverting input is connected to ground. The
inverting input of the ~mrl~f;~r A2 is connected to the
output B2 of said ;~mrl; f; ~r via two parallel-connected
branches: a first branch is constituted by a resistor R5
in series with a diode D1 that is reverse-connectedi and
30 a second branch which is constituted by a diode D2. In
~,c -lv~lLlonal manner, when the difference VB1_VB2 is greater
than the threshold voltage of the diode Dl, then it
conducts giving VB3 = ( R~/R~ )VB2 . Conversely, when the
value VB1_VBZ drops below the threshold voltage of the
35 diode D1, then the diode D2 becomes conductive and VB3 = ,
the rectified signal having the appearance given in
Figure 13.
. , . . . .. . . . . . .. _ , . .. . _ _ . . _ _ . . _ . .. . .
~1~2583
26
In order to retain only the oscillation frequency f
of the ~et of gas, the rectified signal is then filtered
by electronics block 106 which acts as a (second order)
lowpass filter. As shown in Figure 11, the block 106 has
5 two resistors R6 and R8, and a capacitor C1 f orming a T -
filter which is subjected to negative feedback by a
resistor R~ and a capacitor C2 together with an
operational ~mrl; f i Pr A3 whose non-inverting input is
connected to ground. The filtered signal obtained at Bs
10 has the appearance shown in Figure 14.
This signal is then injected into an ~mrl;f;cation
electronics block 108 that comprises two stages: a first
stage 109 that acts as a bandpass ~mrl if ~Pr having a gain
of 50, for e~ample, and having a cutoff requency lying
in the range 0 . 5 Hz to 50 Hz; and a seaond stage 110 that
acts as a lowpass ;~mrlif;Pr having a gain equal to 5, for
example, and a cutoff freguency e~ual to 50 Hz.
The first stage comprises a resistc~r R9 and a
capacitor C3 connected in series between the output Bs of
block 106 and the inverting input of an operational
~rl~l;f;Pr A~. The non-inverting input of the ~mrl;f;Pr A~
ls connected to ground, and its inverting input is
connected to its outputs B6 via a capacitor C6 and a
resistor RlC connected in parallel.
The second stage 110, downstream from B6, comprise~3
a resistor R1l connected to the inverting input of an
operational amplifier As~ which input is also connected
to the output B~ of the ~mrl i f i Pr via a resistor Rl2 and a
capacitor Cs ccnnected in parallel. The non-inverting
input of the ;:~mrlif;Pr As is connected to ground.
The electrcnics block 108 serves to shift the signal
from the filter 106 so as to place it on either side of
zero, and to amplify said signal. The signal as
amplified in this way that appeared at B~ is in~ected
into the following block 112 which transforms it into a
pulse signal as shown in Figure 15.
.... . .... ....... . . .. . ..... .. . . . _ . .
2~2~83
27
The electronics block 112 comprises an operational
amplifier A6 whose non-inverting lnput is connected to
the output B~, and whose inverting input is connected
firstly to the output B8 of the ~mrl ~ fiP~ A6 via a
resistor Rl~, and secondly to the output of a ~ v~-tlonal
follower circuit that comprises an operational ~mrl~f1pr
A~. Because of the negative feedback from the follower
circuit, the amp~ifier A6 makes it poc~lhlP to amplify
small amplitude signals more-than large amplitude
signals.
This block also includes a resistor Rls connected to
the output B3 of the amplifier A6 and to a point B9, and
it also includes two diodes D3 and D~ mounted head to tail
between the point Bg and a point B1o. The point Blo is
connected firstly to ground via a capadtor C6 and
secondly to the non-inverting input of the follower
circuit A~ and to the inverting input of a further
operational Amrl; f;P~ A3. The output of ~operational
amplifier A8 (point Bl2) is looped back to its non-
inverting input via a resistor R1,. The non-inverting
input of this ~mrl~ f~P~ is also connected to the diodes
D3 and D~ via a resistor R16. When the amplitude of the
voltage VB9_V310 increases to exceed the threshold of diode
D~, then the diode conducts and the value of the voltage
signal at point B9 minus the voltage drop across diode D;
is stored in capacitor C6. The differential amplifier A8
then compares the value of the voltage at point B11 as
given by:
V R + V R
V311 B9 17 312 16
Rl7 + Rl6
with the value of the voltage on capacitor C6, and it
produces a high value signal when the voltage at point B9
is greater than the voltage on the capacitor C6.
Once a peak has been reached and the amplitude of
the signal (lPrP~qPq~ the difference between the value of
the signal at point B9 and the value of the signal stored
- ~ ' 21g2~83
28
by the capacitor C6 drops below the threshold of the
diode D~ so the diode D4 becomes non-conducting. The
value of the sigl~al stored on capacitor C6 then remains
unchanged. When the amplitude of the signal at point B5
drops below the value of the signal stored on the
capacitor C6, then the amplifier A8 provides a low level
signal showing that a peak has occurred. When the
amplitude of the signal drops below the value of the
signal stored on the capacitor C6 by an amount that
corresponds to the threshold of tl~e diode D3 plus the
voltage at B1o, then the diode D3 starts to conduct and
the value of the signal as stored on the capacitor C6
drops to the vallle of the signal at point Bg minus the
voltage drop across the diode D3. When a negative peak
is reached and passed, the diode D3 will again become
non-conducting, and the ~mrl;fiP~ A3 will indicate a
change in state once the signal at point Bll has increased
to above the value of the signal stored ~n the capacitor
C6 ~
In Figure 16, curve 150 shows how the voltage of the
first signal at point Bg varies, and curve 151 shows how
the voltage across capacitor C6 varies. Initially, the
capacitor voltage 151 is e~Iual to the signal 150 minus
the value Vd that corresponds to the voltage drop across
the diode D" so the ~mrl if i~ A8 provides a high level
signal. When a peak is reached at time T0, and while the
voltage of the signal 150 has dropped under the threshold
of the diode D~, the voltage on the capacitor 151 remains
unchanging. At time Tl the volta~e of the signal 150
drops below the ~oltage stored on the capacitor 150 so
the output oi ~the ;~ 7l i f; c.~ A8 provides a low level
signal. At time T2, the difference between the voltage
of the first signal 150 and the voltage stored on the
capacitor 151 becomes greater than the threshold voltage
for the diode D~ so the capacitor voltage again tracks
the voltage of t~le iirst signal. The electronic circuit
corresponding to block 112 is thus a peak detector. The
:
2~2583
29
amplifier A8 of l igure lr is a ~ CLuL with hysteresis
that compares the values ûf the two voltages 150 and 151
shown in Figure 16. Thus, when the value of the voltage
at point B.., i.e. V31l, the voltage on the non-inverting
input of the amplifier A8, is greater than the value of
the voltage at point Blo~ i.e. VD10~ the voltage on the
inverting input of the amplifier A8, then the amplifier
provides a constant output voltaye equal to +Vcc where
Vcc is the power supply voltage of said amplifier, and
the voltage at B11 becomes:
VccR16 + VDgR17
VD11
Rl6 + Rl7
Conversely, and as shown at instant Tl in Figure 16,
when VD11 is less than VD10~ then the output voltage of the
cmplifier A~ i5 egual to -Vcc so the voltage at B~,
becomes:
-VccR16 + VDgR17
VD11 =
R1~ + R17
As a result the output from the block 112 is in the form
o E a pulse signal in which each pulse represents a unlt
volume of gas being swept by the j et of gas during a
single oscillation ( Figure 15 ) . --
An electronic counter (Figure 10), e.g. a 16-bit
counter, then serves to count the total number of pulses,
thereby Pni~hl ~n~ the mi~ilu~;ullLLuller to deliver the
volume of gas that has flowed through the fluidic
oscillator .
It should be observed that if the transducers 52 and
54 are accurately aligned on the longitudinal plane of
-y ~ tly P in order to enhance received signal energy at
the frequency 2f, then the above-described electronic
circuit 100 needs to be adapted sp~r~ f ~ ~~P l l y to detect
3 5 that f requency .
When the oscillations of the ~ et of gas become too
weak for the frequency thereof to be detectable, i.e.
when the flow rate of the gas becomes less than the
2192~83
above-mentioned transition flow rate, provision is made
to use the ultrasound trAncrl~ rPrs 52 and 54 to measure
the ~lu~aya~lon time of ultrasound signals in the flow of
gas as r9PC~r~ hPfl above ( low f low rate conditions ) . In
5 order to decide ~hen to use low flow rate conditions or
high flow rate conditions, it is possible, for example,
to measure the time interval between two successive
pulses and to compare the time interval measured in this
way with a prede~Prm; nP~ value that corresponds to the
10 transition flow rate. If the measured time interval
exceeds the predetPrm~ nPCl value, then the ultrasound
tr~nR-lllrPrs are used alternately as emitter and receiver.
It is also pOcR~ hl P to provide for an overlap range,
e. g. extending from 100 l/h to 150 l/h, over which both
15 operating conditions of the combination fluidic
oscillator may be used. Thus, if the fluidic oscillator
is operating under low flow rate conditions, it may
continue to measure flow rate in that way until the high
value of the overlap range i5 reached, at which point
20 measurement switches over to high flow rate conditions.
Similarly, when the fluidic oscil 1 ator is operating under
high flow rate conditions, then it is nPrPgc~ry for the
flow rate to drop below the lower value of the overlap
range be~ore the fluidic f~R~ r switches over to
25 measuring under low flow rate conditions.
The advantage of having an overlap range is to make
it unlikely that it will be necessary to switch
mea:iuL t conditions back again to the preceding
conditions immediately after a switchover has taken
30 place.
The combination fluidic oscillator o the present
invention can be adapted to various flow rate ranges, and
is capable, in particular, of covering flow rates at
greater than 6000 l/h. Figure 17 is a calibration curve
35 for the combination fluidic oscillator operating over
flow rates extending within a range of about lO l/h tû
about 7000 l/h. The curve shows the relative error
~ 2192583
31
applicable to measul~ ~ throughout the above range. It
can thus be seen that the combination fluidic oscillator
is entirely suitable for mea:,u~ ~ t purposes throughout
this large range of flow rates.
It will be observed that the invention has the
a~v~cy~: of also being applicable to other types of
1uidic oscillators, e. g . that described in patent
application GB-A-2 120 384, which is based on the Coanda
effect. It is quite possible to envisage using the
combination fluidic oscillator o~ the invention both to
cover a range of 1uid 10w rates over which oscillations
o the jet of 1uid are strong enough for the frequency
thereo to be detected, and also to use the low flow rate
operating conditions as P~rl~;n~d above solely for
det~rm;n;n~ a leakage flow rate. For example, a 1uidic
oscillator of the invention could be used as a commercial
gas meter ~flow rate range 0.25 m3~h to 40 m3/h) or as an
industrial gas meter (flow range 1 m3/h to 160 m3/h) that
is capable o measuring a leakage rate.
It is also possible to increase meter sensitivity by
positioning the ultrasound transducers accurately in the
longitudinal plane o symmetry of the fluidic oscillator.
