The present invention relates to a method of improving the accuracy of a measuring result obtainable by means of a flow meter of the type defining therein a flow passage for a medium flow to be measured, which flow meter comprises a sensor for sensing a varying condition of the medium flowing through the passage, means for generating ultrasonic signals and for directing such signals in opposite directions along at least part of the flow passage, means for receiving sig¬ nals having passed along said at least part of the flow passage, and electronic circuit means for calculating or determining a measured flow rate value of the medium based on the difference in velocity of transmitting the ultrasonic signals in said opposite directions.

A flow meter of this type is disclosed for example in German patents Nos. 29 24 561, 29 34 031, and 30 39 710. As described in these patent specifications, various factors, such as varying acoustic loads on the ultrasound generating disc of piezoelectrical material at varying temperatures, undesired disturbing echoes of ultrasonic signals transmitted to a receiving transducer, and varying, uneven conditions of the flow into and out from the flow passage may adversely influence the accuracy of the measuring results to an unsa¬ tisfactory degree.

SU 792143 discloses a testing device for testing the accuracy of flow meters. This known testing device may simulate a flow rate, and the flow rate measured by the flow meter being tested is compared to the flow rate simulated. This known testing device renders it possible to draw graphs showing the absolute error of measurements in dependency of the tempera- ture and pressure of the medium flow measured.

The accuracy of measuring results obtainable by means of a flow meter of the type described above may be improved by the method according to the invention, which comprises providing a medium flow according to a predetermined flow rate course

through the flow passage, varying said condition of the medium in the flow passage, calculating or determining by the electronic circuit means measured flow rate values of the medium in the flow passage at each of a plurality of diffe- rent values of said condition, determining for each of said condition values a first compensating value being the diffe¬ rence between the corresponding flow rate value according to said predetermined course and the measured flow rate value actually calculated or determined, and storing such compen- sating values.

In the method according to the invention the difference between the measured flow rate value calculated and the corresponding actual flow rate value, which is known, is determined at a plurality of different values of said condi- tion of the medium. These differences or first compensating values determined represent the errors caused by various factors, such as the factors mentioned above, and they may be used for compensating calculated measured values of unknown flow rates.

The predetermined flow rate course may represent any known flow rate varying in respect to time. In the preferred embo¬ diment, however, the medium flow is passed through the flow passage at a predetermined constant flow rate, which means that the "predetermined flow rate course" is a predetermined flow rate which does not vary with respect to time. This constant flow rate may have any desired value. However, for the sake of simplicity it is preferred to determine the first compensating values at a predetermined flow rate value which is zero. This means that the medium, which may be liquid or gas, is enclosed within the flow passage, while the diffe¬ rences or first compensating values are being determined.

It is understood that the method according to the present invention relates to any condition influencing the measure¬ ments of the flow meter, and that the flow meter could be used for measuring the flow of any medium. In most practical

cases, however, the said condition is the temperature and/or the pressure of the flowing medium, which is normally a liquid, such as water, with varying temperature.

When the first compensating values are to be determined, the medium within the flow passage may be heated to an elevated temperature, such as 90°C and thereafter allowed to cool. The measured flow rate values at said plurality of different condition values may then be calculated or determined by the electronic circuit means continuously or at short time inter- vals during such cooling, and each of the measured flow rate values is then compared with the actual flow rate value, which is preferably zero. In this manner it is possible to determine a compensating curve or a great number of first compensating values associated with various temperatures or pressures.

The conditions at which the first compensating values are determined may be sensed or measured by a condition sensor, such as a pressure or temperature sensor. Therefore, it is important that the condition sensor is correctly calibrated or adjusted. The condition sensor is preferably calibrated at least at two different condition values prior to calculating or determining measured flow rate values of the medium in the flow passage at different values of said condition. When the condition of the medium in the flow passage is the tempera- ture of the medium, this may for example be done when the medium within the flow passage is being heated. As an example, the accuracy of the temperature sensor may be checked at two or more temperatures, for example at 40°C and 80°C.

Prior to determining the first compensating values the flow meter is preferably exposed to various extreme conditions simulating use of the flow meter. Thus, the medium within the flow passage may be exposed to widely varying temperature and/or pressure conditions for an extended period of time, such as a few or several days, whereby the flow meter may be

"artificially aged" before the first compensating values are determined.

When first compensating values have been determined for the individual flow meter as described above and such compen- sating values or such compensating curve has been stored, for example in an electronic memory or storing circuit of the electronic circuit means, the flow meter may be used for measuring the flow rate value of an unknown medium flow through the flow passage. Such method of measuring may then include calculating or determining a measured flow rate value of an unknown medium flow through the flow passage based on the difference in velocity of transmission of the ultrasonic signals, measuring or sensing the actual condition of the medium in the flow passage by means of the condition sensor, and compensating the measured flow rate value of the unknown medium flow by the stored first compensating value associated with the condition value actually sensed or measured. This means that each flow meter may be artificially aged and subsequently tested so as to determine and store a set of individual first compensating values in the electronic cir¬ cuit means of each individual flow meter before it is dis¬ patched from the factory. This means that the electronic circuit means of each flow meter contains a set of tempera¬ tures and/or pressures and corresponding individually deter- mined first correction or compensating values of the measured flow rate value calculated by the electronic circuit means.

As explained above the determination of the flow rate value of the medium is based on the difference in velocities of transmitting ultrasonic signals through the medium. However, such velocity of transmission may be dependent on the varying condition of the medium. Therefore, the measured flow rate value may preferably be further compensated by a second compensating value compensating for such dependency.

Each ultrasonic signal may be a pulse train, the flow rate value being determined by measuring the difference in time of

detection of a pair of corresponding pulses in said trains, identical pulse trains being simultaneously emitted from oppositely arranged signal generating means.

In order to increase the accuracy of measurement the measured flow rate value is preferably calculated or determined on the basis of a predetermined plurality of successive measurements of said difference in time of detection. Furthermore, the measured flow rate value may be calculated or determined on the basis of the sum of said plurality of successive measure- ments. The number of successive measurements may be chosen in dependency of the flow rate being measured. Thus, the number of successive measurements may be reduced when the flow rate measured is increased. An initial test measurement of the flow rate may be used for determining the predetermined plurality of successive measurements of the difference in time of detection.

The invention also relates to a flow meter for measuring a medium flow through a flow passage defined therein and com¬ prising a sensor for sensing the temperature and/or pressure condition(s) of the medium in the flow passage, ultrasound generating means for generating ultrasonic signals and for directing such signals in opposite directions along at least part of the flow passage, means for receiving ultrasonic signals having passed along said at least part of the flow passage, and electronic circuit means for calculating or determining a measured flow rate value of the medium in the flow passage based on the difference in the velocity of transmitting the ultrasonic signals in said opposite direc¬ tions, the circuit means further comprising a memory for storing a plurality of different temperature and/or pressure values and associated compensating values determined for example by a method as described above, and means for compen¬ sating the measured flow rate by the compensating value associated with the condition value sensed by the sensor.

It is understood that each such flow meter contains an indi¬ vidual compensating curve or individual compensating values in its memory so that the measuring value being registered by the flow meter or displayed to the user may be the compen- sated measured value at the respective condition value, such as temperature and/or pressure.

The fluid flow, such as a flow of hot water, through the flow passage should preferably be as uniform as possible over the cross-sectional area and along the length of the flow pas- sage. In order to improve such uniformity of flow the cross- sectional area of the flow passage may be substantially constant, the cross-sectional area of an inlet end portion of the flow passage decreasing slightly in downstream direction towards said constant value. Such inlet end portion, which may define a f usto-conical inner surface part, may function as a kind of funnel. Alternatively or additionally, the cross-sectional area of an outlet end portion of the flow passage may increase slightly from the constant area in the downstream direction.

According to the invention at least one of the ultrasound generating and/or receiving means may comprise an electronic transducer including a housing made from metal, such as stainless steel, and defining an annular abutment surface, a vibrateable disc member made from a piezoelectrical material, and means for mounting the disc member in said housing so that a peripheral rim portion of the disc member and the annular abutment surface are positioned in overlapping rela¬ tionship. The disc member may be mounted in any suitable manner. As an example, the mounting means may comprise means for biassing the disc member towards the annular abutment surface.

The biassing force used is preferably, but not necessarily, sufficient to maintain the piezoelectrical disc member in the position in which it is mounted in relation to the housing.

The area of the abutment surface overlapped by the peripheral rim portion of the disc member preferably exceeds half the area of the disc member.

Such relatively large area of overlap ensures that the speci- fic pressure to which the peripheral part of the disc member is exposed may be kept within acceptable limits.

Alternatively or additionally, the mounting means may com¬ prise an adhesive by means of which a peripheral rim portion of the disc member is sealed to the annular abutment surface, whereby it is ensured that the disc member is substantially immoveable in relation to the abutment surface of the hous¬ ing.

The disc member may be received in an insert member of a non- metallic material, the insert member comprising a plane annular wall part, which its sandwiched between the annular abutment surface and an adjacent side surface of the disc member, and a peripheral, cylindrical wall part having its outer surface engaging with an inner cylindrical surface part of the housing and having its inner surface engaging with a peripheral rim surface of the disc member. The peripheral surface of the piezoelectrical disc member may also be in contact with an annular flange part of the insert member, and such contact may vary from one transducer to another and be dependent on temperature because of the different coeffi- cients of expansion of the materials from which the piezo¬ electrical disc member and the insert member, respectively, are made.

The biassing means may comprise a sheet member of metal, such as stainless steel, having its inner side surface fixed to the adjacent side surface of the disc member, for example by an adhesive or a glue, and having its peripheral rim portion fixed to said housing, for example by welding. Such sheet member may maintain the piezoelectrical disc member in the same position in relation to the abutment surface and at the

same time provide the necessary bias against such surface. Such bias may be obtained by prestressing the metallic sheet member axially before welding the diaphragm-like sheet member to the housing.

The peripheral rim portion of the metallic sheet member is preferably fixed to the housing so as to completely seal the inner of the housing containing the disc member. When the disc member is fastened not only to the metallic sheet mem¬ ber, but also to the annular abutment surface, for example by an adhesive, the sealed transducer may be made capable to resist or withstand substantial outer pressure variations. When the disc member is fastened to the annular abutment sur¬ face the prestressing of the metallic heat member is less necessary than when the disc member is not fastened to the annular abutment surface.

The ultrasound generating and receiving means may comprise a pair of transducer units arranged in inlet and outlet cham¬ bers, respectively, said chambers being interconnected by the flow passage having an axial length being several times larger than the distance from the transducers of the trans¬ ducer unit and the walls of the respective inlet or outlet chamber. It may thereby be prevented that ultrasonic signals generated by one of the transducers are reflected by the chamber walls and returned to said transducer at substan- tially the same time as when signals are received from the other transducer. The ultrasound transducers are preferably of the type comprising a piezoelectrical disc for generation and detection of ultrasonic signals.

The invention will now be further described with reference to the drawings, in which

Fig. 1 diagrammatically illustrates the function of a fluid flow meter of the type including a pair of ultrasound trans¬ ducers, Fig. 2 is a graph showing the velocity of ultrasonic waves in water as a function of the temperature of the water,

Fig. 3 is a side view and partially sectional view of a first embodiment of the apparatus or flow meter according to the invention,

Fig. 4 is a perspective view of a transducer unit forming part of the apparatus shown in Fig. 3,

Fig. 5 is a side view and partially sectional view of the transducer unit shown in Fig. 4,

Fig. 6 is a fragmentary sectional view of the transducer unit according to Fig. 4 and 5 shown in an enlarged scale, Fig. 7 is a side view and partially sectional view of a second embodiment of the apparatus or flow meter according to the invention,

Fig. 8 is top plan view of the flow meter shown in Fig. 7 where a part has been cut away for illustrative purposes, Fig. 9 is a block diagram of the fluid flow meter illustrated in Figs. 1 and 3-6, and

Figs. 10 and 11 are curves showing flow rate compensating values as a function of fluid temperature and plotted for two individual fluid flow meters according to the invention.

Fig. 1 diagrammatically illustrates how a fluid flow may be measured by means of ultrasound transducers. Fig. 1 illu¬ strates a flow meter comprising a housing 10 defining inlet and outlet chambers 11 and 12, respectively, which are inter¬ connected by a flow passage 13. A pair of ultrasound trans- ducers 14a and 14b are arranged in the inlet and outlet chambers 11 and 12, respectively, opposite to the adjacent ends of the flow passage 13. Each of the ultrasound trans¬ ducers may act as an ultrasound emitter as well as an ultra¬ sound detector.

When a fluid flows from the inlet chamber 11 to the outlet chamber 12 through the flow passage 13 the fluid flow in the flow passage may be measured. When such flow measurement is performed the two transducers 14a and 14b simultaneously emit a train of ultrasonic pulses in opposite directions, which means that a train of pulses is emitted upstream as well as downstream of the fluid flowing through the flow passage 13.

After the emission of such train pulses and before the pulses reach the opposite transducer the mode of operation of both transducers is changed from an emitting mode to a detecting mode so that each of the transducers 14a and 14b may detect the pulses emitted from the opposite transducer. The ultra¬ sonic pulses emitted by the transducer 14b and travelling upstream the fluid flowing through the flow passage 13 are delayed compared to the pulses emitted by the transducer 14a and travelling downstream the fluid flow. This time delay T may be used for determining the fluid flow through the pas¬ sage 13.

When the flow passage 13 is defined by a circular cylindrical surface having a diameter d, the cross-sectional area of the flow passage is

A = (π x d ² )/4.

If the velocity of the ultrasonic pulses through the fluid is C, and the axial spacing of the transducers 14a and 14b is 1, the time delay of the ultrasonic pulses directed upstream of the fluid flow may be expressed as

T = (F x 2 x 1)/(C ² x A) ,

where F is the fluid flow through the flow passage 13. This means, that

F = T x (C ² x A)/(2 x 1) .

Consequently, if the time delay T is measured the fluid flow through the flow passage 13 may be calculated.

As the time delay i is rather small and thus difficult to measure accurately, this delay is preferably detected for each of the pulses of a train of pulses and then summed and amplified so that a resulting delay pulse ΔT of a reasonable length may be obtained. It is understood that the fluid flow

is also proportional to such resulting delay pulse. In case the flow is small the delay pulse may be measured for a train of for example 24 pulses, in case of an average flow the delay pulse may be measured for a train of for example 12 pulses, and in case of a high fluid flow the delay pulse may be measured for a train of for example 6 pulses. By measuring the delay pulse for a number of train pulses dependent on the flow value, the dynamic range of the flow meter may be increased by a factor of four.

From the above equation it is apparent that the flow F is proportional to the square of the velocity of the ultrasonic waves in the liquid. As the velocity of the ultrasonic waves is dependent on the temperature of the fluid, it is necessary to measure the temperature of the fluid flowing through the flow passage 13 so that the temperature dependency may be taken into account.

The flowing fluid being measured may be any gas or liquid, and for each individual fluid to be measured the velocity C of ultrasonic waves in such fluid may be plotted versus the temperature of the fluid. As an example, a graph showing the velocity of ultrasound waves in water plotted as a function of the water temperature is shown in Fig. 2.

Figs. 3-6 illustrate a first embodiment of the fluid flow meter according to the invention diagrammatically shown in Fig. 1. In the embodiment shown in Fig. 3 the housing 10, which is preferably made from metal, defines the inlet and outlet chambers 11 and 12 therein, and these chambers may be detachably connected to inlet and outlet conduits 15 and 16, respectively, for example by means of threaded connection means, such as threaded nuts 17. In Fig. 3, the flow passage 13 is defined by the bore of a connecting tube section 18, which is mounted in the housing 10 by means of a pair of mounting elements 19 so that the opposite free ends of the tube section 18 projects into the inlet and outlet chambers 11 and 12, respectively. The inner diameter of the tube

section 18 has a substantially constant value along the main part of the length of the tube section. However, the inner bore of the tube section may be slightly conical at the free ends of the tube section 18 so that the inner diameter is slightly increasing from said constant value towards the adjacent outer free end of the tube section. While the con¬ necting tube section 18 is preferably made from stainless steel the mounting elements 19 may be made from a plastic material, such as NORYL ^® . Each of the mounting elements 19 may comprise a sleeve-like part snugly surrounding the adja¬ cent projecting end portion of the tube section 18, and the annular outer end surface of each sleeve-like part may be chamfered or bevelled as shown at 20 in Fig. 3. Furthermore, an annular outer sealing element, such as an O-ring 21, engaging with the outer surface of the tube section 18 may be positioned in engagement with an annular abutment shoulder defined in the housing 10 and an adjacent annular end surface of one of the mounting elements 19.

An ultrasound transducer unit 14 shown in more detail in Figs. 4-6 and each including one of the transducers 14a or 14b is mounted in each of the inlet and outlet chambers 11 and 12 so that the transducers therein are arranged opposite to the adjacent open ends of the tube section 18, and the transducer units 14 are held in position within the inlet and outlet chambers 11 and 12 by means of the mounting elements 19. The flow meter further comprises a temperature sensor 22 for sensing the temperature of a fluid flowing into the flow meter via the inlet conduit 15. The temperature sensor 22 may be arranged within or adjacent to the inlet chamber 11 and is preferably forming part of the transducer unit 14 arranged therein, vide Fig. 5.

The fluid flow meter illustrated in Figs. 3-6 further com¬ prises an electronic circuitry 23 arranged in a circuitry housing 24, which is mounted on the housing 10, for example by releasable fastening means, such as one or more fastening screws 25. The electronic circuitry 23, which will be further

described below with reference to Fig. 9, is electrically connected to the transducers 14a and 14b and to the tempera¬ ture sensor 22 by means of electrical conductors 26, so that the electronic circuitry 23 may receive electrical signals from the temperature sensor 22 and from the transducers 14a and 14b and transmit signals to these transducers.

As shown in Figs. 4-6 each of the transducer units 14 com¬ prises a stiff transducer block, which is preferably made from metal, such as stainless steel. The transducer block may comprise a substantially cylindrical portion 28 having an outer peripheral channel for receiving an annular sealing ring, such as an 0-ring 29, and a semi-cylindrical part 30 extending axially therefrom. The semi-cylindrical part 30 defines a flat, substantially diametrically extending surface 31 having a transducer receiving recess 32 formed therein. The cylindrical base portion 28 of each of the transducer units 14 is received in the inlet or outlet chamber 11 or 12 so that the O-ring 29 is in sealing engagement with the adjacent inner wall part of that chamber and the flat surface 31 is arranged opposite to the adjacent free end of the tube section 18, vide Fig. 3.

An insert member 33, which is preferably made from plastic material, such as NORYL ^® , is arranged within the recess 32 and defines an outer, plane, annular abutment surface 34 and a peripheral rim or flange portion extending transversely outwardly therefrom. The insert part defining the abutment surface 34 and the peripheral flange portion 35 are preferab¬ ly rigidly supported by complementary surface parts defining the transducer receiving recess 32. A vibratable disc or plate 36, which is made from a piezoelectrical material and having metallic electrodes 37 and 38 adhered to its opposite side surfaces in a manner known per se is supported by the annular abutment surface 34. The outer diameter of the disc or plate 36 substantially corresponds to the inner diameter of the rim or flange portion of the insert member 33 so as to obtain a fit allowing an easy mounting of the disc of plate

36 in the insert member 33. The area of the outer rim portion of the disc or plate 36 supported by the plane abutment surface 34 preferably amounts to a value which is about half and even more than half the total area of the inner side surface of the disc or plate 36. The outer side surface of the disc or plate 36 or the electrode 38 covering the same is fastened to a thin plate or diaphragm 39, for example by means of an adhesive or glue. The outer annular rim portion of the diaphragm 39, which is preferably made from metal, such as stainless steel, is prestressed by exerting a pres¬ sure thereto directed axially inwardly into the recess 32 and, subsequently, the free edge portion of the diaphragm 39 is fixed to the semi-cylindrical portion 30, for example by a welding seam 40.

The plate or diaphragm 39 should have a thickness sufficient to prevent the disc or plate 36 from moving substantially in radial direction in relation to the abutment surface 34 or to the peripheral flange portion 35. As an example, the dia¬ phragm 39 may be made from stainless steel and may have a thickness of about 50 μm.

It would be possible to modify the transducer unit 14 shown in Figs. 4-6 and described above in various manners. As an example, the peripheral rim portion of the disc or plate 36 may be fastened directly to an annular abutment surface formed on the stiff metallic transducer block by means of a suitable adhesive. When the inner side of the disc or plate 36 is fastened directly to the metallic transducer block and the outer side surface of the disc or plate 36 is also fas¬ tened to the diaphragm 39 a sealed transducer unit which is able to resist relatively high pressure variations may be obtained.

In the embodiment shown in Fig. 7 and 8 apparatus parts corresponding to parts of the embodiment shown in Figs. 3-6 have been designated similar reference numerals. In Fig. 7 and 8 the flow passage 13 is defined by a bore in the housing

10 interconnecting the inlet and outlet chambers 11 and 12. The inner surface of the bore is in tight engagement with a tubular lining 54, which may, for example, be made from stainless steel or another suitable material defining a sufficiently smooth inner surface in the flow passage 13. The lining 54 does not extend into the inlet and outlet chambers

11 and 12 so that the mounting element 18 of the first embo¬ diment shown in Fig. 3 is dispensed with. In the embodiment shown in Fig. 3 the flow passage 13 and the inlet and the outlet conduits 15 and 16 have a common longitudinal axis coinciding with the longitudinal axis of the flow passage 13. In the embodiment of Figs. 7 and 8, however, a liquid inlet 55 and a liquid outlet 56 have a common axis defining an acute angle with the longitudinal axis of the flow passage 13, and the liquid inlet 55 and outlet 56 are connected with the inlet and outlet chambers 11 and 12, respectively, by means of curved connecting passages 57 and 58. This means that liquid flowing from the inlet 55 to the outlet 56 has to follow a tortuous path formed by the connecting passages 57 and 58, the inlet and outlet chambers 11 and 12, and the flow passage 13. As shown in Fig. 8 the circuitry housing 24 has a substantially rectangular outline, and the longitudinal axis of the rectangular housing 24 may be positioned substantially in a vertical plane containing the axis defined by the inlet 55 and the outlet 56.

Fig. 9 is a block-diagram of the electronic circuitry 23 contained within the circuitry housing 24 of the fluid flow meters shown in Figs. 3-6 and ^\' in Figs. 7 and 8, respectively. As explained above, liquid such as water or another fluid may be passed through the flow passage 13 and the inlet and outlet chambers 11 and 12 in which the ultrasound transducers 14a and 14b are arranged. The transducers are electrically connected to an oscillator 41 via amplifiers 42 and 43, and the oscillator 41 as well as the amplifiers 42 and 43 are controlled by a processor, such as a microprocessor 44 so that the oscillator generates trains of voltage pulses which are amplified by the amplifiers 42 and 43, and the amplified

pulses 45 are transmitted to the transducers 14a and 14b when ultrasound pulses are to be emitted thereby. The pulses generated by the oscillator 41 are preferably 1 Mhz square- wave signals. As explained above, the transducers 14a and 14b may be switched between ultrasound generating and detecting modes, and this may be obtained by activating and de-activat¬ ing the oscillator 41 and/or the amplifiers 42 and 43 by means of the microprocessor 44. In their detecting mode, each of the transducers 14a and 14b detects the ultrasound pulse trains emitted by the opposite transducer so as to determine the time delay or face difference of the ultrasound pulse train moving upstream through the flow passage 13 compared with the downstream pulse train. This time delay or phase difference determined at the detecting mode of the trans- ducers 14a and 14b are detected by a differential detector 46 connected to a pulse integrator 47, in which the time delay pulses 48 received from the detector 45 are added and amplified whereby a delay pulse ΔT proportional to the liquid flow through the flow passage 13 is obtained. The value of the amplification factor of the pulse integrator 47 is pre¬ ferably near to 2000.

The microprocessor 44 may transmit a control signal 49 to the differential detector 46 so as to determine the number of time delays to be detected by the differential detector 46 to be used for determining the delay pulse ΔT. The micropro¬ cessor 44 may also determine the value of width of this delayed pulse on which the calculation of the fluid flow is based. The temperature sensor 22 which is arranged within the inlet chamber 11 is connected to an analogue/digital con- verter 50 from which information about the temperature of the fluid flowing through the flow passage 13 is transmitted to the microprocessor 44 as a digital signal 51. The temperature sensing circuitry may also comprise a reference resistor (not shown) in order to obtain a circuitry which is self calibrat- ing with respect to the measured temperature of the fluid.

Information about the temperature dependency of the ultra¬ sound velocity represented by the graph shown may be stored in a storage element or circuit 52, such as a PROM, EPROM, E ² PROM or any other similar memory circuit. This storage circuit is connected to the microprocessor 44 so that this temperature dependency may be taken into account when the fluid flow through the flow passage 13 is calculated based on the determined delay pulse ΔT.

In order to obtain substantially the same accurate measure- ments by means of all of a plurality of fluid flow meters of the above type, such flow meters must be produced very accu¬ rately with small tolerances. It is especially important that the piezoelectric disc or plate 36 is arranged and retained very accurately in relation to its seat on the transducer block 27 in all of the flow meters produced, which substan¬ tially increases the manufacturing costs of each flow meter.

According to the present invention a compensating or correc¬ tion value compensating for individual characteristics or deviations are determined for each single flow meter as a function of the temperature of the fluid being measured.

Thereby it is possible to obtain very accurate measurements by means of the fluid flow meters according to the invention without the necessity of producing the flow meters with the great care and accuracy required in connection with the production of known flow meters of the type in question. As an example, the same accurate measurements may be made by means of various samples of the flow meter according to the invention in which the piezoelectrical disc or plate is arranged in different positions in relation to the peripheral flange portion 35 of the insert member 33. Thus, in some cases the peripheral surface of the piezoelectrical disc 36 may be in abutting engagement with the peripheral flange portion 35 of the insert member 33 along more or less of the peripheral length, and as the piezoelectrical material of the disc or plate 36, the plastic insert member 33, and the metallic transducer block 27 will expand and contract diffe-

rently with varying temperature the peripheral edge portion of the disc or plate 36 is pressed into engagement with the flange portion at a pressure varying with the temperature. In other cases the piezoelectric disc or plate 36 may be radially spaced from the peripheral flange portion 35 of the insert member 33 at least for a certain temperature range. This means that the ultrasonic vibrations of the piezoelec¬ trical disc 36 of the transducers 14a and 14b may vary with the temperature of the fluid passing the transducers.

Such individual differences and temperature dependent flow irregularities of the various samples of fluid flow meters produced may be taken into account by generating and storing compensating or correction values at different temperatures for each individual flow meter manufactured. For this purpose each fluid flow meter comprises a switch 53 (Fig. 9) which may be activated so as to bring the microprocessor 44 into a test mode. The testing or calibrating procedure at which the compensating or correction values are determined for various temperatures and stored in the storage circuit 52 may, for example be performed as follows:

To begin with the temperature sensor 22 is calibrated by providing a flow of water (200 litre/hour) through the flow passage 13, while the water in the flow is gradually being heated from room temperature to 90°C, typically over a time period of 1 hour. When the temperature of the flowing water is 40°C and 80°C, respectively, a signal is transmitted to the microprocessor 44. These temperatures are then related to the output signals received from the temperature sensor 22 when the 40°C and the 80°C signals are received, respective- ly. Based on information about the linearity of the tempera¬ ture sensor 22 the microprocessor may translate the output signals received from the temperature sensor into correct temperatures.

Now, the switch 53 is activated to place the microprocessor 44 in the test mode, and water or another liquid heated to

the high operational temperature of the flow meter, such as 90°C, is circulated through the flow passage 13 at a known flow rate, which is preferably constant during the testing procedure. In the preferred embodiment the flow rate of the liquid or water is zero, which means that the inlet and outlet chambers 11 and 12 and the interconnecting flow pas¬ sage 13 are filled with the water while the inlet to the inlet chamber and the outlet from the outlet chamber are sealed or closed. From the maximum temperature, such as 90°C, the liquid or water within the flow meter is allowed to cool slowly to room temperature and possibly to a lower tempera¬ ture. During such cooling the temperature of the liquid is measured frequently and the flow through the flow passage 13 is simultaneously measured by the flow meter, and the values measured at the various temperatures are detected and stored in the storage circuit 52. As the actual fluid flow through the flow passage is of a known constant value, preferably zero, any measured value deviating therefrom represents a correction or compensating value for the specific temperature or small temperature range at which the flow measurement was performed. Typically, correction or compensating values are determined for every 1/2°C from 90°C to room temperature. The correction or compensating values which have been determined in this manner and stored in the storage circuit 52 may be used to compensate for any of the temperature dependent errors, such as temperature dependent echoes of the ultra¬ sonic waves from the apparatus wall parts, temperature sensi¬ tivity of the ultrasound transducer or any other temperature dependent error of the apparatus including the electronic circuitry.

It should be understood that the correction or compensating errors determined as a function of temperature within the temperature range in which the apparatus or flow meter may operate are determined individually for each flow meter or apparatus, which means that graphs showing the corrections or errors as a function of temperature within the temperature range of operation are different for different flow meters.

However, the storage circuit 52 of each flow meter has its individual correction curve or corresponding correction values stored therein. Figs. 10 and 11 are curves or graphs showing flow rate compensation values plotted as a function of the temperature of the fluid flowing through the fluid passage 13 for two individual flow meters of the same embodi¬ ment.

When the test program described above has been completed for a flow meter the switch 53 is deactivated to cause the micro- processor 44 to operate in its normal mode of operation. The flow meter is now ready for use in correct measurements of the flow of a fluid passing through the flow passage 13.

During the calibrating or testing procedure as well as during actual flow measurement the operation of the apparatus is controlled by the microprocessor 44. Thus, during the cali¬ brating procedure the processor 44 controls the emission and detection of ultrasound pulse trains by means of the trans¬ ducers 14a and 14b, calculation of the measured flow or correction value (when the flow is zero) based on the time delay signal received, storing the correction value and the corrected temperature value based on the temperature signal received from the temperature sensor at the same time, and repeating such procedure at a predetermined temperature range, such as 1/2°C. During the normal measuring mode the microprocessor 44 controls the emission and detection of ultrasound pulse trains by means of the ultrasound trans¬ ducers 14a and 14b, calculation of the measured fluid flow value based on the time delay signal received from the pulse integrator, and correction of the measured fluid flow value based on the temperature pulse received from the temperature sensor 22 and the correction graph or values stored in the storage circuit 52. The corrected measured flow value thus obtained may be displayed by a display or printer not shown.

The fluid flow meter according to the invention may not only be used for measuring the fluid flow through the flow passage

of the meter but also for other measurements based on such fluid flow measurement. As an example, the flow meter accor¬ ding to the invention may be used for determining heat con¬ sumption of a unit, such as an apparatus, an apartment, or a building to which a flow of hot water or another liquid is supplied. By measuring the fluid flow through the unit and the inlet and outlet temperatures of the unit the heat con¬ sumption within the unit may be determined.