DE3112227A1 - "METHOD AND DEVICE FOR CALIBRATING A FLOW METER CALIBRATION DEVICE" - Google Patents

Description

Patent attorneys. . - ·. '

Dipl.-Ing Dipl.-Chem. '' 'Dipl.-Ing.

E. Prince - Dr. G. Hauser - G. Leiser 3112227

Ernsbetger Strasse 19

8 Munich 60

-AU -

March 11 , 1981

ROCKWELL INTERNATIONAL CORPORATION
6 00 Grant Street
Pittsburgh / Pennsylvania 15219 / V.St.A.

Our reference: R 1014

Method and device for calibrating a flow meter calibrator

The invention relates to a device for the precise calibration of flow meters or checking the accuracy of flow meters, in particular for gas flows, and in particular according to the preamble of claim 1.

From US-PS 185 319 is already the use of a bell-shaped Verification device known, which has a bell-shaped container. ter or has a bell that is straight into a container or kettle moving in and out of it, which is filled with a liquid such as oil. A pulley arrangement with a pulley over the 'bell is provided, with a rope wrapped around the pulley and with one end on the Bell and at the other end attached to a set of weights -is. A line leads from the bell to the one to be checked Flow meter, whereby by pulling up the bell a fluid, e.g. gas, passes through the flow meter and is sucked into the bell. There is a valve in the line

130062/0718

-AA-

provided that in its closed position the flow of fluid from the flow meter into the space within the bell and its kettle, so that the movement of the bell and the weight suspended from it via the pulley blocked. When the valve is opened, the fluid flows into the bell, thus allowing the weights exert their force on the bell, which lifts it. When the weights are released, that will be The rope, and consequently the bell, is pulled upwards so that a vacuum is created in the bell, with the oil forming a seal that prevents air from entering the bell.

To determine the amount of fluid passing through the flow meter is sucked through, it is provided that the movement distance of the bell is controlled, this movement in Correlation is brought to a given amount of fluid that flows through the flow meter, and the known The amount of fluid that has flowed through the flow meter and the calibration device is compared with the display value of the flow meter compared, which is typically read on a counter. According to the conventional technique, it is necessary to to determine the dimensions of the bell (bell circumference measurement), which causes difficulties and a number of errors by averaging the changing diameter and wall thickness changes of the bell as well Interpolation arise when reading the scale markings. The specified accuracy of such methods is at most about 0.3%. Such bell-type calibrators, previously used to calibrate flow meters, thus have inevitable inaccuracies due to errors caused by: 1. Reading the start and end points the bell movement; 2. Read the initial and final volumes on display values on the scale of the flow meter; and 3. the inevitable inaccuracy in determining the bell volume. The biggest mistake comes from the determination the bell volume. The bell itself is indeed at the highest

130062/0716

Manufactured accuracy / changes in their diameter and consequently their size are inevitable. The undertaking Numerous measurements of the circumference of the bell by hooping it with tape have been suggested to measure the middle Obtain the bell circumference and from this the volume of the cylindrical Part of the bell to be determined.

The use of the bell-type calibrators continued for many years, with improvements being made, mainly to determine the movement of the bell and the volume of fluid passed through the flow meter. One of the oldest examples of an automated calibration system is described in US Pat. No. 3,050,980 and includes a bell optical pickups that sense the movement of the bell while it is being moved upwards by a motor. One Line leads from the bell to the flow meter and is connected to a first solenoid operated valve for controlling the Fluid flow from the flow meter to the bell as well as with a second solenoid operated valve coupled to the conduit so as to allow the outflow of the fluid from the bell while it returns to its lowest position. In operation it is originally with air filled bell is lowered into its container to force the air through the flow meter. A dial pointer on The counter of the flow meter, which is referred to as the "calibration pointer", is scanned by means of an optical pick-up trigger the test process, which opens the first valve while the second valve remains closed, allowing air to flow out of the bell through the flow meter can be done. An automatic airtightness test is also described in which both the first and inlet valves and the second or outlet valve are closed, and while a pressure is being built up, it is checked whether in the system and its valves are leaking by measuring the pressure built up in the bell jar.

130062/0716

US Pat. No. 2,987,911 also describes a calibration system in which a temperature sensor is provided at the outlet of the flow meter and a temperature sensor is arranged at the outlet of the calibration device, wherein the Teiiiperaturdif are calculated to a 'J'tMiil> (. * Rutut kojnpen.sat ions Γ ukl.ur Tc / u ei /.euycn, the verwcndot to correct the calculated volume.

US Patent 3,933,027 suggests improvements to the bell-type calibration system by automating its operation. A series of scanning marks is attached to the bell, with an optical encoder detecting the movement of these marks! nstet to display corresponding air volumes as they are sent by the calibration device through the coupled flow meter. A second optical one. Encoder coupled to the pointer of the flow meter to provide an output signal in the form to generate a pulse sequence which iliiM through the Π ußmessei: mv / .eigt i \ li öinende volume, the beginning of the verification of the flow meter, the first optical encoder triggers a gate circuit to initiate a counting process or timing process in which a clock signal is applied to a bell clock counter and to a flow meter clock counter. The gate circuit, which passes the clock signals to the bell counter, is blocked when a given counting rate is reached which corresponds to a known amount of the fluid which has flowed through the flow meter. When the same amount has been measured by the flow meter as indicated by the second optical encoder, a signal therefrom is applied to a gate circuit to terminate the application of the clock signals to the flow meter clock counter. At the end of this process, two counting rates are accumulated, namely a first counting rate of the bell clock counter and a second counting rate of the flowmeter clock counter, the ratio of these counting rates can then be easily calculated and displayed on a suitable digital display. This ratio is called the "registration

130062/0716

accuracy "of the flow rate, i.e. the ratio the actual or calibration volume of the fluid passed through the flow meter to that indicated by it Measured value.

A similar proposal is described in U.S. Patent 3,937,048. In the case of the bell-type automatic calibration device described there A device is also provided to sample the train of pulses emitted by the flow meter during a Operating cycle of the same is generated. Actually through Volume flowed through the flow meter is determined by means of an encoder which generates a pulse train that reflects the linear movement of the bell and consequently that during the test into the Bell in or out of this indicates volume displaced. The encoder delivers a pulse train indicating the volume displaced by the bell. The encoder's pulses are over a given number of operating cycles of the flowmeter accumulated to the volume reading of the flow meter based on a known volume of fluid displaced by the bell was to be calibrated.

Another system is described in US Pat. No. 3,877,287 in which, instead of a bell-shaped container, a cylinder is used, in which a piston is arranged, which is pushed through the cylinder at a controlled speed, by a motor, which is coupled to the piston by a threaded spindle, to drive it through when the motor rotates to move the cylinder. As a result, the piston is driven at a constant speed through the precision bore of the Moved cylinder to displace a fluid from the cylinder and to send through the flow meter to be tested. There are two ways of measuring the flow rate, namely the arrangement of a series of holes in a piston rod, which connects the piston with the threaded spindle, and scanning movement of the holes past a photodetector, as well as the use of an optical encoder connected to the

130 062/0718

Drive motor is coupled to generate a pulse train as an output signal, which the piston displacement and consequently indicates the actual volume displaced from the cylinder.

Also in U.S. Patent No. 3,631,709 is a flow meter calibrator which comprises a piston and cylinder arrangement, the piston being driven by a program controlled motor is driven by a threaded spindle. During operation, the motor drives the piston through the cylinder via the threaded spindle, whereby a known amount of liquid (water) is sent through a series of flow meters arranged in series are. The control program of the motor causes the piston to move at different speeds, whereby corresponding flow rates caused during a single stroke of the piston through the cylinder on the flow meters will. A magnet is coupled to a shaft that connects the threaded spindle with the piston to create a read switch to actuate while the piston is being guided through the cylinder, and to trigger the counting of pulses that be emitted by a first or main pulse generator, which is coupled to the motor. The output of the main pulse generator provides a pulse train that is applied to a register to provide an indication of the actual flow through to win the flow meter. Optical encoders are on coupled to each flow meter to output pulse signals to a second group of registers so that those of the Cumulative flow rates measured by flowmeters and can be displayed. The actual or comparison volume is defined as a certain number of counting events from the first or main pulse generator, with which the output signals of the individual flow meters are compared. The program control of the engine enables the engine and the piston to be accelerated to a stable operating state, before the measurement of the fluid flow through the

130062/0716

away

Flowmeter starts to allow short-term disturbances decay in the fluid.

Efforts have therefore already been made to verify of flow meters by automatically triggering and stopping the counting of pulses a first encoder indicating the comparative volume of the fluid passed through the flow meter and the count of pulses from a second encoder indicative of the volume being measured by the flow meter under test.

So far, however, no ways have been shown, the basic accuracy of the flow meter calibrator, i.e. the accuracy of the actual calibrator. Conventional Flow meter calibrators, particularly of the bell type, can only guarantee an accuracy of ± 0.2% under optimal conditions. Consequently, those with such calibration devices calibrated or checked flow meters themselves do not achieve greater accuracy. One of the main reasons for the lack Precision of conventional flow meter calibrators is the lack of precise methods and devices for high precision Measuring the volume that is displaced in a bell or in the cylinder in the arrangements described above.

The invention is intended to provide a flow meter calibration device with which a much better accuracy is achieved. In particular, an accuracy of 1:10 "is sought. If the volume with such a high accuracy can be determined, it becomes necessary to ensure that the structure containing the chamber is rigid and undeformable is. So far, the bell-type containers did not have such a rigid structure that impacts or the like exerted on them. can result in such a change in the internal volume that the accuracy of the readings of the flowmeter calibrator is affected by the bell type.

130082/0716

According to the invention, "the volume displacement within the Chamber of the flow meter calibrator measured by electromagnetic Waves are generated and the frequency is determined at which in a first and in a second position a piston displaced through the housing results in resonance. This procedure for measuring the container volume requires as the invention teaches, the use of a housing of practically perfect circular or cylindrical shape, so that the frequencies, at which resonance arises, can be precisely determined in order in this way the displacement volume within of the rigid cylinder.

The invention is also based on the knowledge that after once Once the volume of the calibration device has been determined with high precision, it is necessary to determine other parameters, which falsify the display value of the flow meter or the volume of the fluid which has passed through the tested flow meter, with the same accuracy. The invention therefore proposes methods and devices for high-precision measurement of the Temperature and pressure of the fluid within the flow meter and within the flow meter calibrator before, so that a correction factor can be determined with the same accuracy in order to correct for changes in this Make parameters that are in the tested flow meter and present in the flow meter calibrator according to the invention are. This proposal is in contrast to the prior art, according to which flow meter calibrators in rooms with controlled Ambient conditions were brought in which there are small changes in temperature and pressure. But when the precision with which the variables are to be measured reaches the order of 10, as does the invention when measuring the displacement in the flow meter calibrator allows, it becomes necessary to take into account that these pressure and temperature parameters are within of the flow meter calibration device and within the

130062/0716

Flux meter change in the course of his test, so that special measures to ensure the desired precision must be provided for measuring pressure and temperature. For example, if there is an error of 5/9 ° C (1 ° F) when measuring the temperature of the fluid occurs, this can result in an error of 0.2% in the display of the displacement volume on the calibration device result. The flow meter calibration system of the invention however, a display of the volume that has flowed through the flow meter to be tested should be displayed with a precision of 0; 004% enable. With such accuracy, investigations are indicated to determine the effects of other factors on the measurement to determine the volume flowed through. E.g. also has the number of tests performed on a given flow meter are carried out, an influence on its measured value. A change in the flow rate and that through the flow meter The flowed volume also has an influence on the volume display of the flow meter, which corresponds to that of the calibration device measured volume is compared.

130 062/0718

The object on which the invention is based is achieved in particular by the features of the patent claims.

The invention provides an apparatus and a method for calibrating a flow meter. The device contains a chamber, in particular as an essentially completely formed upright or right-angled circular cylinder is designed, a piston which extends in a straight line within the chamber from a first position to a second Can move position, and a measuring device in the form of a Encoder that is coupled to the piston and emits a signal with high resolution, e.g. in the form of a pulse train. The chamber is coupled to the flow meter to be subjected to the test. During the calibration of the flow meter calibrator the piston is brought to the first position, and within the volume between the chamber and an electromagnetic field is generated on the piston. The frequency of the electromagnetic field is varied in order to to establish the conditions for standing waves in resonance within the volume limited in the first piston position. The electromagnetic field is generated so that vibration is generated with a normal mode that is is chosen that the electric and magnetic field components increase in its resonance condition within the chamber are related to the dimensions of the chamber. Then the piston is moved from its first to its second position, while the output of the high resolution encoder is accumulated or counted. In the second piston position is changed again the frequency of the electromagnetic field to the conditions for a standing wave to create resonance within the housing, and then the frequency at which this resonance condition is determined is satisfied. Based on the determination of these two frequencies according to the establishment of conditions for standing waves in resonance at the first and

130082/0716

In the second piston position, the volumes are determined which are passed through the housing or the chamber and the first or second piston position can be determined., then the volume difference or the displacement volume is determined. The displacement volume determined with high accuracy is compared with the output signal of the encoder to determine the increase in volume of the fluid sucked in by means of the piston for a signal unit, e.g. a pulse of the pulse train.

In particular, the diameters of the cross-sections of the cylinder in the first or second position are determined first, and on the basis of these two diameter values, the volumes for these positions are calculated, and the difference in between is determined. Then the displacement volume is divided by the counted number of output pulses, um to obtain a factor that indicates the increase in volume, which is sucked in by the piston per signal unit or output pulse of the encoder.

According to a further feature of the invention, first and second oscillation modes are used as electromagnetic oscillation in the housing or in the chamber, which are selected so that disturbances due to the nature, the shape and the inner surface of the housing are reduced. In one embodiment, the TM ₀₁₀ and TE ₁₁₁ electromagnetic vibration modes are used. According to a preferred embodiment, the device with which the volume is measured contains an antenna which is arranged within the chamber, and a generator which couples electromagnetic energy to the antenna, as a result of which the electromagnetic waves are transmitted into the chamber. A detector is coupled to the antenna to detect electromagnetic energy reflected from the chamber and is also coupled to a resonance detector in the form of a cathode ray tube indicator by means of which a minimum of the level

130062/0716

SL / I

the electromagnetic energy can be determined. Furthermore, a frequency detector in the form of a counter is attached to the Output of the generator connected to determine the frequency of the output signal of the generator at which the Minimum level is observed on the display device, corresponding to the generation of a standing wave in resonance inside the flowmeter calibrator chamber. The frequency or frequencies at which standing waves resonate Occurring within the chamber are in turn used to determine the volume of the chamber of the flowmeter calibrator used. The electromagnetic field is generated inside the chamber with such an oscillation mode, that under resonance conditions within the chamber the electric and magnetic field components of the electromagnetic Field have a certain relationship to the dimensions of the chamber, the chamber having a regular geometry, e.g. that of an upright or right-angled circular cylinder.

130GB2 / 071S

Further advantages and features of the invention emerge from the following description of exemplary embodiments with reference to the drawing.
In the drawing show:

Figure 1 is an elevation of a flow meter calibration system according to FIG the invention;

FIGS. 2A and 2B show individual views, partially in section, of the calibration system according to the invention, which is shown in FIG. 1 in FIG Is shown in the form of an overview;

2C shows the structure of a housing for the calibration device shown in FIGS. 1 and 2A, with a control console via which the user can control and monitor the output data of the calibration device;

Fig. 2D is a display panel of the system control shown in Fig. 2C and condition monitoring part;

Fig. 3 is a functional block diagram of the general layout of the computer system for sensing the various Variables of the flow meter calibration system and for controlling piston movement through the cylinder of the Calibration device shown in Figs. 1 and 2A, and to enable an accurate indication of the registration accuracy the flow meter under test;

4A to 4E different signal conditioning and interface circuits, which are required to receive the signals from the computer system according to FIG. 3 and into it to bring into the appropriate form;

Fig. 4F is a perspective view of the flow meter, wherein it is shown how a proximity detector is arranged with respect to its associated coding mechanism;

130062/0716

4G-M individual circuit diagrams of the circuit diagrams in FIGS. 4A and 4E signal conditioning and interface circuits generally shown ;

Figure 5 is an overview diagram of the program executed by the computer system of Figure 3;

Figures 6A and 6B are more detailed diagrams for initiating the computer system of Figure 3;

Fig. 7 is a detailed diagram of the steps required to calibrate the input variables that are sent to the signal conditioning and interface circuits are applied in order to transfer temperature and pressure measured values to the in Fig. to create the computer system shown;

8A to 8P details of a flow chart for reading out data stored in the computer system and for Enter the conditions under which the flow meter calibration system will test a given flow meter (the letters I and O are not used for clarity);

9A through 9Q are flow charts showing the steps performed by the computer system of FIG perform various functions and make them recognizable (again, for clarity, the letters I and O are not used);

10 shows a circuit diagram of a circuit arrangement for applying a high-frequency signal to a microwave antenna in the cylinder of the flow meter calibrator shown in Figs. 1 and 2A, for changing the to the Signal frequency applied to the antenna, whereby the cylinder volume can be determined with the highest accuracy, so as to accurately encode the output of the linear encoder of the flowmeter calibrator; 130062/0718

FIGS. 11A to E show various input signals which are sent to the Signal 'processing and logic circuits 170a and 170b are applied and generated by these, which are shown in Figures 4C, particularly Figure 4K, are shown generally;

Figures 12A and 12B are perspective views of a lid overlying applied to the piston shown in Figs. 1 and 2A is, or a cross-section of a seal that resembles a spring and around the circumference of the piston head is arranged;

13 is a graph showing the response of a chamber of a perpendicular Circular cylinder shows, which is excited by electromagnetic high-frequency fields, whereby the changing dimensions of the cavity are shown as a function of the excitation frequencies; and

Fig. 14 is a graph showing the cavity response from which the reflected power Pr versus from the excitation frequency.

Reference is first made to FIGS. 1 and 2A. The flow meter calibration system 10 according to the invention shown there is coupled to a flow meter 38 to be checked. The flow meter calibration system 10 comprises a cylinder 12 through which a piston 14 is rectilinear by means of a program variable speed motor 20, e.g. by means of a servo motor.

The cylinder 12 is held in the upright position, namely by means of a series of struts 76, only two of which are shown and which are attached to a flange 77, the again on the outside of the cylinder 12 around it is attached. The upper end of the cylinder 12 is closed by a head piece 86 from which a number of

130062/0718

Struts 88 extends upward (only one of which is shown). A mounting plate 94 is at the top of the struts 88 and a servo motor 20 is mounted on top of the plate 94. The upper end of a lead screw 18 is rotatably mounted in the plate 94 by means of a bearing 98 and is driven with the drive shaft of the servo motor 20 connected via a coupling 100. A guide nut 22, which is fastened inside a housing 23, is on the lead screw 18 screwed on. The threaded or lead screw 18 is arranged telescopically within a sleeve 17, wherein the upper end of the sleeve 17 is firmly connected to the housing 23. The lower end of the cuff 17 protrudes and slidably extends into a socket 96 in head portion 86. The upper end of a piston rod 16 is at the lower End of the cuff 17 attached.

An intermediate cylinder head 91 separates the upper part of the interior of the cylinder 12 from the lower part in which the piston 14 is included. The piston rod 16 extends displaceably through a bushing 93 into the head part 91, the lower end of the rod 16 being attached to the piston 14 is.

When the servomotor 20 rotates, the threaded spindle 18 rotates in the guide nut 22, so that the housing 23, the sleeve 17 and the rod 16 move vertically in one direction or the other, depending on the direction of the Rotation of the servo motor.

The bottom of the cylinder 12 is closed by a head piece 60, and the cylinder 12 therefore encloses and delimits a chamber 28 between the piston 14 and the head piece 60 with variable volume. An opening 62 in the head piece 60 brings the chamber 28 into communication with line 30, line 32 and the flow meter 38. A first inlet valve 34 is arranged between the cylinder 12 and the flow meter 38,

130062/0716

around the fluid flow, in particular gas flow, in between to control. A second valve, namely an outlet valve 36, is connected to the line 30 to allow the fluid to exit to release from the cylinder 12 when it is open.

The exact position of the piston drive rod 16 and consequently the piston 18 is determined by a high precision optical linear encoder 26 reported, which is coupled to the drive rod 16 is and moves with it. The encoder 26 shown comprises two groups consisting of a light source and a photodetector and arranged on both sides of a linear scale 24 which has a large number of scale marks 102 having. In one embodiment of the invention, the linear scale 24 is in a fixed position with respect to the movable encoder 26 and includes 40,000 scale marks 102 (2,500 marks per inch or 98,425 marks per millimeter). In the drawing, of course, only a few scale markings are shown. While the encoder 26 moved in a straight line along the scale 24, two groups of signal pulses A and A are generated which have a phase shift of 90 ° to each other, while the light rays generated by the two light sources from the scale markings 102 are blocked. The output signals A and A of the optical encoder 26 show exactly the position of the piston 14 and consequently the volume of fluid which has flowed through the flow meter 38. As in the following is explained, the volume of the chamber 28 within the cylinder 12 is precisely measured, and each output pulse of the Encoder 26 gives an accurate indication of the increase in volume of chamber 28 in cylinder 12 as piston 14 is withdrawn i.e., is moved upward by the servo motor 20.

Before the upward movement of the piston 14 begins, as a result of which a negative pressure is generated in the chamber 28, the outlet valve is opened

130062/0716

36 is closed and a flow meter valve 34 is opened, thus a flow of fluid over the flow meter 38 can take place via the line 32, the opened valve 34 and the two lines 30, 32 into the chamber 28. at When a flow meter 38 is checked, it delivers an output signal in the form of a pulse train via its encoder 40, which indicates the fluid flow passing through it. The ones from the Encoder 40 derived pulse train is compared with the pulse train that the linear encoder 26 delivers to provide an indication of the Gain flow meter accuracy that is called registration accuracy of the flow meter, i.e. the ratio of that measured by the encoder 40 of the flow meter Volume to the volume measured by the linear encoder 26.

The programmed movement of the piston 14 requires a plurality of proximity sensors 50, 52 and 54 and two limit switches 49, 53. As will be explained below, the servomotor 20 drives the piston 14 in a straight line within the housing or of chamber 28. As shown in FIGS. 1 and 2A, the piston 14 is in its uppermost position, in which a Stop 92 on the housing 23, 26 comes into contact with the upper limit switch 53 and closes it, so that the servomotor 20 is stopped when the piston 14 is moved up, whereby the movement of the piston 14 is prevented. If the Servomotor 20 moves the piston downward in FIG. 1, the stop 92 can push the lower limit switch 49, whereby in turn the piston is stopped. The upper safety switch 53 and the lower safety switch 49 are required to prevent damage to the flow meter calibrator 10 when the servo motor malfunctions 20 pushes the piston 14 to one end or the other. When the stop 9 2 with one of the safety switches 49, 53 When engaged, the servo motor 20 is turned off to stop the piston immediately. In one explained later The proximity detectors become the volume self-test mode

130062/07 16

52, 54 are used to determine the movement of the piston 14 between specified locations. Generally accelerated the servomotor 20 drives the piston 14 to a given speed, and the output of the linear encoder 26 is through the output signal of the proximity detector 52 is switched through so that its pulses can be accumulated or counted. The counting of the pulses obtained from the linear encoder 26 is blocked in the volume self-test mode if an output of the proximity detector 54 appears indicating the passage of the stop 92. After that the servo motor 20 braked to a stop. In the operating mode for checking the flow meter, however, the piston 14 shifted upwards by the servo motor 20, and if the

Stop 92 passes the proximity detector 52, becomes a Release signal generated thereby, and thereby it is caused that when the next output pulse or its Rising edge from the encoder 40 of the flow meter can begin counting the output pulses of the linear encoder. In the operating mode for checking the flow meter, the counting of the pulses from the linear encoder ends when When counting the pulses from the encoder of the flow meter, a predetermined counting rate is reached, the one corresponds to the volume of fluid flown through the flow meter.

To facilitate understanding of the invention, now briefly outlined the operation while providing a detailed explanation of the operation of the flowmeter calibration system 10 takes place later. In the first or initiation subroutine, it is determined whether the piston 14 is in its rest position which is determined by the proximity detector 50 detecting the presence of the stop 92 in the above-explained Way notices; otherwise the servomotor 20 is put into operation to bring the piston 14 into its starting position bring to. When the stop 92 is in a position to be detected, i.e. before the piston 14 is in its starting position

130082/0716

position is returned, the second or the outlet valve 36 is opened to allow the fluid displaced from the chamber 28 can flow out via the line 30, and then the inlet valve 34 of the flow meter is closed to prevent a possible damage to the flow meter 38 Fluid flow occurs. When the user inputs the control command that a flow meter 38 is to be checked, the valve 34 on the flow meter is first opened, and then the second valve 36 is closed to allow the fluid to flow through the flow meter 38, the line 32, the opened valve 34 and release line 30 into chamber 28 while piston 14 is pushed up by servo motor 20. The piston 14 is gradually accelerated until it has a stable speed and is then held at this selected speed during the test measurement, while the output pulses of the encoders 26 and 40 are accumulated in an arithmetic unit which comprises registers in order to accumulate counting events, which in the case of the linear encoder 26 indicate a precise measured volume and in the case of the flow meter 38 the measured volume. The flowmeter test begins when the piston 14 passes the proximity detector 52, which, when the next output pulse or the rising edge from the encoder 40 of the flowmeter occurs, releases the counting or accumulation of the pulses from the encoder of the flowmeter as well as from the linear encoder releases. Depending on the desired volume that is to flow through the flow meter 38 during the test operation, the test process ends when a predetermined count rate of pulses from the encoder of the flow meter is reached. In particular, the register provides for cumulating the pulses from the encoder of the flow meter 38 signal ', which is applied to the linear encoder system upon reaching the predetermined count, which depends on the volume of fluid, an output, whereby the count of the input pulses from the encoder 40 of the flow meter is terminated. The stored counting events, which the means of the coding

130062/0718

show the volume of fluid measured by the flowmeter 40, and those of the linear encoder 26 are compared with each other, that is, the ratio between these two values becomes intended to gain an indication of registration accuracy.

In addition, measurements of temperature and pressure are made in order to adapt the measured volumes to these conditions to vote. In particular, a temperature measuring device 42 is located at the inlet and at the outlet of the flow meter 38 and 44 arranged. A pressure differential transducer 46 is provided to the pressure differential between that of the fluid in to determine the pressure built up in line 32 and the ambient pressure. In addition, there are temperature measuring devices 48,57 on the line 30 b2w coupled to the chamber 28. placed on the piston 14 to control the temperature of the fluid within the chamber 28 to indicate. There is also a second differential pressure transducer 51 arranged within the piston 14 to the pressure difference between the ambient pressure and the pressure to be indicated within chamber 28. The temperature output signals TM3 and TM4 of the temperature measuring devices 42,44 are averaged to obtain an averaged flow meter temperature AMT, while the output signals TP1, TP2 of the temperature measuring devices 57, 58 are averaged to obtain the average temperature APT of the calibrator. As will be explained in detail later, these Input parameters used to correct the measured volumes obtained from the encoders 26 and 40 depending on the measured pressure and temperature conditions.

As shown in Fig. 2A, a microwave antenna 70 is disposed in the head portion 60 and generates microwaves within the chamber 28, whereby its volume can be precisely determined. This measurement can be done periodically, too Detect the slightest changes in the volume of the chamber 28.

130062/0716

-vt-

As will be explained, enables the use of electromagnetic Waves in the microwave range have an accuracy of at least 1:10 when determining the volume of the Chamber 28. In order to check the accuracy of the temperature sensors 4 8, 57, a high-precision temperature measuring device is also used 6 8 arranged in the line 30. The temperature converters 48, 57 can be temperature converters of the RTD type, type no. 601222 manufactured by Senso-Metrics, Inc. while the high-precision temperature sensor 68 is the 18115A from Hewlett Packard. This converter 68 must in conjunction with a Hewlett Packard quartz thermometer 28O4A can be used. The proximity switches 50, 52 and 54 can, according to one embodiment, be the types FMSA5 from the manufacturer Microswitch, while the limit switches 49 and 53 are types BA-2R-A2 or ADA3721R from the same manufacturer. The servo motor 20 can be the NC100 type with a DC servo motor / tachometer from the manufacturer Control Systems Research, Inc. In the optical encoder 26 and the Scale 24 can be the linear encoder Pos-Econ-5 from the manufacturer Heindenhain Corporation. As in detail As explained later, the measurements of pressure and temperature are made within the flow meter 38 and within the calibration device 10 is used to calculate a factor by which the volume flowed through the calibration device 10 in view of corrected for these variables. In particular, the temperature transducers 48, 57 measure the temperature at the bottom or in the top Area of the chamber 28 to obtain a spatially averaged measurement of the temperature within the entire volume of the chamber 28 of the calibration system 10 to win. In the same way, the temperature converters 42, 44 provide indications of the temperature of the gas at the inlet or outlet of the flow meter 38, thereby a spatial averaging of the temperature of the flow meter 38 gas flowing through can be obtained. The temperature transducers selectively used in the flow meter calibrator 10 are built-in, are highly stable platinum resistance thermometers with low thermal inertia and

130Q62 / 0716

give an accurate reading of these variables, which makes the The volume flowing through the flow meter calibration device 10 can be precisely determined.

As shown in FIGS. 2A and 2B, the proximity detectors are 50, 52 and 54 and the limit switches 49 and 53 mounted on a rising part 90, which in turn is attached to one of the upstanding struts 88 is mounted. One end of the linear scale 24 is suspended from a part 94 and extends substantially parallel to the direction of movement of the piston 14 downwards, with its lower end at the upper Part of the inside of the cylinder 12 is mounted. Only a small number of the approximately 40,000 markings 102 are in 2A shown. In the lowermost part of Fig. 2A, the inlet valve 34 and the outlet valve 36 are shown, which from the opened into the closed position can be actuated via pneumatic actuating devices 66 and 64, respectively.

In Fig. 2A a generally designated 78 sealing device of the piston 14 is shown, which prevents the over the flow meter 38 in the chamber 28 directed fluid exits at the edges of the piston 14, while this is rectilinear moves within the cylinder 12. Details of this Piston seals 78 are described in co-pending US patent application entitled "Piston Seal".

Reference is now made to Figure 2C. The flow meter calibrator 10 is arranged there in a conditioned environment which is formed from an environment enclosed by walls 104 Calibration room 106, in which the flow meter calibration device 10 is located, and a control room 105 in which the control panel is located, with a display unit 111 comprising a printer and a series of lamps to display the display different states of the calibration device, with a cathode ray tube viewer 112 with input keyboard, via the various control commands can be entered, as well as

^ 33 -

with a variety of heat generating devices such as power supplies, Motor control, amplifier, etc. The conditioning of the ambient conditions of the calibration device 10 is thereby ensures that this is in the calibration room 106 away from the heat-generating display unit 111 and the station 112 is located. As shown in FIG. 2C, the flow meter 38 to be checked is also arranged within the calibration space 106 and coupled to the calibration device 10 via the line 32. The temperature in the calibration room 106 becomes more temperature sensitive by means of four Devices RO, R1, R2 and R3 measured, which are distributed over the calibration space 106 or on a strut 76 of the Calibration device 10 are arranged. A barometer pressure transducer 109 and a barometer are also arranged in the calibration room 106, to measure the ambient pressure. The electrical connections of the various temperature measuring devices shown in Fig. 2C and the temperature and pressure measuring devices shown in Figures 1 and 2A are through two ducts 110 to the control panel guided, which is located within the control room 105. In this way, the environmental conditions under which the Flow meter 38 is checked and the calibration device 10 is working, precisely controlled by the means of the flow meter 38 of the calibration device 10 to make measurements made reliable.

Reference is now made to FIG. 3. This shows a block diagram the general design of the computer system to perform the various functions, including processing the temperature and pressure measurements, the output signals of the linear encoder and the encoder of the flow meter and the appropriate control of the valves 34, 36. Furthermore, output signals for the viewing station 112 are used Display of the measured parameters and generated for the system control as well as the display unit 111 to the various Display operating conditions of the system 10 while the user selectively uses the keypad of the station 112 to use the flowmeter test functions can enter. The computer system

130082/0716

comprises a central processing unit (CPU) 120 as disclosed by US Pat manufactured by the applicant under the name PPS-8 Microcomputer; therein become address signals from the CPU 120 via an address bus 128 to a programmable read-only memory (PROM) 124 and to a main memory (RAM) 126 sent. The main memory 126 can have the form of a 256 χ RAM, as it is manufactured by the applicant, while the programmable read only memory 124 may be the 2708 from Intel Corporation. A "system clock 122 provides Clock signals (e.g., 200 kHz) to the CPU 120 and may, for example, take the form of a clock generator circuit P / N 10706 as disclosed by US Pat the applicant is produced. As shown in Figure 3, both clock 122 and CPU 120 are programmable Read-only memory 124 and the main memory 126 are connected to one another by a command / data (I / D) bus line 140, at which can be, for example, a 14-wire bus that not only connects the named components with each other, but also also to the signal conditioning and interface circuits 130, 132, 134, 136 and 138 is connected.

The circuit 130 brings the signals TP1 and TP2 from the Temperature measuring devices 57, 58 are derived and the Display the temperature of the calibrator in the appropriate form. Also the flowmeter temperature signals derived from devices 42,44 TM1 and TM4 are applied to circuit 130. As shown in Figure 2C, there are four additional Room temperature measuring devices RO, R1, R2 and R3 in the calibration room 106 arranged in which the calibration system 10 shown in FIGS. 1 and 2A is located. It is assumed that the environmental conditions of the calibration system 10 are well regulated in order to keep the ambient temperature of this system as constant as possible keep. It is common practice to keep the flowmeters to be checked in this environment for a period of time so that they reach the same ambient temperatures as the calibration system 10. As shown in Fig. 3, the temperature signals to a signal conditioning and logic circuit 150

130082/0716

and arrive from there via an interface circuit 151 to the I / D bus 140.

In the same way, the pressure signals are applied to the conditioning and interface circuit 132, which has a signal conditioning and logic circuit 162 and an interface circuit 164. In particular, be attached to the circuit the output signals of the pressure difference measuring device 151, which indicate the pressure difference PP1 of the calibration device 10, and the output signals of the pressure difference measuring device 46, which indicate the pressure values MP2 on the flow meter, as well as the output signals of the barometer pressure measuring device 109, which indicate the ambient or atmospheric pressure PB in the calibration room 106, are applied.

The output signals of the flowmeter encoder 40 and the linear optical encoder 26 are sent to the signal conditioning and interface circuit 134 applied. In particular, the output signal of the encoder 40 is sent to a signal conditioning and logic circuit 170a applied, their output signals can be applied to the interface circuits 171 and 173 in turn. A clock circuit 175 applies a signal to the Interface circuit 173 on. In another embodiment, a proximity detector 27 is used to detect the rotation of the flowmeter encoder 40 and the output of the proximity detector 27 is applied to the signal conditioning and logic circuit 170a. This is in FIG. 3 by input signals called rotary encoder pulses and by the term proximity detector for flow meter test specified. Of course, there is only one signal input to circuit 170a at a time. The output signal of the The linear optical encoder 26 is connected via the signal conditioning and logic circuit 170b and the interface 179 the I / D bus 140 is applied.

In order to get an indication of the measured parameters, e.g. temperature, pressure and fluid volume, which the calibration device 10

130062/0716

is flowed through and measured by the flow meter 38, the output signals from the I / D bus 140 are given via the circuit 136 to the display unit 111, which includes a data printer such as those manufactured by Practical Automation, Inc. under the designation DMTP-3 will.

In particular, circuit 136 includes an interface circuit 192a for applying parameter output signals to the printer via a signal conditioning and logic circuit 190a. The circuit 136 further comprises an interface circuit 192b, which the parameter output signals via a signal conditioning ' processing and logic circuit 190b to the display device or data station 112 gives up. In addition, command signals are entered by the user via the station keypad and via logic circuitry 190b and interface circuit 192b on the I / D bus 140 given. A clock circuit 193 controls the following speed, with which the signals are transferred between station 112 and the computer system. The station 112 can, for example, be of the type 1500 from Hazeltine Corporation. About such a station input commands can be entered using an alphanumeric keyboard and the entered control commands as well as those from parameter data retrieved are displayed to the user.

Finally, there is a signal conditioning and interface circuit 138, which connects the I / D bus 140 to the inlet valve 34 and the outlet valve 36 connects and applies the control signals to the servomotor 20. The servo motor 20 is also assigned a motor control, as it is under the type designation DC servo controller NC101 from Control Systems Research, Inc. is manufactured. This generates feedback signals that determine the speed of rotation of the servo motor 20 and can be applied via the logic circuit 94 in order to in turn influence the control of the servomotor 20. In addition, signals indicating the state of the servomotor 20 and an input signal from a test mode switch,

130062/0716

which indicates whether the rotary encoder 40 or the proximity detector 27 is used to measure the flow rate, via the logic circuit 198 and the interface circuit 200 on the I / D bus 140.

The signal conditioning and interface circuit 130 is shown in more detail in Fig. 4A and includes two input lines leading from the temperature transducers 57, 48, 42 and 44, which indicate the temperatures TP1 and TP2 or the temperatures TM3 and TM4 of the flow meter, to a multiplexer

149 lead. In addition, the four temperature converters RO, R1, R2 are via the next four inputs of the multiplexer 149 and R3, which are arranged in the calibration room 106 in which the calibration system 10 is located. The temperature converters mentioned are switched on in amplifier modules, which serve to generate output voltages proportional to are the sensed temperatures and are applied to the corresponding inputs of the multiplexer 149. In this way each temperature converter is assigned its own amplifier module, so that its output signal delivered to the multiplexer 149 can be adjusted in such a way that all of them are sent to the multiplexer connected temperature converters have the same voltage amplitudes and zero point shift. details of the amplifier modules of each of the transducers shown in Fig. 4A will be explained later with reference to Fig. 41, where at the same time a detailed explanation of the signal conditioning and logic circuit shown only generally in FIG. 4A

150 takes place. The multiplexer 149 is used to sample the input signals at each of its eight inputs in time division multiplex and to bring the temperature signals to the appropriate level, which are applied individually via a multiplexer 154 to an amplifier 152, for example from the TP AD522 of Analog Devices is. The second multiplexer 154 is normally set to have one of the eight temperature input signals via the operational amplifier 152 to an analog / digital (A / D) converter 158, e.g. of the ADC 80 type from

130062/0716

Burr Brown can be. In calibration mode, the multiplexer 154 applies a precise calibration voltage to the A / D converter 158. As is well known in the art, the DC voltage of the analog signals is set to a level that is different from that of the A / D converter 158 can be easily processed, which in turn converts these analog signals into digital signals, which via its twelve output lines go to a parallel input / output circuit (PI / O) 160, e.g. of the type P / N11696 of the Applicant is. This P / O circuit 160 enables Input commands pass through I / D bus 140 to multiplexer 149 to determine which of the inputs to a particular one Time is sampled, and passed to the logic circuit 156, so that the multiplexer 154 one of the output signals of the multiplexer 149 or the calibration voltage signal to the A / D converter 158. In operation, the CPU 120 applies a polling signal to the PI / 0 circuit via the I / D bus 140 160, which responds by calling up information previously converted to digital values and applying the same to the I / D bus 140 releases. In addition, a command is derived from the PI / O circuit 160 to control the timing the conversion of the analog signals into digital signals by means of the A / D converter 148, and a state the signal indicating this A / D converter is passed through the PI / 0 circuit 160 given to the I / D bus 140. The calibration voltage signal allows operational amplifier 152 to be zeroed and span adjusted to full amplitude any input signal to the A / D converter 158 can appear.

The rendering and interface circuit 132 is shown in FIG 4B shown in detail. It is shown there that the output signals the transducers 51 and 46, which indicate the calibration device or piston pressure PP1 and the pressure MP2 of the flow meter, via operational amplifiers 161a and 161b to a multiplexer 163 can be created. In addition, the output signal of the pressure transducer 109 is used to measure the barometer or ambient

130062/0716

-35 -

pressure in the calibration room 106 is applied to the multiplexer 163 via an operational amplifier 161c. Initially, the CPU 120 transfers a command to multiplexer 163 to select which The output signal of the pressure transducer 51, 46 or 109 is to be read out via the I / D bus 140 and the PI / O circuit 168. In response, this PI / O circuit 168 applies control signals to the multiplexer 163 via a four-wire bus 169 select one of the three pressure indication signals or a signal indicating the calibration voltage to be sent to an A / D converter 166 is to be applied, which converts the analog input signals into a corresponding digital signal in order to transmit them to the Apply PI / O circuit 168 and then via I / D bus 140 transferred to. At the command of the CPU 120 then controls the ΡΙ / 0-τ Circuit 168 connects to A / D converter 166 over line 159 to convert the selected analog pressure signal to a corresponding digital signal which is transmitted over I / D bus 140 target. The conversion of the analog input signals into digital signals requires a discrete period of time, which is also required is used to stabilize the digital signals appearing on the twelve output lines of the A / D converter 166 before they are read out by the PI / O circuit 168. When the signals on the twelve data lines are stabilized, a status signal is generated by the A / D circuit 166. In response to this status signal, the PI / O circuit reads 168 takes the data appearing on the output lines of the A / D circuit 166 and applies it to the via the I / D bus 140 RAM 126 shown in Fig. 3. After this process is complete, the system is ready to start another Select pressure output signal derived from another transducer is converted to a digital signal and arrives at the main memory 126 in the manner described above.

The interface and conditioning circuit 134 is below Further explained with reference to FIG. 4C. The flow meter rotary encoder 40 is coupled to the flow meter 38 in order to emit two signals A and A which are 90 degrees from one another

130062/0716

- Ht) -

are out of phase, whereby these signals are sent to the signal conditioning and logic circuit 170a are applied. In particular this circuit 170a processes the input signals A and A in such a way that signal fluctuations are eliminated how they can be set on the rotary encoder 40 due to mechanical vibrations. The signal conditioning and logic circuit 170a generates a composite pulse signal corresponding to each group of input pulses of signals A and A "and applies that signal to an interval timer 174 which is a programmable Counter 174a includes, in which a factor is entered, which is to be conducted by the set, over the flow meter Fluid volume depends, as will be explained later. This factor is entered into the programmable counter 174a, and if this counts back to zero based on this factor has, a pulse is generated by means of the interval timer 174 and applied to a logic circuit 177, the output signal of which an interrupt subroutine 2 triggers the checking of the flow meter 38 is terminated, as will be explained in detail later with reference to FIG. 9K. The logic circuit 170a responds the input signal A of the flowmeter rotary encoder 40 to apply a suitably conditioned pulse to an interval timer 176 that shows the leading edge of the Signal A detects in order to control the timing or counting of the programmable counter 174a as well as the counter 176a and the Trigger linear encoder counter 182. In particular, the CPU 120 controls the flow meter calibrator 10 and determines when the piston 14 is accelerated from its rest position to its test start position, which is due to the presence of a Output of the proximity detector 52 is displayed. When this output signal is determined, the from of CPU 120 periodically, e.g., about every 40 microseconds Program an access to the interval timer 176 to check whether it has an output signal from the signal conditioning and logic circuit 170a which has received the

130062/0716

Leading edge of the input signal A from the rotary encoder 40 indicates. When the first leading edge of the output is detected the circuit 170a after the piston 14 on the proximity detector 52 has passed, a trigger signal is applied to the PI / 0 circuit 184 which then provides a count trigger signal applies to the counter 182 and also applies a trigger signal to the programmable counter 174a via the I / D bus 140. In this manner, all of the counters 174a, 176a and 182 are activated to begin counting at the same time. With the one shown For example, the programmable counter 174a counts down in response to output signals from the rotary encoder 40.

According to an essential feature of the invention, the triggering takes place and the completion of the flow meter check, i.e., the count in programmable counter 174a and counter 182 in response to the outputs of the rotary encoder 40 in the flow meter 38 whose accuracy is to be measured, in particular the rotary encoder 40 coupled to the fluid or gas flow meter 38, as will be explained in detail with reference to FIG. 4F; at Rotation of a rotatably mounted rod of the same, the rotary encoder 40 coupled thereto generates a pulse train which corresponds to the rotation of a tangent arm and the cycle of the diaphragm of the flow meter. As in Figure 4F is shown, the rotatable part is connected to the diaphragm of the flow meter via coupling arms, and its rotation is non-linear, so that the output signal of the encoder 40 is, as it were, frequency modulated. So to get an accurate measurement to obtain by means of the Rotationscodxerers 40, the aim is to count the pulses derived from this, in such a way, that the counting begins and ends at approximately the same point of rotation of the rotating part of the flow meter. This is achieved by initiating the count in response to this rotary encoder 40. In particular

130082/0716

A flow meter test is performed by moving the piston 14 from its rest position to a steady speed is accelerated so that when passing the proximity detector 52, which is located at the test start point, an output signal derived therefrom in order to detect the leading edge of the next output signal from the logic circuit 170a corresponding to the leading edge of the next output signal A of the rotary encoder 40. The interval timer 176 responds to the leading edge to initiate the counting of the programmable counter at the same time 174a and the counter 182 to effect. If the programmable counter 174a based on the selected factor, which depends on the desired volume to be conducted via the flow meter 38, has counted down, the interval timer 174 sends its output signal to the logic circuit 177, to enable the interrupt program 2 of the CPU 120, which Again, the counting of the counters 174a and 182 is ended and the respective count rates are transferred to corresponding locations within the main memory 126 convicted. The end of the count can be done program-controlled; however, this would be an additional one Number of program steps required, so that the programming of the system would be relatively complicated and additional time would be required to perform the timing process described above. Triggering the count in response to the output signals of the rotary encoder 40 also ensures a more accurate test and calibration of the checked flow meter 38.

As explained above, the pulse sequence obtained from the rotary encoder 40 is used to convert the initially into the programmable Counter 174a to count down the counting rate entered. The count rate entered into the programmable counter 174a depends on the volume which is to flow through the flow meter 38 and get into the chamber 28. The count rate is based on the dimensions and characteristics of the flow meter 38 and on the characteristics of the rotary encoder 40 or the number of

130062/0716

pulses generated by this per revolution. In one embodiment a count rate of 40,000 is entered into the programmable counter 174a, corresponding to a volume of 28.31 liters (1 cubic foot) flowing through the flow meter 38 target. Assuming that the characteristics of the flow meter 38 and the encoder 40 for different volumes remain the same, e.g. the counting rates 20,000 or 10,000 can be stored in the programmable counter 174a, if only half or a quarter of the specified volume is to flow through the flow meter 38. By entering, one Counting rate, which is based on the characteristics of the flow meter 38, into the programmable counter 174a, which is determined by the pulses is counted down from the rotary encoder 40, a more accurate check of the flow meter is achieved by starting and end of the test process at the same point in the rotation cycle of the flowmeter and its rotary encoder 40 take place, as already explained above.

As shown in Fig. 4C, a clock signal A is derived from the system clock 122 by the CPU 120 and is applied to interval timer 176 via I / D bus 140. A selected factor is entered into the down counter 176a, to provide an output on interval timer 176 generate which a sampling pulse signal with one pulse per Second corresponds. In one embodiment, the system clock signal from clock 122 has a frequency of about 200 kHz, and the factor entered into counter 176a is chosen so that logic circuit 178 will generate the desired one pulse per Second, as well as starting from this the motor control circuit. As will be explained later, this sampling pulse is used to control the timing of the sampling of the measured values of pressure and temperature.

The output of the linear encoder 26 is in the form of two signals A and A ″ which are 90 ° out of phase and is applied to a signal conditioning and logic circuit 170b

130062/0716

will. Circuit 17Ob is similar to circuit 170a, to the extent that it brings the input signals A and A "into shape and eliminates signal fluctuations that would otherwise occur would lead to an incorrect output signal of the linear encoder 26. In addition, the circuit 17Ob is capable of the Determine the direction in which the piston 14 moves, based on the inputted signals A and A; if the submitted Signals A and A do not indicate that the piston 14 is moving in the desired direction, so the Circuit 170b does not output any signals. The circuit 170b generates a sequence of conditioned pulses according to the linear encoder output signal applied to counter 182, this counter begins to count after being triggered and the output signal of the linear encoder 26 accumulates. The cumulative The output of the counter 182 is sent to a PI / O circuit created, and transferred to the other parts of the computer system via the I / D bus 140 when the corresponding command is given.

According to a particular embodiment of the invention, a rotary encoder or transducer 40 is attached to the flow meter 38 coupled and contains an optical encoder which is rotatably coupled to the flowmeter 38 own tangent arm is to detect its rotational movement while a gas flow occurs therethrough to a plurality of output signals A and A as explained above. In another embodiment of the invention, the proximity detector 27 can be used to determine the mechanical rotation of the tangent arm of the flow meter 38, namely by means of a mechanism, which will be explained in detail later, to provide an output signal to the signal conditioning and logic circuit 170c to deliver, which in turn with the interval timer 176 and the Interval timer 174 is connected. Because of the arrangement of the tangent arm coupled to the tangent arm of the flow meter 38, the proximity detector 27 generates a lower signal Resolution than the optical encoder 26 as it detects the rotation of the tangent arm of the meter. The special

130062/0716

Their advantage of the arrangement with proximity detector is that it is mechanically and electrically very simple. The decision whether a proximity detector 27 or a rotary encoder or converter 40 is used, is carried out by the user by operating a switch 191, as shown in FIG. 4K is shown. If the proximity switch 27 is selected, the programmable counter 174a is coded with the numbers 8, 4 and 2, corresponding to the volumes to be measured of around 28, 14 or 7 liters. The user triggers the input of the suitable factors, whether for the proximity detector 27 or the rotary encoder 40, by first move switch 191 to the appropriate position and then enter the test volume using the station 112 keypad.

4F shows a perspective view of a typical embodiment of the flow meter 38 which measures fluid flow through the use of two diaphragms? of these is only which one membrane 1202 is shown. The flow meter shown in Figure 4F is described in detail in U.S. Patent 2,544,665. As can be seen, a flagpole 1203 senses the deflection of the membrane 1202 around an arm 1204 in FIG To set vibration. A second membrane (not shown) and an associated flagpole (also not shown) cause the arm 1206 to oscillate at a different cycle. As explained in said US-PS, the combination of arms 1206 and 1204 with arms 1208 and 1210 causes that a tangent arm 1214 is set in rotation. A metal sample 1212 at the intersection of arms 1208 and 1210 is made rotated past the proximity detector 27, whereby an output signal is given to the logic circuit 170c to in to be processed in the manner explained above.

The signal conditioning and interface circuit 136 is shown in detail in FIG. 4D. There the connections of the printer 111a and the display unit 111 are shown, for output

130062/0716

Uw qrwiHuu-lii on, from which
10 measured parameters, including the measured flow rate (η) and the percentage error of the tests carried out. In particular, the printer 111a is coupled to the I / D bus 140 via a first logic circuit 190b and via a parallel data controller (PDC) 192b, which can be, for example, the applicant's type 10453 and which provides controlled two-way access between the I / D bus 140 and printer 111a guaranteed. Thus, on command of the PDC circuit 192b, a signal is generated by the logic circuit 190b, whereby the printer 111a is keyed and an appropriate acknowledgment signal (ACKO) is transmitted via the logic circuit 190b and the PDC circuit 192b to indicate that the printer 111a is ready to print. If the printer 11.1a is busy, a suitable busy signal is transmitted back onto the I / D bus 140. When a command has been issued to print data, the control portion of the signal is transmitted through the PDC circuit 192b and the logic circuit 190b to control the printer 111a to print out the data appearing in the data channel leading out of the logic circuit 190b .

Via the keyboard of the viewing station 112, suitable Commands are entered, this station via logic circuit 190a, a serial data controller (SDC) 192a and the I / D bus 140 is coupled to the computer system. The SDC circuit 192a can be of the type 10930 of the applicant. This SDC circuit 192a can receive serial data, originating from the viewing station 112, including the commands entered by the user on the keyboard. This SDC circuit 192a sets these serial input signals from the logic circuit 190a with a suitable one by means of the clock generator 193 certain subsequent speed and transmits a group of digital signals over the I / D bus 140 to the CPU 120. The data to be displayed on the screen of the station 112

130062/0716

again via the I / D bus 140 and via the SDC circuit 192a and logic circuit 190a are transmitted to the station to get.

The signal conditioning and interface circuit shown in Fig. 4E 138 is between the I / D bus 140 and the first or inlet valve 34 and the second or outlet valve 36 and sends signals to and receives signals from the engine control unit. Motor control signals that determine the speed and indicate the direction of rotation are via the I / D bus 140 applied to a logic circuit via the PI / 0 circuit 196 194a, whereby these digital signals are applied to the engine control in order to ensure that the Servo motor 20 to effect. At the appropriate time, as controlled by the executing program, signals are generated to control the Open or close valves 34 and 36 by actuating their associated coils 66 and 64, respectively. These valve control signals are via the PI / 0 circuit 196 and the logic circuit 194b applied to two pneumatic valves which are located within the control room 105 (FIG. 2C), whereby Compressed air at a pressure of 3.445 bar (50 psi) is selectively applied to one of the valve controls 66 and 64, respectively to open or close these valves on command. This leaves that of the valve actuators generated heat from the temperature-controlled calibration room 106 remotely. Additionally, each actuator or coil 66 includes a proximity detector to determine if the valve is open or is closed. The output signals generated by the proximity detectors 50, 52 and 54 to determine the The approximate position of the piston 14 are given to the I / D bus 140 via a logic circuit 194c and the PI / 0 circuit 196. A second sampling clock is generated by the coding circuit and sent to the PI / 0 circuit via the logic circuit 194d 196 created. By operating a switch, the user can determine whether the fluid flow information is from the

130062/0716

Proximity detector 27 or from the encoder 26 can be obtained as shown in Fig. 4C. This control command is given to the I / D bus 140 via the PI / O circuit 196. To a To achieve suitable control of the servomotor 20, its states with respect to speed, direction of rotation and torque given to the I / D bus 140 via logic circuit 198b and PI / 0 circuit 200.

The display unit 111 includes a front panel shown in Fig. 2D and is provided to display information about the various by means of a series of lamps and backlit pushbuttons To give states of the system. As shown in Fig. 2D, the display panel also includes the printer 111a for Print out the flow rate and percentage of fluid flow through the flow meter 38. In addition, a plurality of lamps 111b-111e are provided. The lamp 111b is turned on to indicate that the calibrator 10 is self-testing, as will be explained later. A ready light 111e indicates that power is being applied to the calibrator 10 and that an initiation process has begun to bring the device into its ready state. During this Standby mode is through a series of keystrokes requires the user, and upon completion of the data entry via the keypad of the station 112, the calibration device 10 goes automatically into the state of the test sequence, which is indicated by a corresponding lamp 111c. In this operating state the flow meter 38 is actually checked. When a flow meter check is completed, a corresponding one lights up Lamp 111d on. At this point the final value will be the percent accuracy is calculated and provided by printer 111a printed out. In addition, a backlit stop push button 111g and reset push button 111f are provided. If the reset button 111f is pressed during any phase of operation, the calibration device 10 will respond as if the operating power would be turned on as referring to

130062/0716

can be seen on Figure 9H. The stop push button 111g is only depressed when an emergency occurs that could damage the calibration device 10. When the stop push button is pressed 111g, the servomotor 20 is quickly braked and stopped, and the calibration device 10 is locked in this operating state until the main power supply is switched off and switched on again is. When the operating voltage is applied again, the calibration device TO goes into its standby state. During the Execution of the program, suitable signals are generated and transmitted via the I / D bus 140, the P / O circuit 200 and the logic and driver circuit 198a applied to drive the appropriate indicator lights 111b-111e.

Those shown in the block diagrams of Figures 4A through 4E Signal conditioning circuits are more detailed in the 4G-M schematics. The signal conditioning circuit 130 shown generally in FIG. 4A is particularly illustrated in FIG Figures 4G, 4H and 41 are shown in detail.

In Fig. 4G the multiplexer 149 is shown which comprises a plurality of relays with mercury-saturated relay contacts, which reduce the volume resistance; these relays are coupled to the channels connected to the amplifier modules, to supply a voltage to the respective associated relay of the multiplexer, which corresponds to the temperature, which of a the transducer 57, 48 of the calibration device, the temperature transducer 44, 42 of the flow meter or one of the room temperature converters RO - R3 is measured. The selected channel is energized by the relay of multiplexer 149 via the second in Figure 4G Multiplexer 154 shown as well as amplifier 152 are applied to analog / digital converter 158 shown in FIG. 4H. In 4H is the PI / O circuit 160 on the I / D bus 140 the CPU 120 coupled. In addition, the PI / 0 circuit becomes T60 derives an output signal to be sent to a heximal-to-1 decoder 153 shown in Fig. 4G,

130062/0716

which in turn controls a plurality of drivers 155 in order to close the corresponding relay of the multiplexer 149. The output of the PI / O circuit is further turned into an inverter through the logic circuit 156 which is an AND circuit 156a 156b and a logic translator 156c applied to multiplexer 154. When in the arrangement shown in Fig. 4H the output signal of the selected temperature measuring module is applied to the A / D converter 158, this converter receives a conversion signal from the PI / O circuit 160 via an expander circuit 153. In response to this repositioning command, it sets the converter 158 converts the inputted analog temperature signal into a corresponding digital signal that is transmitted via the PI / 0 circuit 160 is given to the I / D bus 140, and then gives a signal indicating the completion of the relocating process via line 147 to the P / O circuit 160.

Reference is now made to FIG. 41. There's a schematic there an amplifier module shown, each with a temperature measuring device is connected to amplify their signal and to generate an output signal via a corresponding Channel is applied to the multiplexer 149. The temperature measuring devices can, for example, be resistance temperature meters, as manufactured by Senso-Metrics under the designation 601222. The temperature sensitive resistor (RTD) is connected in a branch of a resistor bridge 201, which contains the resistors R1, R2 and R3 in addition to this resistor. The excitation voltage applied to the connections a and d of the bridge 2O1 as well as that tapped at the connections c and d of the same Output voltages are sent to a conditioning circuit 203, as it is manufactured e.g. by Analog Devices under the designation 2B31. Essentially contains the Processing circuit 203 an operational amplifier 205, to which the output signal of the bridge 201 is applied before being applied to a Bessel filter 201 which removes selected frequencies

130062/0716

can before further amplification in an operational amplifier 209 takes place / so that the amplified signal can be applied to the corresponding channel of the multiplexer 149. An extremely well stabilized voltage supply is used to control circuit 203, which is e.g. Power supply can act as it is manufactured by Analog Devices under the designation AD584. As in Fig. As shown, the gain of operational amplifier 205 is determined by the resistance that exists between the terminals 10 and 11 of the circuit 203 is, while the zero deviation of the output voltage is done by tuning a potentiometer, which is applied to terminal 29 of circuit 203. Voltage and current supplied to bridge 201, are also set by setting potentiometers R7 or R6.

The signal conditioning circuit 164, shown in Fig. 4B only in general is shown in detail in the circuit diagram of Fig. 4J. The P / O circuit 168 is on the I / D bus 140 and coupled to CPU 120 for data transfer therebetween to ensure. The PI / O circuit 168 also receives an input signal from the A / D converter 166 via a corresponding one Group of inverters 167. The pressure transducers 51, and PB are connected to the multiplexer via amplifiers 161a, b and c, respectively 163 connected. As shown in Figure 4J, the multiplexer is 163 formed from a corresponding number of relays, energized to apply the selected output of one of the pressure transducers to the A / D converter 166. The PI / 0 circuit 168 determines which of the relays of the multiplexer 163 is to be energized by sending control signals via bus line 169 are given to the logic circuit 165, which consists of a binary-coded decimal / decimal converter and a decoder 165a is formed, which selects in the result which Relay of the multiplexer 163 is to be energized, and on signal going high at a corresponding output

130062/0716

via a group 165b of logic converters to a corresponding one Group 165c of power drivers outputs, whereby the corresponding relay of the multiplexer 163 is energized to the corresponding Apply temperature output to A / D converter 166. Thereafter, the PI / O circuit 168 issues a translation command via line 159 to the A / D converter 166, which then converts the input analog signal into a corresponding digital signal and sends a status signal to the PI / O circuit 168.

Figure 4K shows a circuit diagram of the conditioning circuit 134, shown generally as a functional block in Figure 4C. the PI / O circuit 184 is connected to CPU 120 via I / D bus 140 applied and connected to the counter 182, which consists of two individual Counters 182a, 182b is formed. The signal conditioning and logic circuit 170b in turn contains a series of NOR circuits 172, whose output signal is via a NAND circuit 180 and an inverter is applied to one input of the counter 182a. The output of the linear encoder 26 is via corresponding Logic converter and inverter to said NOR circuits 172 of the signal conditioning and logic circuit 170b created. The rotary encoder 40 of the flow meter is connected to the signal processing and logic circuit 170a, which is similar to circuit 170b and generates a composite Signal for the interval timer 174 and a processed signal corresponding to the signal A for the Interval timer 176. These are designated 174 and 176 in FIG I / D circuits include programmable counters 174a and 176a, respectively. The output of the interval timer 174 is via logic circuit 177 which is a NAND circuit and comprises two NOR circuits, as shown in FIG. 4K, applied to the "Interrupt 2" input of the CPU 120. The output signal of the interval timer 176 is applied to the control circuit of the servo motor 20.

As shown in Fig. 4K, the signal conditioning includes

130062/0716

and logic circuit 170a two inputs, the first of which the output signal A and the second the output signal Ä receives which from the rotary encoder 40 of the flow meter be generated. As can be seen in FIGS. 11A and 11B, the signal A is delayed from the signal A by to enable an indication of the direction of rotation of the encoder 40, as will be explained later. The encoder 40 of the flow meter is designed to rotate clockwise. If shocks or mechanical vibrations occur on it, so can at least temporarily the signal apparently appear before the signal A. Figure 11C shows the output signal A as it appears would if it were 90 ahead of signal A. This condition is undesirable and shows a faulty signal on. The signal conditioning and logic circuit 170a is designed to disable such conditions as will be explained later. The signals A and A are each via level shifters and inverter circuits is applied to a NOR circuit 181a and 181b, respectively. The output of the NOR circuit 181a is connected to an input of the NOR circuit 181b is applied while the output of NOR circuit 181b is connected to an input of a NOR circuit 181c via an inverter. As shown in Fig. 4K, one NOR circuit 181d, the output of which is connected to the other input of the NOR circuit 181c is applied, an inverted, i.e. 180 ° phase shifted signal with respect to the input of the NOR circuit 181a supplied. The resulting output of signal conditioning and logic circuit 170a becomes at the output of the NOR circuit 181c is decreased and has the form shown in Fig. 11D, assuming that the encoder 40 of the flow meter rotates clockwise or in the desired direction; this signal is sent to the interval timer 174 is applied to be counted in the manner explained above. But if even during a relatively If the signal A appears before the signal A for a short period of time, a DC voltage output is applied to the NOR circuit 181c.

130062/0716

signal (or logic 0) received, which 'indicates that signal fluctuations or other error signals are present. The signal conditioning and logic circuit receives in the same way 170b the signals * A and A from the linear encoder 26, these signals are also shown in Figures 11A and 11B. In the same way, the signals A and A are sent to a same group of NOR circuits 169a, b, c and d are applied. The output of NOR circuit 169c is to a latch circuit which is formed from two NOR circuits connected as shown in Fig. 4K. When the signal Ä from the linear encoder 26 is delayed from the signal A, the output signal shown in Fig. 11C is output through latch circuit 169e is applied to the interval timer 176 to be counted by its counter 176a. Another group of NOR circuits is provided in the signal conditioning and logic circuit 170b in order to provide an output signal that is sent to the input 22 of the interval timer 176 is applied to indicate the appearance of the signal A before the signal A, which is derived from the Linear encoder 26 is obtained, which indicates that the piston 14 is moved backwards, so by the servo motor 20 after is moved down to its rest position.

Figure 4L shows a circuit diagram of the signal conditioning and logic circuit 136, shown only generally in Figure 4D. The data is between the parallel data controller (PDC) 192b and the CPU 120 via the I / D bus 140 and via direct connections to the CPU 120. The output of the PDC circuit 192b is coupled to a logic circuit 190b, namely over a plurality of lines shown on the right side of this circuit PDC 192b. The logic circuit 190b is mainly made up of logic converters connected to one output each. A NAND circuit is in the logic circuit 150b to the printer 111a a reset signal to feed. As shown in Fig. 4L, a polling signal is applied to printer 111a, which in turn provides an acknowledgment

130062/0716

ACKO to the PDC circuit 192b, whereby Data to be printed out by the printer 111a at Control of a group of appropriately marked signals can be transferred. In addition, the printer 111a issue a busy signal to prevent data transmission from the PDC circuit 192b to the printer 111a. Further is the SDC circuit 192a via the I / D bus 140 and direct Connections connected to the CPU 120. Its output signal taken from the right side of the SDC circuit 192a becomes delivered to the display device station 112 via logic circuit 19Oa, to supply it with data and to receive data from it, under control of preselected signals that from the associated control circuitry shown on the left. These control signals lead to the fact that the operation of station 112 is determined by fixed signals. The logic circuit receiving the data output signal 190a comprises a logic converter and an inverter circuit in the form of a line driver, while the data input line is connected to the SDC circuit 192a after signal inversion. The clock circuit 193 sets over a logic circuit 195, which is formed from two logic converters, a signal to the clock inputs of the SDC circuit 192a.

The signal conditioning circuit 138 shown generally in FIG. 4E is shown in detail in the circuit diagram of FIG. 4M. The PI / 0 circuit 200 is connected to the CPU 120 via the I / D data bus 140 as well as the direct connections specified above and below. The various outputs of the PI / 0 circuit are coupled through logic circuit 198a to operate the various lights of the control system and display unit 111 activate. Logic circuit 198a includes a logic converter for each output of PI / 0 circuit 200 and one Multiple drivers to control the corresponding lamps. In addition, signals are generated from the switch mechanisms the reset and stop illuminated buttons 111f and 111g via logic

130062/0716

Circuit 198a coupled, in particular to a A group of NAND circuits shown in Fig. 4M, the outputs of which are applied to the PI / 0 circuit 200 through inverters are. A second PI / 0 circuit 196 is across the I / D data bus 140 as well as direct connections to the CPU 120. A group of its outputs is fed to the control of various functions via a logic circuit 194a, specifically the direction and speed of the servo motor 20. The logic circuit 194a receives seven input signals which are passed through a group of NOR circuits and in series Switched inverters and logic converters are applied to a corresponding number of drivers, the output signals of which for this purpose serve to determine the direction and speed of the servomotor 20. Two outputs PI / 0 circuit 196 are also across a logic circuit 194b made up of a series-connected inverter and logic converter is applied to a driver before the control the pneumatic control devices assigned to the valves 34 and 36. A group of five input signals is derived from logic circuit 194c which processes inputs from proximity detectors 50, 52 and 54. The logic circuit 194c comprises a resistor and diode circuit which are connected to the corresponding inputs via inverters the PI / 0 circuit 196 are coupled.

The logic circuit 194d is connected to the logic circuit shown in Fig. 4C 170a and comprises a series of NOR circuits, the input signal being a 1 sec clock signal for calling up the various pressure and temperature signals.

Referring now to Figure 5, there is an overview flow chart of the various program steps stored in the programmable read-only memory PROM 124 and from the central processing unit CPU 120 while evaluating the input into the main memory RAM 126 Data is running. In the initial step 210

130062/0716

Power applied to the computer system. The power supply of the computer system shown in Fig. 3 can, for example, come from components the company Power Mate with the trade names Power Mate EMA 18 / 24B and EMA 12 / 5D or the types Analog Devices-925; Datel-MPS 5/12, MPS 5/3 and MPD 12/3 and Practical Automation-PS6-28 to be delivered. In step 214 the memory locations of the main memory 126 are then cleared to zero, before an initiation subroutine 300 begins, by which the various parts of the computer system are initiated, as detailed later with reference to FIGS. 6A and 6 6B will be explained. At various points in the program, a return to the step takes place via entry point 212 214 to start the program again. As shown in Fig. 4B there is a switch 139, the setting of which determines whether parts of the calibration system 10 are to be calibrated or a flow meter check is to be performed. If the switch 139 is in its "calibration" position, so there is a transition from de-sexing step 216 to step 400, in which a subroutine is executed to access the various analog inputs from the temperature and measuring devices 1 and 2A as well as the corresponding A / D converter to which these signals are applied, as in detail will be explained with reference to FIG. In moving the system to step 500, the user can Data from the various measuring devices, e.g., temperature measuring devices 42, 44, 48, 57 and pressure measuring devices 51 and 46 and call up the output signals of the linear encoder 26. In addition, the user can also perform various self-tests of the measuring device. This subroutine will be explained in detail with reference to Figs. 8A to 8P. After access for the appropriate data, the program goes to step or subroutine 900, in which a test or a series of Tests are carried out on a flow meter 38 by means of the calibration system 10 and the test results on the display device displayed or recorded by the printer. The subroutine 900 is explained in detail with reference to FIGS. 9A-Q.

130062/0718

Referring now to Figures 6A and 6B, the initiation subroutine 300 is shown. in the At step 302 a command is sent over I / D bus 140 which causes logic circuit 198a in Figure 4E to turn on the ready light 111e turns on. At step 304, a scale factor corresponding to a test volume of 28 liters (1 cubic foot) corresponds, from the main memory 126 to the programmable Counter 174a in interval timer 174 of Fig. 4C is transferred to the output of the flowmeter encoder counter 179a in the correct relation, whereby when counting the correct number of pulses, e.g. -40,000, the logic circuit 177 emits a pulse which indicates that the fluid volume specified above is flowing through the flow meter 38 Has. In step 306, the interrupt programs of the CPU 120 are then enabled, so to any later-time of the program these program interruptions can be executed if the user e.g. the reset push button 111f or presses the stop pushbutton 111g. At At this point in the initiation subroutine 300 are the program interruptions associated with pushbuttons 111f and 111g not yet released; however, after step 306 is performed, they are available for execution. Then, at step 308, that shown in FIG. 4C becomes the rotary encoder 40 of the flow meter associated circuitry initiated. In particular, the logic circuit 170a initiated, interval timers 174 and 176 are cleared, PI / 0 circuit 184 is set to the selected mode of operation set, the logic circuits 177 and 178 are reset, and programmable counters 174a and 176a programmed with the factors to be entered and counted down. In step 310, the PI / 0 circuit 196 a control command is issued to the logic circuit 194b to cause the second or exhaust valve 36 to open. In particular, there is access to the proximity detector associated with valve 36, and if valve 36 is not open, the program waits until it is opened. In step 310

130062/0716

the control circuit assigned to the servo motor 20 is initiated, the speed of the servo motor 20 increasing to Is set to zero and a signal is applied to it to hold it in place while a logic circuit is used to 'feel it' of the torque of this servo motor 20 is reset. This logic circuit determines the current flow through the servomotor 20. In step 312 the logic circuit 19 2b is on their static input / output operation initiated, with "acknowledgment operation" output option, and the logic circuit 190b is deleted and prepares the assigned printer 111a to start printing. In step 314 the Circuit SDC 192a programmed so that between the station 112 and CPU 120 can transfer data, and logic circuit 190a in FIG. 4D similarly comprises the Station 112 to clear their contents and preparations to receive data and clear the cathode ray tube display. Then in step 316 any data stored in the A / D converter 166 in Fig. 4B is erased, and the multiplexer 163 becomes on its first channel adjusted, whereby the output signal of the converter 51 over the multiplexer 163 is applied to the A / D converter 166. In step 318 the rendering and interface circuit 130 is erased, and more specifically, the data stored in the A / D circuit 158 is erased and becomes the Multiplexers 149 and 154 are set to their respective first channel, thereby reducing the output voltage of the temperature converter module of the calibration device via the multiplexers 149 and 154 and the operational amplifier 152 to the A / D converter circuit is created. In step 320, command signals are sent over the I / D bus 140 and are long-lasting to the PI / O circuit 196, which is shown in Fig. 4E to cause the logic circuit 194b to drive the coil 66 so that the first or valve 34 assigned to the flow meter is opened and then iji-HichlG «smi is, and to din SpuU» (.4 «iir / UHtou ^ rn, druiiH this closes the second or outlet valve 36. If one

130062/0716

- sr -

the valve 34 or 36 is to be opened or closed, the assigned proximity switch is queried; if he is in the desired state, no further action takes place. However, if it is not in the desired position, it delivers the logic circuit 194b an output signal in order to control the associated coil of the pneumatic control, thus the valve can be opened or closed as desired. In step 322 the program goes to a subroutine based on 6B, whereby the piston 14 is returned to its original position, i.e. its lowermost position which corresponds to the state of the proximity detector 50 shown in FIGS. 1, 2A and 2B.

In Fig. 6B is the subroutine 322 for returning the KoIr bens 14 shown in its lowest position. In step 324, a control signal is sent through the I / D bus 140 and the PI / 0 circuit 196 is sent to the logic circuit 194b in order for it to have the Closing of the first or flow meter valve 32 causes. Then, in step 326, the position of the piston 14 determined, which can be anywhere between its lowest and uppermost position in FIG. In particular causes the step 326 that the PI / O circuit 196 the logic circuit 194c interrogates to determine whether the output of the proximity detector 50 shown in FIGS has a high level or the value 1; if so, the subroutine proceeds to final step 346 where an instruction is on the PI / O circuit 196 in Figure 4E is sent to the logic circuits 194a and 198b so that the motor control of the servo motor 20 to the speed zero is set, and the interval timer 176 in Fig. 4C is set so that the output of the Encoder is in the initial state, i.e. 50 that the counter 176a is set to zero so that it is ready to begin generating a 1 sec polling pulse. If the piston 14 is not in its initial position, goes

130062/0716

the subroutine to step 328, in which it is determined whether the piston 14 is in an intermediate position, ie whether the stop 92 is located between the proximity detectors 50 and 52 in FIGS. 1 and 2A and 2B; if the piston / line is in such an intermediate position, the subroutine goes to step 330, whereby control signals are sent through the PI / O circuit 196 which set the logic circuit 194a so that the servo motor 20 is rotated counterclockwise; in step 332 the speed is also regulated so that a selected speed is reached, for example the ninth speed out of sixteen possible speeds. If, however, it is determined in step 328 that the piston ^ 14 is in its uppermost position, ie its stop 92 is above the proximity detector 52, then in step 334 the logic circuit 194a in FIG is rotated counterclockwise at the lowest speed, ie, speed 1 of the sixteen possible speeds, before proceeding to step 336. In this step, the logic circuit 194a accelerates the servomotor 20 to its next higher speed up to a maximum of the twelfth speed. At this point, in step 33.8, the logic circuit 194c is accessed to determine whether the piston 14 is on the proximity detector 52; If this is not the case, then in step 336 the piston is accelerated to its next higher speed until it is determined in step 338 that the piston 14 is located at your proximity detector 52. At this point, the logic circuit 194a then maintains the present speed of the servomotor 20 until, in step 34 2, the proximity detector 50 detects the starting position of the piston 14. Then in ri-hjitt · 344 the J.oijikiJchiiH uimj 194 a fio oiriyi ^ itoll t that the piston 14 is accelerated to zero before step 346 begins, in which the engine control is set to the speed zero.

130062/0718

When the "calibration / operation" switch is set to the "calibration" position, the program goes to subroutine 400 which is shown in Fig. 7 in detail. Initially, in step 402, a control command is sent to logic circuit 190a in FIG. 4D, to clear station 112. In step 404 an appropriate message is displayed on the screen, which indicates that the flow meter calibration system 10 is in a calibration mode has occurred, wherein - as shown in Fig. 7 - the various parameters are displayed, the so can be calibrated. Then in step 406 the runner displayed on the screen is returned to the left margin, to wait for the appropriate values to be entered by the user. After entering them via the keyboard of station 112 the keyboard is queried at step 408 to determine which keys were pressed. If, for example, it was determined that one of the keys T and 0, T and 1, T and 2, T and 3, T and 4, T and 5, T and 6 or T and 7 has been pressed, the subroutine goes to step 410, in which a corresponding channel is selected by the multiplexer 149. Then in step a select command through the PI / O circuit 160 to the multiplexer 149 sent, whereby the selected channel via the multiplexers 149 and 154 and the amplifier 152 to the A / D converter 158 is applied. Thereafter, in step 414, a conversion command is issued to the A / D conversion circuit 158, wherein in step 416 the measured temperature is displayed. At this point the user can select the selected temperature converter module Calibrate so that it shows the correct reading by means of the zero point and display range of the operational amplifier of the module, this process being repeated until the correct one appears on the screen Reading is displayed. A detailed explanation can be omitted, because it can be seen that in the same way Keys P and 1, P and 2, P and B can be pressed, increasing the gain of the operational amplifiers 161a, 161b and 161c in Fig. 4B can be set in the same manner to get accurate readings. In the same way

130062/0716

The buttons T and O, T and 1, T and 2, T and 3, T and 4, T and 5, T and 6, T and 7 can be pressed, and their associated amplifiers and circuits can be set to the correct ones To give readings. Upon depressing the V and P keys, determined in step 408, the program proceeds via step 450 to step 452, whereby a control signal is sent through the PI / O circuit 168 in FIG applies the calibration voltage signal V to the A / D converter 166, which converts this input signal into digital values in step 454. The exchange voltage signal is dimensioned in such a way that the zero point and maximum input range of the A / D converter 166 are set, these values being displayed in a suitable manner at the station 112 in step 456. Likewise, if the keys V and T on the keyboard are pressed, the subroutine proceeds via step 442 to step 444 wherein the voltage signal VT is applied to the A / D adapter 158 in FIG. 4A so that it is. 'sspn zero point and riinganqslioreieh can be set in the same way. If any other key on the keyboard is pressed, step 408 branches via step 456 to step 458 in which the entered quantity is discarded in order to return to the starting point of step 408.

The data input and sorting subroutine shown roughly as step or subroutine 500 in FIG. 5 will now be used on the basis of this 8B to 8P explained in detail, wherein an overview of the program 500 is given in FIG. 8A. At first, a command is sent to the logic and driver circuit via the I / D bus 140 198a in Fig. 4E to turn on the ready lamp 111e. Then, at step 504, shown in FIG. 8C shown in detail is the determination of whether a self-test is desired; if so, the subroutine will executed. It is then checked in step 506 whether the room temperatures should be displayed. If applicable, the

130062/0716

- -Si "-

6H

Display on the screen of the station 112, as in detail based on of Fig. 8H. In step 508 the keyboard is checked to see if it is desired to retrieve data relating to the preceding Flowmeter accuracy tests include; when such data is requested, it is displayed on the screen displayed. The subroutine 508 is explained in detail with reference to FIG. 8J. In step 510 the appropriate identification signals (flag) set automatically so that the to be monitored at a later point in time during the flow meter test Pressure and temperature parameters are displayed. The subroutine for executing this display is detailed with reference to of Fig. 8K. The subroutine corresponding to step 512 for setting the test volume in order to like this in the programmable counter 174a of the interval timer 174 is entered, are based on of Fig. 8L. The user must then enter the selected flow rate using the station keypad, as shown by 8M is further explained. In step 516, the user sets the number of repetitions of a particular Tests a. For example, the flow meter at a given Flow rate or a given volume can be checked three times. The input of the repeat commands is based on FIG. 8N explained in detail. Thereafter, in step 518, delete commands are sent out via the logic circuit 190a, whereby any Data stored in station 112 buffers are to be deleted. In step 520 the decision is then made whether the volume of the flow meter 38 exceeds the Output signal of the proximity detector 27 or via the signal of the rotary encoder 40 is to be determined. If the Switch is set to logic "0", in step 522 an identification signal (flag) is set in order to carry out a test by means of the encoder while the switch is in position an identification signal is set to logic "1" in step 524, which indicates that the check is carried out by means of the proximity detector causes.

130062/0716

(P

As program 500, shown in FIG. 8B, continues, the system receives the temperature and pressure readings from the flow meter and shows them on the screen, as well as the temperature and pressure of the calibration device, which during the test process can be obtained in the manner described below. Then, in step 528, data is displayed on the screen on command, what pressure and temperature in the flow meter 38 and in the calibration device 10 specify. Once the test volume, namely 28, 14, or 7 liters (1 cubic foot, 1/2 cubic foot, or 1/4 cubic foot) is above the Station 112 is entered, the entered value is decoded in step 530 and the selected value is used in step 532 displayed across the screen as: TEST VOL. = x.l (resp. χ CU. FT.). Thereafter, the desired flow rates entered on the keyboard are decoded in step 534, and the selected values are displayed on the screen at step 536 as: FLOW RATES are QX, QX-QX. In the same Thus, in step 538, the number of repetitions of any test or tests for flow rates or volumes are decoded and in step 539 the number of repetitions per flow rate or volume is displayed on the screen displayed as: NUMBER OF TESTS PER FLOW RATE = X. In step 540 a runner return line command is then executed, thereby removing a runner applied to the screen by the viewing station.

The subroutine 504 is shown in detail in FIG. 8C, wherein the desired self-test or self-tests to be selected. Initially in step 542 the data storage buffers of the screen of the station 112 are cleared, by commands generated by logic circuit 190a. At step 544 the screen shows as in Fig. 8C indicates the various self-tests that can be performed, e.g. Volume V, Leak L or NO. The user selects one of the self-tests by pressing the appropriate key on the keyboard. In step 546 the runners flash as in FIG

130062/0716

8C to prompt the user to select one of the Select self-tests, namely Volume V, Leak L or NO. In step 548 the routine goes to the one desired Self-test, depending on which key was selected in step 544. For example, if the volume self-test V is selected, then A volume self-test sequence is output to the screen in step 552, and the volume self-test is carried out in step 554 carried out, as will be explained in detail with reference to FIG. 8E. If the leak test is selected, the subroutine goes over Step 558 so that the selected test is displayed on the screen and the leak self-test is initiated, which is carried out in the Step 560 is executed, which will be explained in detail with reference to FIG. 8D. Then the subroutine determines in step 556, whether there was any other input that needs to be processed, and the subroutine returns to step 546th. If the user presses the "exit" button, which is determined in step 548, the subroutine is via step 562 and go to entry point 212 of the main diagram in Fig. 5, with which the program begins. If the user presses the N and 0 keys, which is determined in step 548, the subroutine is exited via step 566 and returns to the program of FIG. 8A to go to the next step 506 to continue. If any other key is pressed, the routine exits via step 568 to open the to discard entered value in step 570 and return to step 548 so that another key is recognized will.

The leak self-test subroutine 560 is shown in detail in Figure 8D. It is a check to see if the Seal of the calibration system is perfect. Initially, in Figure 4E, in step 572, logic circuit 198a turns the lamp on 111b "Self-test" and 111c "Self-test sequence". Thereafter, the subroutine 322 is executed to put the piston in its Return to the starting position, as above with reference to FIG. 6B

130062/0716

-ÖD

was explained. Then in step 5 74 is via the PI / O circuit 168 a command signal is issued to set the multiplexer 163 to receive the signal over the second channel from the pressure transducer 51 (pressure PP1). Then the logic circuit 194b is controlled to receive actuation signals is applied to the coils to close inlet valve 34 and outlet valve 36. In step 578, the PI / O circuit 196 issued a command to the logic circuit 194a to instruct in Fig. 4E to control the servomotor 20 so that this the threaded spindle 18 in a clockwise direction with relative able to drive slower speed corresponding to "1". The servo motor 20 cannot be switched on via step 578 until a lock built into the program is removed by temporarily depressing a switch (not shown) which is coupled to the logic circuit 194c so that the operation of the servo motor 20 is enabled, when both valves 34 and 36 are closed. Normally, the servo motor 20 cannot operate when the valves 34 and 36 are closed to prevent damage to the piston seal. Then, in step 580, an A pressure reading is input to the A / D converter 166 and a check is made in step 582 to see if the pressure is around a certain value Amount has increased, which is indicated by the data stored in the A / D circuit 166, i.e. it is checked whether the vacuum increase has exceeded 0.0033 bar. If this: is not the case, the subroutine runs back and repeats the step 580. However, if the vacuum in the chamber 28 has increased by more than 0.0033 bar, the subroutine goes to step 584, whereby via the PI / 0 circuit 196 a command is issued to stop the servo motor 20. In step 586 the command "wait 20 seconds" entered into a register in the working memory 126 and counted down to time for a leak to be detected to be made available to Chamber -28. Then, in step 588, a conversion command is issued to the A / D converter 166 in FIG. 4B, in order to determine the pressure, which is now determined by the pressure transducer 51. In step 590 the subroutine waits for that

130082/0716

1 sec clock signal generated by logic circuit 178 in FIG. 4C is generated and applied via the logic circuit 194d in Fig. 4E, in step 592 the 20-see count rate which is in the Register in the main memory 126 was introduced to count down. Step 592 determines whether this register has been decremented to zero in memory 126; if not, the routine returns to step 588, whereby for every second of the 20-sec interval the piston pressure is above the transducer 51 is obtained and displayed on the screen, so that a continuous monitoring of the pressure in the calibration device is enabled so that the user can determine if any leaks within the chamber 28 persist during the 20 see Tests occur. The countdown continues until the register is zero, which is detected in step 592, where then at step 594 a command is issued to cause logic circuit 194b to operate the solenoid to open the exhaust valve 36 controls. The position of the piston 14 is then determined in step 596 by checking the status of the proximity detectors 50 and 52 is checked by accessing logic circuit 194c. The test is then complete and a command is given to logic circuit 190a to cause the indications "Test completed" and "Self-test" are activated. Finally, in step 322, the piston 14 is in its starting position fed back via that in Fig. 6B before returning to step 556 shown in Fig. 8C.

At step 8E, the volume self-test subroutine 554 is shown in detail. First, step 600 has the Logic circuit 198a to activate the display 111b "Self-test" and 111c "Test in progress". The volume self-test will entered to gain an indication of the accuracy of the linear encoder 40 2} u. By step 332 shown in FIG. 6B the piston 14 is returned to its original position before command signals in step 602 via logic circuit 194b

130062/0716

output to control the coils so that the inlet valve 34 is opened and the outlet valve 36 is closed. A command is then given in step 604 to save the speed buffer to load, i.e. an addressed part within the working memory 126, at its maximum speed 16. Then in step 606 the servo motor 20 is accelerated to the speed stored in this buffer, as will be explained in detail with reference to FIG. 8F. Then, in step 608, the counter 182 and the programmable Counter 174a in Fig. 4C locked and reset to zero. In step 61, the state of the proximity detector 5 becomes 2 checked to see if the piston 14 is in its original position was brought back; if not, the test is repeated until the piston 14 has reached this position. When the piston 14 has reached its home position, the program proceeds to step 612 in which the programmable counter 174a is disabled and the counter 182 is enabled to count the pulses from the linear encoder, while the piston 14 from proximity detector 52 to proximity detector 54. At step 614, the proximity detector status becomes disabled 54 checked periodically until the piston 14 is there, at which point in time the subroutine goes to step 616 to disable the linear counter 182. Then in step 618 the encoded number or count rate of the counter 182 for the linear encoder via the P / O circuit 184 promoted to a specified location within memory 126. In step 620 an appropriate sequence of Command signals are given to the logic circuit 194a, whereby the servo motor 20 is braked and stopped, as in detail will be explained with reference to Fig. 8G. Then in step 622 identification signals (flags) are set which indicate piston 14 is not located between proximity detectors 50 and 52. In step 624 the logic and activates Driver circuit 198a lights "self-test" and "test completed" 111b and 111d, respectively. At this point, the step 626 the count rate of the counter 182, which the stroke of the piston

130062/0716

14 between the proximity detectors 52 and 54, transferred from their place within the main memory 126, to be displayed on the screen of the station 112 and at the same time the counter 182 is cleared. Then the Piston 14 is returned to its original position by step 332, as shown in FIG. 6B, and by step 628, the logic and driver circuit 198a is caused to turn on the ready lamp 111e before going to step 556 is returned in Fig. 8C.

The engine acceleration subroutine shown in FIG. 8E denoted by reference numeral 606 is in detail shown in Fig. 8F, wherein in step 630 a command is sent over the I / D bus 140 to set the counters 182 and 174a. to lock in Fig. 4C. Then, in step 632, an enable command is issued to logic circuit 194a to operate the servomotor controlled so that it rotates clockwise, whereby the piston 14 is raised »At this point then the first speed, i.e. the lowest speed, in the speed comparison buffer within the main memory 126 entered via step 634. In step 636, a command is issued to logic circuit 194a to increase the speed of the servomotor 20 in steps, and the respective speed that is stored in the speed comparison buffer of working memory 126 is compared to the top speed signal shown in is stored in the speed buffer; if they are the same, then the servo motor 20 to the desired speed is accelerated, the subroutine goes to step 646. Otherwise, a time of approximately 0.25 see waited before the logic circuit 194c of the State of the proximity detector 52 is scanned. If this position has not been reached, then in step 644 the desired Increase speed by 1 before returning to step 636 where the new speed is stored in the speed comparison buffer of the main memory 126 introduced

130062 / 071S

will. When the piston 14 has been brought into its intermediate position, which is determined in step 640, the subroutine continues to step 642 to set a flag indicating that the piston 14 is early in its Is intermediate and the routine returns to step 604 shown in Figure 8E. As in Figure 8E As shown, the piston 14 is accelerated to the desired speed in normal operation and then periodically the output of the proximity detector 52 is checked. If it is determined in step 646 that the piston 14 has reached the test start position, as determined by the output signal of the proximity detector 52 is displayed, the subroutine branches off to step 9 70, as shown in FIG. 9E explained is to indicate the occurrence of the rising edge of the next pulse A from the rotary encoder 40 of the flow meter determine, whereby the test of the flow meter begins by the output pulses of the encoder 40 and the Linear encoder 26 are applied to the corresponding counters 174a and 182, respectively.

The subroutine for braking the servo motor 20, which indicated at step 620 of Fig. 8E is shown in detail in Fig. 8G. In step 648, the counters 174a and 182 locked in Figure 4C and then reset to to receive the output signals of the rotary encoder or linear encoder. In step 650, the speed comparison buffer loaded within memory 126 at speed 16, i.e. the highest available speed. Then in step 652 the actual speed, which is stored in the buffer of the working memory 126, compared to the speed 16 stored in the speed comparison buffer was loaded, and if these are not the same, the speed comparison buffer loaded speed is decreased until the actual speed equals that in the speed comparison buffer invited speed is; to this

130062/0716

Time the subroutine goes to step 656, and the The speed stored in the speed comparison buffer is decremented. In step 658, a Command issued to logic circuit 194a to decelerate servomotor 20 to this new speed, which is introduced into the speed buffer of the working memory 126. In step 660 the actual speed compared with zero, i.e., the servo motor 20 is stopped. If it has not yet been stopped, it will be in step 662 waits 0.2 seconds before engine speed is decreased again in steps 656 and 658. This process continues until the servo motor 20 stops i.e. its speed is zero, whereupon step 660 leads the subroutine to step 622 in FIG Fig. 8E brings back.

In Fig. 8H the subroutine 506 is shown, via which the user by means of the keyboard of the station 112 the selected Entering temperatures within the calibration room 106 so that they are displayed, this subroutine also includes the steps of displaying them on the screen. It has been found that knowing the temperature within of calibration room 106 is useful to find out how well the flow meter calibration device 10 within the space 106 can be controlled. The subroutine shown in Fig. 8H enables the user to display all four room temperatures RO, R1, R2 and R3 using the temperature measuring devices in 2C, which are arranged distributed over the calibration space 106. These temperatures can be continuously called up and displayed on the screen of station 112 until the user presses the N and 0 keys on the keyboard. In step 664 initially a command is issued to the logic circuit 190a, whereby the buffer assigned to the station 112 is via the control input to be deleted. Thereafter, step 666 causes logic circuit 190a to cause a display as shown in FIG. 8H, an indication of the available possible ones

130082/0716

- 64 -

Room temperatures by accessing the room temperature measuring devices RO, R1, R2 or R3. If the Benjtzer doesn't If you want access to these temperatures and do not want to display them, the keys N and 0 are pressed. In step 668 a runner return line command is issued so that the runner flashes to indicate that the user has responded is required. In step 670 an inquiry is made as to which key was pressed, to be precise via logic circuit 190a. If the the room temperature measuring device RO corresponding keys R and 0 have been actuated, the subroutine proceeds via step 672a to step 674a, whereby a control signal over the I / D bus 140 and via the PI / 0 circuit 160 is output to the To cause multiplexer 149 to select its fifth input, creating a voltage which is that indicated by converter RO Represents the room temperature value via which multiplexers 149 and 154 are applied to the A / D converter 158. In step 676a, a conversion command is sent via the PI / 0 circuit 160 the A / D conversion circuit 158 is applied to the analog input signal is converted into a binary digital signal. Thereafter, in step 678a, the binary digital signal in Decimal data is converted and, via step 68Oa, that data is converted into decimal format using well known techniques implemented with floating point. In step 682a, the floating point digital decimal data is converted to digital Temperature display according to the formula:

r (INITIAL DIGITAL NUMBER) (7.326007326 x 10 ~ ³ ) (9) J _{+ 32} = ⁰ F

This formula applies to the temperature display selected in one embodiment in F.

The relationship results for the display in ^{0 C.}

(INITIAL DIGITAL NUMBER) (7.326007326 x 10 " ³ ) _ O _n

130061/0716

In step 684a, suitable data is generated and output from logic circuit 190a in order to display the room temperature on the screen in the desired units ( ⁰ F or ⁰ C). In the same way, step can be used to determine when the user presses two keys corresponding to the room temperature detectors RO, R1, R2 or R3 in order to gain access to these detectors which are connected to the multiplexer 149 in FIG. 4A , and to process the selected data and display it on the screen. At step 694 it is determined whether any other input is desired and flow returns to the beginning of step 670. If the user presses the exit key, the program goes through step 686 to return via entry point 212 to the beginning of the program shown in FIG. If the N and 0 keys are depressed, the program proceeds to step 688 and returns to the next subroutine or step 508 in Fig. 8A. If an incorrect key is pressed, the subroutine is exited via step 690 to discard the entered value in step 692 before returning to the beginning via step 670.

The subroutine 508, shown generally in FIG. 8A, for retrieving data from the tests previously carried out is shown in FIG carried out in the manner shown in Fig. 8J in detail. The subroutine can begin after a flow meter check was executed. First, in step 696, the display station, in particular its buffers and display. unit, deleted or cleaned of previously stored data. Thereafter, step 698 is followed by activating the logic circuit 190a makes the cathode ray tube display the indicates possible test parameters to which access can be obtained and which can be displayed, namely: K, i.e., the number of counts of the linear encoder 26; TP, i.e. the temperature of the calibrator; TM, i.e. the temperature

130062/0716

door of the flow meter, the pressure PM of the flow meter; the pressure PB of the calibrator; and the percentage of errors from the previous test. If none of these values should be displayed / the user can press the N and 0 keys corresponding to the command NO (no display), after which the program advances to the next step. In step 7 the cursor flashes and indicates to the user that one of these parameters should be selected or NO must be entered by pressing the appropriate keys. If the count rate of the linear encoder is called up by pressing the K key, then in step 702 the subroutine branches off to step 704a to arrive at step 706a, whereby the count rate of the linear encoder register is read out within the main memory 126 and sent to the display station via the logic circuit 190a is created and displayed on it. If the temperature of the calibration device is called up via the keyboard, the sequence goes from step 702 via step 704b to step 706b, in which the content of the calibration device temperature buffer, or an assigned location within the main memory 126, is transferred from this to the display station and is displayed there. In the same way, the temperature TM of the flow meter, the temperature TP of the calibration device, the pressure PB of the calibration device or the pressure PM of the flow meter can be taken from the corresponding memory location within the main memory 126 and displayed on the screen. Furthermore, the percentage error of the difference between the standard volume that has flowed through the calibration system 10 and the volume actually measured by the flow meter 38 is calculated with reference to this standard volume and can be displayed in the same way. If a second or more key has been pressed, step 716 returns the subroutine to the beginning of very itto 700. If the Ta "te" V ^ JiIausuti "was pressed, dna subroutine UELX" goes r step 708 to the starting point 212 back at the beginning of the program shown in Fig. 5. When a wrong key is pressed

130062/0716

the program proceeds to step 712 to determine the faulty Discard input value in step 714 before returning to step 702 occurs. After the operator has viewed all of the desired parameters on the screen, press the keys N and O, after which the program returns to the next step or subroutine.

Program 510, shown generally in FIG. 8A, is capable of accessing and displaying the parameters automatically measured and monitored in the course of a test; it is shown in detail in Fig. 8K. First, the display station associated buffer in step 718 deleted before in steps 720, 722, 724 and 726 characteristic signals (flags) at the appropriate locations within the working memory 126 ^- are set to gain access to the temperature values and pressure values of the flowmeter and of the To obtain calibration device. That is, the flags are set so that these parameters are automatically displayed on the display station, and then the subroutine 510 returns to step 512 shown in FIG. 8A.

Step 512, shown generally in FIG. 8A, for setting the desired volume which is to flow over the flow meter 38 is shown in detail in FIG. 8L. First, in step 742, the display station 112 is cleared before the possible volumes are displayed thereon in step 746, for example 28, 14 and 7 liters (1 , 0.5 and 0.25 cubic feet, respectively) that are available to the user and over the flow meter to be tested 38 can flow. At step 748 the cursor blinks to prompt the user to make the volume selection. When the key corresponding to the desired volume is actuated, the subroutine 512 proceeds to input the corresponding volume. For example, if the key indicating entry of a volume value of 28 1 (1 cubic foot) is selected, the routine proceeds via step 752a to step 754a wherein the

130062/0713

Scale factor equal to the value 28 1 (1 cubic foot) into a known location of the main memory 126 is entered. The scaling factor is then taken from this memory location transferred to counter 174a of the rotary encoder. After selecting the required test volume, the subroutine returns to the next step 514 shown in Figure 8A. If the "Exit" button is selected, that works Subroutine returns via step 756 to the main program at its starting point 212. If you mistakenly press the wrong key is pressed, the routine goes to step 758 and discards the entered size in step 760 before returning to start with step 750.

As the next step, the program contains the subroutine. 514, in which the desired flow rate with which the fluid is to flow over the flow meter 38 is set. As previously mentioned, the selected flow rate in turn indicates the speed at which the servo motor 20 is controlled to raise the piston 14 in FIG. Reference is now made to Figure 8M. In step 762 the buffers associated with the display station 112 are initially cleared, and in step 764 the possible flow rates from QO to QF are displayed on the display screen, where QO corresponds to the minimum flow rate, e.g. 566 liters per minute (20 cubic feet / min), while QF corresponds to the maximum flow rate, e.g. 11326 liters per minute (400 cubic feet / min). At step 766 the runner is flashed to prompt the user to dig into the flow rate. At step 768 the keyboard is queried to determine which key was pressed by the user. If, for example, the flow rate QO has been selected, the subroutine is exited via step 77Oa in order to arrive at step 772a, in which the identification signal (flag) for the flow rate QO is entered into a designated location in the working memory 126. Thereafter, step 77a causes the selected flow rate QO to be displayed on the screen.

130062/0716

The system then responds with "MORE" and the runner flashes to indicate that a response from the user is required is. If the test is to be performed at a different flow rate, step 784 leads the program to Return to the beginning of step 766, which provides a second and possibly additional flow rates within a single test group can be checked. If the user presses the "exit" key, the program proceeds through step 776 to the entry point 212 in the main program shown in FIG. If an incorrect key is pressed on the station 112 keyboard the program proceeds via step 780 to step 782 where the incorrect entry is discarded before proceeding to the initial step 768 returns. If no further flow rates are selected at which the flow meter 38 will be tested should, the user presses the keys N and 0, whereupon the program returns via step 778 to the next step 516, which is shown in Fig. 8A.

The step or program 516 for setting the number of repetitions of tests to be carried out for each selected flow rate can be entered via the keyboard as shown in detail in Fig. 8N. Initially, in step 786, the display station 112 buffers as well as its display screen is cleared before the possible number of tests or test repetitions per flow rate, e.g. 1, 2. or 3, is displayed on the display screen. At step 790, the runner flashes on the display to prompt the user to make the choice of flow rate. In step 792 it is determined which key of the station was pressed and if so E.g. a key has been pressed indicating that only one test should be performed per flow rate, so goes the subroutine via step 794a to step 793a in which the output "1" is displayed on the screen, before returning to the next step 518 of the main routine in Figure 8A. If two tests are to be performed, the subroutine is exited via step 794b to go to

130062/0716

To go to step 793b in which a command is given to copy the flow rate signal flags for repetition, and in step 796b is commanded to be a flag (flag) to be set for a second repetition in the assigned area of the main memory 126. In step 796b, the Display "2" on the screen. When a selected test is repeated three times is to be, the subroutine performs a similar sequence of steps 794c, 793c and 796c. If a wrong one Input was made on the keyboard, the program proceeds via step 795 to step 797, in which the incorrect input is discarded before returning to initial step 792. When the user presses the exit button, the program exits Exit via step 798 to get to starting point 212, thereby returning to that shown in FIG Main program takes place.

Subroutine 528, shown generally in FIG. 8A, displays the temperature and pressure data while the flowmeter test is in progress, and this is shown in greater detail in FIG. 8D. Initially, the carriage of the cathode ray display tube is moved to a starting position and the CPU addresses the location within the working memory 126 where the TM flag was automatically set during the previous subroutine 510. Since this identification signal TM has been set, the content of the flowmeter temperature buffer, ie an addressable location within the working memory 126, is transferred via the logic circuit 190a to be displayed on the display screen, as indicated in step 802. After the display, the subroutine continues to step 806, where that location of the working memory 126 is checked where the TP output flag is set, which indicates that the temperature TP of the calibrator should be monitored during a test, and that from the The signal output from the calibrator device buffer is converted to be displayed on the screen. After the display, the subroutine continues to step 810, in which access to the previously set PM print

130062/0716

Identifying signals and output signals of the content of the print buffer or the corresponding location in the working memory 126 is made to be displayed on the screen. After the display, access occurs in step 814 to the previously set TP output buffer identification signal, and the content of the calibrator pressure buffer is transferred to to be displayed on the screen. After this information is displayed, the subroutine returns to the next step 530 back in Fig. 8B. The values output at this point in time correspond to those currently in the buffers of the main memory stored values. If a test has not yet been carried out, zeros are first stored in the main memory 126 and output and displayed as TM - 0.0 Deg. F. In this program step the display is brought into such a format, that at the beginning of a test the system replaces the zeros with the appropriate numbers.

Performing the checks on the flowmeter 38 by means of the calibration system 10 and the output of the results via the printer and the screen take place via the in 5, the program, generally designated 900, will now be explained in detail with reference to FIGS. 9A-9P will. Referring to Figure 9A, there is shown a high level diagram of the various steps or subroutines required to in order to carry out the test of the flow meter 38, starting with step 323, as was explained above with reference to FIG. 6B, to return the piston 14 to its original position, i.e. to its normal lowermost position in the cylinder 12, shown in Figures 1 and 2A. Thereafter, in program 902, as will be further explained with reference to FIG. 9B, the type of check is determined, i.e. whether the volume measured by means of the flow meter 38 is determined by the proximity detector 27 or by the optical rotary encoder 40 in Fig. 4C is to be determined, and also the volume which is to flow through the flow meter 38 is determined. In particular, in step 902 the appropriate volume

130062/0716

factors are entered into the programmable counter 174a and the appropriate divisor is stored in the buffer EORP of the working memory 126 saved. Thereafter, in step 904, the condition is determined that one check per flow rate is to take place and the flow rate applicable to this check is determined. In particular, that addresses System indicates the location in which the flow rate characteristics are stored within the working memory 126 in order to obtain the determine the appropriate flow rate to be used in the particular test. After the flow rate up is located this way, the program proceeds to step or subroutine 906a to determine what flow rate should apply, and to obtain this flow rate and execute this subroutine, as in particular in the Figures 9C and 9D are shown. If it is determined in step 908 that two tests per flow rate will be performed are to be, the program addresses those locations in the main memory 126 which correspond to the corresponding flow rate characteristic signals and performs the checks at the programmed flow rates in step 906b. If three Checks per flow rate are desired so will first, at step 912, determine whether the appropriate three flow rates have been entered, and if so, so the flow rate characteristic signals are called up from the specified locations within the main memory 126 in step 906c, and the checks of the flow meter 38 are made at these flow rates. After that there is a command sent to logic and driver circuit 198a to activate the ready indicator before returning to the Step 216 in FIG. 5 occurs.

The determination of the test type and volume to be measured is shown generally as subroutine 902 in Figure 9A and will be explained in detail with reference to Fig. 9B. This initially contains the programmable counter 174a and the counter

130062/0716

182 of FIG. 4C is locked and reset in step 918. At this point, i.e. in step 919, the state of the Encoder / proximity detector switch to decide whether the volume of fluid flowing through the flow meter 38 determined via the output signal of the proximity detector 27 or the output signal of the Rotatxonscodierers 40 shall be. If the rotary encoder is selected, the subroutine goes to step 920 to determine the presence or the State of the 28 1 (1 cubic foot) identification signal on the assigned Place in the working memory 126 to check. If it is present, it will be the factor corresponding to that volume stored as the number of output pulses of the optical rotary encoder 40 in step 922 at the location EORP of the working memory, i.e., in its assigned location, and the factor number is loaded into the programmable counter 174a. At this point the subroutine returns to the next step 904 of the program in Figure 9A. If the 28 1 (1 cubic foot) signal is absent, step 924 tests the condition of the 14 1 (0.5 cubic foot) flag; if there is is what is determined in step 924, then the programmable counter 174a of FIG. 4C in step 926 with a Counting rate is loaded which is equivalent to this volume, and the corresponding factor is stored in the buffer location EORP of the working memory 126 before returning to step 904 in Fig. 9A. If neither the identification signal for 28 nor that are set for 14 1, then step 928 searches for the presence of the 7 1 (0.25 cubic foot) flag, which was previously entered by the user in the working memory 126, and at step 930, the programmable counter 174a is loaded at a corresponding count rate, and a corresponding one Factor is stored in the buffer location EORP of the working memory 126.

If there is no identification signal that corresponds to one of the volumes mentioned, then in step 932 it is automatically

130062/0716

It is true that the test is performed on a 28 l (1 cubic foot) volume of the flow meter 38.

If, however, it is determined in step 919 that the output signal of the proximity detector 27 is to be used to measure the fluid flow across the flow meter 38, so goes the subroutine to step 934 in which the state of the 28 1 (1 cubic foot) flag within working memory 126 is checked; if this flag is set, then the programmable counter 174a of FIG loaded with a binary number equivalent to that volume and a corresponding factor, normally 8, is entered into the buffer location EORP of the working memory 126 before the return to the main program takes place. If a 28 1 (1 cubic foot) flag is not set, the routine exits to step 938 where the condition of the 14 1 (0.5 cubic foot) flag is checked; if it is present it becomes a factor corresponding to this volume, normally 4, in step 940 into programmable counter 174a and also entered into the buffer location EORP of the main memory 126, before returning to the main program. If the identification signal for this volume is not set either, that is possible Subroutine to step 942 wherein the condition of the 7 1 (0.25 cubic foot) flag is checked; if it is present, so a factor equivalent to this volume, normally 2, is entered in the programmable counter 174a in step 944, as well as in the EORP buffer space of the main memory 126 before returning to the main program. If none of the characteristic signals corresponding to these volumes is set, so in step 946 the subroutine automatically sets the flowmeter calibration system 10 to check of flow meter 38 occurs with 28 liters (1 cubic foot) flowing therethrough.

The next program 9O4 and the subsequent programs, shown in Figure 9A are a series of steps

130082 / 071S

Sf

to determine which flow rate is used should and how often each flow rate should be checked. Reference is now made to Figure 9C. In step 948 the Logic and driver circuit 198a activated to activate lamp 111c Activate "Test sequence" before proceeding to the step in which it is determined whether the 126 stored flow rates one of the values Q0 to Q7 correspond. If so, the subroutine goes to step 952a where it is determined whether the test in the Flow rate QO should take place. If so, the speed becomes 0 in step 954a in the speed buffer location of the working memory 126 entered; then the flow meter calibration system leads 10 in step 956a the check of the flow meter 38 at the selected flow rate and outputs the result to the printer, as will be explained in detail with reference to FIGS. 9E and F. If but the test is to be carried out at flow rate Q1, corresponding steps 954b and 956b are carried out, whereby the speed 1 is entered and executed in the appropriate steps. Corresponding steps 952c to 952h, 954c to 954h and 956c to 956h are executed, putting the corresponding flow rates in the velocity buffer entered and executed. If decision step 950 produces the result "no", then that is possible The program proceeds immediately to step 958 in FIG. 9D, in which by checking the identification signals in the assigned areas of the main memory 126 it is decided whether a the flow rates Q8 to QF was chosen. If so, the routine continues to step 960a. If not, return the subroutine returns to the next step 908 in Fig. 9A. In step 96Oa it is determined whether a flow meter should be checked at a flow rate equal to Q8. If so, then in step 962a the speed is 8 is entered into the speed buffer of working memory 126. At this point, via step 964a, the transfer

130082/0716

Checking the flow meter 38 by means of the calibration system 10 performed at the specified servomotor speed and flow rate. In the same way are the steps 96Ob to 96Oh, 962b to 962h, and 964b to 964h, allowing checks at the various flow rates and corresponding motor speeds are executed before returning to the next step, wherein it may be step 908 or 912 to determine the number of tests per flow rate to be run.

Tn the PUj. 9C and 10 eriol <jt according to W.ilck »r flow rate, at which the flow meter is to be checked, the actual one Carry out this test in steps 956 and 964. The actual flow meter test using the calibration system 10 takes place in the subroutine according to FIGS. 9E and F. In step 966, the logic and driver circuit 198a in FIG. 4E driven to turn on the lamp 111c "test sequence" via Step 332 in FIG. 6B, the piston 14 is then brought into its starting or lowermost position, which is shown in FIGS 2A is shown. At step 968, logic circuit 194b is instructed to open inlet valve 34 and close of the exhaust valve 36 to make. As indicated above, there is a certain one in the speed buffer of the working memory 126 Speed is stored, and the servo motor 20 is controlled by the control of the logic circuit 194a so that that it reaches this speed, which is done via subroutine 606 in Fig. 8F. It is important that in the subroutine 606 it is established that the piston 14 has been accelerated from its starting position into the test start position, wüo vom Au; i <jfljifjiir.lqn, il dtvt Νίϋι «·· οιΐι« _ |!. «" 1ίΜ ekt ors iingeicft wlrU, and that the speed of the piston is stabilized. Then in step 970 the output signals of the rotary encoder 40 and the linear optical encoder 26 by signal conditioning and logic circuits 170a and 170b, respectively locked so that they are not in the corresponding counter 174a

130082/0716

and 182a, and the time divisors or factors originally selected based on the unit volume are entered into the programmable counter 174a. After the piston has reached the start position, which is determined in step 600, the output of the rotary encoder 40 is periodically addressed, for example every 40 microseconds, in order to determine the occurrence of the rising edge of the next pulse and to feed it to the input of the interval timer 176. If this has not happened, the subroutine checks the status of the proximity switch 54 in step 974, and if the piston 14 is in this position, ie in the region of its uppermost position within the cylinder 12, an indication is given that an error has occurred and the subroutine is exited via point 975 and a program interruption arrives, as will be explained with reference to FIG. 9G. When the leading edge of the output signal of the rotary encoder 40 occurs, however, the program goes to step 976 , in which the output signal A of the rotary encoder 40 is applied to the interval timer 174 and in particular to its programmable counter 174a via the signal conditioning and logic circuit 170a. Step 976 also enables the counter 182 to begin counting the pulses derived from the linear encoder 26, and also enables the programmable counter 174a to count the pulses from the rotary encoder 40. In addition, step 976 enables counter 176a at this point in time so that it counts the system clock signal (FIG. 4C) as clock A (eg 200 kHz) by dividing this clock signal by a suitable factor, the output signal being via the logic circuit 178 a 1-see, call-up signal is issued, by means of which the various temperature and pressure values are called up. In addition, an output signal is obtained via the interval timer 174 in order to enable the logic circuit 177, which is otherwise blocked, so that an end-of-test program interruption can be initiated

130062/0716

and in particular the digital level "1" is generated when the programmable counter 174a reaches the inputted count rate has counted down, a signal then being applied to the interrupt 2 input of the CPU 120 to initiate a program interruption subroutine trigger, which is explained in detail with reference to FIG. 9J. Thereafter, in step 528, the Pressure and temperature data measured continuously in the flow meter 38 and in the chamber 28 are output, as in FIG. 8P shown. Thereafter, in step 980, the state of the counter 176a in the interval timer 176 is checked, and if none Pulses have been accumulated, the routine returns to the beginning of step 980 to identify the occurrence of the first pulse from the counter 176a to be seen. If one or more pulses have been detected at the output of counter 176a, the subroutine continues to step 982 in which the counter 176a is reset to zero. The counter 176a is therefore reset to zero, when the 1 sec pulse is output from logic circuit 178 will. Then in step 984 the flow meter pressure MP2 from the output of the pressure transducer 46, the calibration room pressure from the pressure transducer PB, the calibrator pressure PB1 from the pressure transducer 51, the flowmeter temperatures TM3 and TM4 from the temperature transducers 42 or 44 MUWLo the Kichgorcittemperaturen 'JTI and Tl> 2 au «the Tcjnpor.jt urwandltün L> 7 and ΰϋ in the assigned Places entered within the working memory 126 in order to to be available for the calculation to be performed on it, as will be explained later. In step 986, which is based in detail of Fig. 9L, the pressure and temperature data are converted into decimal form with floating point, and in the With reference to step 988, which is explained in detail in FIG. 9M, the data are cumulated and averaged. As in detail later As will be explained, the pressure and temperature values in the flow meter 38 and within the chamber 28 will be periodic cumulative, e.g. every second, starting with a point in time at which the flow meter test begins until it ends is, whereby all measured values are summed up and divided by the number of measured values in order to determine the temperature and pressure

130062/0718

parameters to be averaged over time. The temperature and pressure parameters should therefore be used while the check is being carried out of the flow meter continuously recorded or monitored will. The last measured value of these parameters is recorded after the programmable counter 174a has already counted down Has. Then, at step 990, a determination is made as to whether the test has ended by interrogating register INT2; when the check is finished, the program proceeds to the next step in Fig. 9F. Otherwise it returns The program returns to the beginning of step 528 to get the current values of pressure and temperature from the buffers of the working memory 126 and continue with the flowmeter test. The generation and storage of an end-of-test signal (flag) enables the test process to be terminated at any point within the interval between the Inquiry pulses for taking pressure and temperature measurements. As will be explained with reference to Fig. 9J, this end-of-test flag becomes is set when the counter 174a of the flow meter has counted down, namely at an associated location of the Working memory 126 to indicate that verification is complete is. The 1-second interrogation pulse obtained from the logic circuit 178 according to FIG. 4C must, however, still appear in order for the to receive the last temperature and pressure values. The subroutine shown in FIG. 9E, in particular step 990, enables the conclusion of the test process by the last pressure and Temperature data is obtained before the flowmeter calibration system is stopped. By means of the logic circuit 177, the stopping of the flow meter counter 174a can be made fast can be determined so that an end-of-test identification signal can be generated quickly and reliably. If so by programming is to be achieved, it would be necessary to repeatedly change the state of the programmable counter 174a check to see if it has counted down, which is in the programmable read only memory 124 saving program would grow in size considerably

t30062 / 0716

and the accuracy of test completion would be decreased.

From the above description it is clear that the flow meter test triggered and terminated in response to the output signals of the rotary encoder 40 of the flow meter will. The mechanism coupling the encoder 40 to the diaphragm of the flow meter 38 produces a kind of frequency-modulated one Signal, so that it is beneficial to start and stop counting its output signal at approximately the same point in the rotation cycle of rotary encoder 40. For this purpose, the piston 14 is accelerated from its starting position to an essentially constant speed, before going through the test start position, which is determined in step 646, then periodically, e.g. 40 microseconds, the appearance of the rising edge of the next Pulse A from the rotary encoder 40 is checked. This next pulse A from the rotary encoder 40 triggers the Down counting of the programmable counter 174a, into the a count rate has been entered which indicates the test volume that is to flow through the flow meter 38 to be examined, and at the same time the counting process of the counter 182, which counts the output signal of the linear encoder 26 by one to provide an accurate indication of the volume drawn into the calibration system 10. When the counter 174a has counted down, becomes a Signal applied to logic circuit 177 to indicate a CPU program interrupt 2, which, as shown in Fig. 9J, disable the counters 174a and 182, so that their counting is terminated will. The counting rate stored in the counter 182, which indicates the pulses received from the linear encoder 26, is started and stopped in counting in response to the output of the rotary encoder 40, whereby it is ensured that the counting rate stored in the counter 182 corresponds exactly to that in the programmable Counter 174a is stored, which indicates the fluid measurement made by means of the flow meter 38.

13006270716

So

Reference is now made to Figure 9F. In step 992 the mentioned final values of the pressure are obtained after the End of test signal has been set and are converted into decimal form with floating point before they are cumulated and averaged will. The final values of pressure and temperature are output via the logic circuit 190a in FIG. 4D to on the screen 112, corresponding to the subroutine described above with reference to FIG. 8P. Then the in the counter 182 accumulated count rate in step 994 in decimal form implemented with floating point, and in step 620 the servomotor is braked, namely according to the above with reference to FIG. 8G explained subroutine 62O. Step 996 then provides control signals through the PI / O circuit 196 and the logic circuit 194b to control the coil of the outlet valve 36 and to open this valve, and to control the coil of the inlet valve 34 and close this valve. Then in step 998 the various, now fully collected parameters used to calculate the volumes that are present at the output of the optical linear encoder 26 'resp. optical rotary encoder 40 of the flow meter are specified, in the reference of Fig. 9N. In step 1000, a control command is sent to the logic and driver circuit 198a is sent to turn on the "end of test" indicator 111d, and then in step 1002 those calculated in step 998 are calculated Output results transferred to the printer, as will be explained in detail with reference to FIG. 9P, and the piston 14 is brought back to its lowest or starting position, which happens in step 322 shown in FIG. 6B, at this point? The program returns to the beginning of step 952a in Fig. 9C to determine if there are repeated tests or tests are to be performed at different flow rates.

In Fig. 9G, the program interrupt subroutine 975 is shown, which begins in particular when the stop key 111g is pressed or when in step 974 of FIG. 9E

130062/0716

Error is detected based on the lack of counting events from the rotary encoder and the simultaneous presence of the piston is based in its uppermost position corresponding to the location of the proximity switch 54. When actuated of the stop button 111g, the servomotor 20 becomes fast decelerated until stopped, and the system is "locked" in this stopped state until the main power supply is switched off and turned on again. When the operating current is reapplied, the calibration system 10 returns to its Back to standby mode. In both cases that begins Program at point 975 in Figure 9G, and at step 1004, logic and driver circuit 198a causes the lamp to illuminate behind the stop button 111g. In step 1006 the straight The present motor speed is derived from the logic circuit 198b and stored in the speed comparison buffer of the main memory 126 saved. Then the servomotor 20 is braked according to the program 6 20 in FIG. 8G. After that, the Step 1008 commands are sent to logic circuit 194b to to control the associated coil for opening the exhaust valve 36, and to control the associated coil for closing the first or inlet valve 34. The subroutine runs through then a wait step 1010 until power is turned off and reapplied.

As shown in Fig. 9H, program interruption 1 begins when the reset button 111g of the system control and status module 111 is pressed, whereby the program in progress is interrupted to go to step 1012, in which the logic and driver circuit 198a is controlled so that it turns the lamp behind the reset button 111g turns on. In step 1014, the currently present The speed of the servo motor 20 is read out via the logic circuit 198b and stored in the speed comparison buffer or an addressable location in the main memory 126. The servo motor 20 is then under control

130062/0716

32,

of the logic circuit 194a is braked to a stop, specifically via the subroutine 620 shown in FIG. 8G, and it takes place a return to the program via point 212, in particular to the Step 214 in FIG. 5.

A third program interrupt, relating to steps 976 and 990 above in FIG. 9E, is shown in FIG. 9J and runs automatically upon completion of the flowmeter test when the interval timer 174 passes a pulse the logic circuit 177 applies the interrupt 2 input of the CPU 120, this pulse occurring after the count rate has been counted down is one of the named volume factors 28, 14 or 7 1 (1, 0.5 or 0.25 cubic feet) drawn through the flow meter 38 via the calibration system 10 should. Initially, the system state is stored in appropriate locations in memory 126 in step 1018, and in step 1020 the counters 174a and 182 in Fig. 4C are disabled, and the stored count rate of the counter 182 is stored in suitable locations within the main memory 126, before the two counters 174a and 182 are reset. In step 1024 a 1 or an identification signal im End of test register of main memory 126 is set as soon as one of the above-mentioned factors or parameters is in the counter 174a is counted down. The end of a test can therefore occur halfway between the 1-second polling pulses that are sent to the Obtaining the pressure and temperature readings can be used. So you have to wait for the final pressure and temperature values and this is done by setting the end of test flag as soon as the above-mentioned factors count down are. The flow meter calibration system 10 is running then continue to accumulate the data as indicated in Figure 9E until step 990 occurs, in which access to the end-of-test register takes place in the main memory 126 in order to check whether an end-of-test signal has been set. If necessary, will then the test or verification is ended. If he still

130082/0716

S3

expires, the status of the system is retrieved again in step 1028 and the operation of the calibration system continues at the point of interruption continue within the flow meter check routine as shown in Figure 9E.

The program 984, shown generally in FIG. 9E, for entering the pressure and temperature values is detailed in FIG. 9K shown. In the initial step 103 2, the pressure difference is displayed from the pressure difference sensor 4 6 by means of the A / D converter 166 converted into binary data and stored at the associated location within the working pexcher 126. the The pressure difference detected by the piston pressure transducer 51 is converted into binary data by means of the A / D converter 166 in the same way implemented and stored in step 1034 at the assigned location of the main memory 126. In the same way, the Barometer pressure measured by the transducer 109 in the calibration room 6, in which the calibration system 10 is located, by means of the A / D converter 166 converted into binary data and stored in the assigned location of the main memory 126 via step 1036. In the same way, in steps 1038, 1040, 1042, the temperature input signals TP1, TP2, TM3 and TM4 are generated from the Temperature measuring devices 57, 58, 42 and 44 converted into binary data by means of the A / D converter 158 and assigned to them Digits are stored within working memory 126 before returning to step 986 in Figure 9E.

Step 986, shown generally in FIG. 9E, is shown in greater detail in FIG. 9L. In step 1046, the temperature values corresponding to the temperatures TP1 and TP2 of the calibration device and the flowmeter temperatures TM3 and TM4 its binary form in decimal form with floating point in the units Fahrenheit or Celsius and then converted to the assigned Place stored within the working pexcher 126. in the Step 1048 then in a similar manner the binary signals representing the pressure values MPI, MP2 and the barometric pressure PB

130082/0716

- »Br -

display / taken from the corresponding digits of the main memory 126 and in decimal form with floating point in the units bar or psi implemented and stored in the assigned locations within the main memory 126 before Returning to step 988 in Figure 9E.

The subroutine or step 988 for accumulating and averaging the data shown in FIG. 9E is detailed in FIG. 9M shown. In step 1050, the digital values of the two Exchgerät temperatures TP1 and TP2 are added to one to provide the mean calibrator temperature (ATP). This value in turn becomes the previously accumulated values of the averaged Calibrator temperature (ATP) added so that a number of N test samples are added up during the course of a flowmeter test. In step 105 2, the flow meter temperatures TM3 and TM4 added to give an average flow meter temperature (ATM) and this current one Value is added to the previously added values of the mean flowmeter temperature (ATM) by one variable to obtain which corresponds to N successive test samples of these temperatures, which in the course of a flow meter test were recorded. In step 1054, the value of the ambient or barometric pressure PB becomes the sum of the barometric pressure PB / the pressure difference MP2 between the ambient pressure and the pressure inside the flow meter 38 added to give an average flow meter pressure (APM) to obtain. This averaged flow meter pressure is added to the previously added values of the same pressure to obtain a quantity which indicates N successive readings of the mean flow meter pressure. In step 1056 the value of the ambient or barometric pressure PB is added to the pressure difference PB1 measured by the transducer 51 to obtain an average calibrator pressure (APP). This average calibrator pressure (APP) becomes added to the previously added values of the same pressure,

130062/0716

in order to obtain a quantity which N consecutive measured values of the averaged calibrator pressure determined in the course of a flow meter test. In step 1058 the number of measurements of each value from steps 1050, 1052, 1054 and 1056 is added to produce the total number N of Determine measurements whereby an average of each of these parameters can be obtained in the subsequent steps, by dividing the added size by the factor N.

Step 998 shown in FIG. 9F will now be explained in detail with reference to FIG. 9N. The accumulated values of the mean flow meter temperatures (ATM), the mean flow meter pressure (AMP) and the mean calibrator pressure (APP) are based on N measurement samples taken during the course of the flow meter test, and in step 1060 these averaged values are calculated for each temperature or won every print. In step 1062, the percentage error of the output signal of the encoder 40 based on the volume of fluid which has passed through the flow meter 38 is calculated with respect to the standard volume which has passed through the calibration system 10, in accordance with the equation given. The values of the flow meter pressure PM, the calibrator pressure PP, the calibrator temperature TP and the flow meter temperature TM are the values averaged in step 1060, while the count rate K for the calibration volume is obtained as the count rate of the signals of the linear encoder 26 which was counted in the counter 182? these count rates indicate the selected standard volume, for example 28 1, 14 1 or 7 1 (1, 0.5 or 0.25 cubic feet). As shown in FIG. 9N, the ratios PM to PP and TP to TM result in pressure and temperature correction factors, respectively. As already mentioned, the temperatures and pressures within the calibration device 10 and the flow meter 38 are different during the measurement. The ratios determined in step 1062 result in suitable correction factors by which the temperature differences are compensated in order to achieve an accurate indication of the percentage error in the reading of the flow meter, based on the

130062/0716

Calibration volume or the reading of the calibration device 10. The Factor K indicates the count rates stored in the programmable counter 174a which are those from the flow meter 38 indicate measured fluid flow. As can be seen in step 1062, the factor K * is multiplied by a constant C, to convert the count rate, e.g. 40,000 per 28 liters (1 cubic foot), into the value of the volume that is passed through the flow meter 38 should flow. In the embodiment shown, in which the entered in the programmable counter 174a Factor K * 40,000, the constant C is chosen to be 1/40,000. It also counts the number of counts from the linear encoder 26, which are counted by the counter 182, multiplied by a calibration factor which - as explained in detail later is obtained by precisely measuring the volume of the chamber 28 of the calibration device 10 and this precisely measured volume is set in correlation to the pulse sequence from the 'linear encoder 26. In one embodiment, it results that for a volume 28 1 (1 cubic foot) of the chamber 28 38 790 output pulses come from the linear encoder 26. The calibration factor is then determined as the reciprocal of this counting rate and has the value 1/38 790. If another volume via the flow meter 38 is to be sucked and the programmable counter 174a is programmed with a count rate that another Volume, another calibration factor must be determined which corresponds to this volume within a selected range corresponds to chamber 28. In practice, the count rate 38,790 is obtained empirically by pressing the piston 14 over a Distance of the chamber 28 is moved, which corresponds exactly to the volume 2.8 1 (or 1 cubic foot), whereupon the number of pulses is calculated from the linear encoder 26 or otherwise determined. Similar measurements and coding can be done empirically to determine the count rates for half or a quarter of the specified volume, based on the output signal of the linear encoder 26, while the piston 14 over a distance starting from its rest position in the Cylinder 12 is moved, this volume within the

130082/0716

Cylinder corresponds. The actual count rates and the corresponding calibration factors therefore correspond exactly to these Volumes which are delimited by the cylindrical wall of the chamber 28. This allows the calculation of the percentage of error between the reading of the flow meter 38 and the actual volume sucked into the chamber 28 are carried out with high accuracy by using calibration factors that are based on a precise volume of the corresponding Areas of the chamber 28 were determined. After completing step 1062, the program returns to the next step 1000 back in Fig. 9F.

The calculation of the average temperatures and pressures shown generally in step or subroutine 1060 of FIG. 9N is shown in full in the subroutine of FIG. 9Q. As already indicated in step 1050, each temperature reading is taken is added to the previous one to obtain a grand total which is the sum of the discrete measurements of the Specifies the exchanger temperature ATP in the course of a single test run. As indicated in step 1064 of Figure 9Q, this accumulated value ATP is divided by the number of measured values AVCTR = N + 2 and added to 459.67 to obtain the final calibrator average temperature FTP in degrees Kelvin. In the same way are in step 1066 the temperature values TM3 and TM4 one after the other in the course of the flow meter check is added to a cumulative value of the average temperature ATM of the flow meter received in the course of a review; this value ATM is determined by the number of measured values or by the counting rate of the 1-second sample rate divided, and the result is added back to the value 459.67 to get the final averaged Get temperature FTM in degrees Kelvin. Similarly, in step 1068 the cumulative values of the averaged The pressure APM of the flow meter divided by the number of AVCTR readings recorded during a test, to a final value FPM of the flow meter

130062/0716

result in absolute pressure units. In step 1070, the values of the averaged calibration device pressure APP accumulated in the course of a flow meter test are represented by the Number AVCTR divided by measured values to obtain the final result of the calibration device pressure FPP in absolute pressure units, before returning to step 1062 in Figure 9N.

The accuracy of the calibration device 10 when measuring the volume that has flowed over the flow meter 38 depends on and is limited by changes in fluid pressure and temperature measurements during a flow measurement. The changes of pressure and temperature during a test run are greater than when measuring and calibrating the volume of the calibration device. Generally can be the actual volume of fluid that flows through the flow meter 38 while the piston 14 is from a first into a second position can be approximated by the following equation:

₊ v ₂

Therein V-, P., and T- are initial volume, fluid pressure and temperature within the chamber 28 of the calibration device 10 at the beginning of the flow meter test, ΔΡ and ΔΤ are the corresponding Changes in temperature and pressure over the course of the test, and "St and" Sp are the mean deviations of T ^ and P ^ im Flowmeter 38. The logic circuit shown in Figure 4C supplies a 1-second clock signal that contains the output signals of the Temperature transducers 42, 44, 48 and 57 as well as pressure transducers 46 and 51 interrogate for measurement samples of these variables in the course the flow meter test. As explained above, the periodic measurement samples are the flowmeter temperature TM3 and TM4 and the calibrator temperatures TP1 and TP2 obtained in step 984 of FIG. 9E are shown in FIG Steps 1050, 1052, 1054, and 1056 of Figure 9M are summed to give a spatially averaged flowmeter temperature and

130062/0718

Calibration device temperatures are averaged over time by being queried and totaled in the course of the test, as explained with reference to steps 1064, 1066, 1068 and 1070 of FIG. 9Q. In this way, the current changes in pressure and temperature in the flow meter and in the calibration device 10 are measured. The temperature and pressure changes, which would otherwise cause an error in the calculation of the volume change, can be compensated for by spatial and temporal averaging. Temperature changes in the fluid entering the calibrator caused by adiabatic expansion at a pressure drop of 0.00381 bar (1.5 "H ^ O) can be as much as 0.27 ° C (0.5 ⁰ F), thereby causing an error in the calculation of up to 0.1%. It is therefore necessary to closely monitor fluid pressure and temperature throughout the system and make the necessary corrections to obtain the "actual" volume which is in the flow meter -Eichsystem 10 flows in. For an accuracy of 0.05% of the volume of the parameters must 6T and 6P in the chamber 28 and at the inlet of the flow meter with a Genauiekgit of more than ± 0,055 ⁰ C (0.1 ⁰ F) and ± 0 , 000127 bar (0.05 "H ₂ O) can be monitored. The temperature transducers and pressure transducers must therefore be able to deliver the desired accuracy so that the desired volume measurement accuracy is achieved.

Step 1002 shown in FIG. 9F for outputting the results of the tests made on the flow meter 38 to the printer is shown in detail in Figure 9P. In the first step 1072, the printer is controlled by logic circuit 190b queried to be initiated and then the percentage of errors previously calculated in step 106 2, printed out on the printer. Then in step 1074 the printer receives a line feed command to advance one line, before, in step 1076, the flow rate at which the check was made is displayed in step 1076. In step

130062/0716

AOO

1078 10 line feed commands are given to the printer, to advance the paper 10 lines before returning to step 322 in Fig. 9F.

Of particular importance to the invention is the method of accurately measuring the volume of the chamber 28 of the cylinder in connection with the utilization of this measurement for the precise calibration of the output signal of the linear encoder 26 to the Specify the volume per output pulse of the linear encoder 26 can. As shown in Fig. 2A, the linear encoder includes a scale 24 which has a large number of precise, closely spaced ones Has markings 102, whereby the optical encoder 26 generates an output pulse train which the passage each corresponds to a mark on the optical encoder. By accurately correlating the number of output pulses too The linear encoder 26 can encode the precisely determined volume of the chamber 28 with the corresponding high accuracy in this way the display of the error of the flow rate registered by the flow meter 38 being examined to improve.

In this calibration process, it is first necessary to measure the volume of the chamber 28 with a high degree of accuracy. A Volume measurement method is based on the application of a high-frequency electromagnetic field within the cavity. The basics of such a measuring method will now be explained. The internal dimensions of the regularly trained Chamber 28, which is completely surrounded by conductive walls, can be accurately determined by measuring the frequencies where there are resonance conditions for the electromagnetic field built up in the chamber. For a given Geometry, the electromagnetic fields within the chamber 28 can assume various spatial configurations. at The electromagnetic energy enclosed in the chamber 28 is stored over time periods at discrete frequencies (in

130062/0716

Resonance), which are long compared to the period of oscillation, and these resonances are called the normal modes or Vibration types of the chamber 28 designated. The ratio of the stored energy to the dissipated energy per cycle the resonance frequency is referred to as the quality factor Q in the resonance state. The quality factor Q is a measure of the losses due to of the ohmic resistance of the walls' which they oppose to electrical currents generated by the electromagnetic Fields are induced. For a given normal mode or vibration type, the resonance frequency is clearly through determines the dimensions of the chamber 28 and the speed of light in the medium which fills the volume. By simultaneous measurement of the resonance frequencies of a number of normal modes equal to the number of linear dimensions which can determine the volume (one dimension for a sphere, two for an upright circular cylinder, etc.) the volume of the chamber 28 can be determined with an accuracy comparable to the accuracy of frequency measurements.

For a given chamber geometry there are an infinite number of normal modes, the resonance frequencies of which have a lower one Have limit corresponding to the wavelength in the order of the linear dimensions of the chamber 28 in free space, however, they have no upper limit. A change in volume caused by a mechanical shift or change in one of the linear dimensions, in turn causes a change in the resonance frequency of the lower modes by approximately same factor.

As shown in FIG. 10, an antenna 70 is provided to direct an electromagnetic field within the chamber 28 generate and take from this a relatively small proportion of the stored electromagnetic energy, so they are measured by means of the circuit shown in FIG

130002/0716

can. As will be explained in detail later, measurements are made to determine the frequencies f at which resonance occurs in the chamber 28. The choice of the shape of the chamber 28 is important because if any shaped chamber is used, the case may arise that the mathematical relationship between the resonance frequencies f and the dimensions of this chamber does not give a clear or analytical solution. The chamber 28 is so ijowcihlt- that .sie has a reyelinüß Lije geometry in order to ensure that the losses carried away and the coupling effects are negligible, so that known Maxwell equations can be established which describe the relationship between the chamber dimensions and the Specify resonance frequencies on the basis of the boundary conditions that the electric field E and the magnetic field H are tangential to the entire closed surface of the regularly designed K.Humor 28 ItU. UnI ff (1 It ¹ SfU hcd i nyuntjtMi Inilirn tlJe tittsuiKjcn the field equations of the Kaituner 28 take a relatively simple form, and recognizable disturbances can be predicted.

In a preferred embodiment of the invention, the chamber 28 is therefore selected and is an upright circular cylinder through the inner surface of the cylinder 12, the head part 60 and the exposed surface of the piston 14 is formed. A chamber 28 designed in this way obviously enables that the piston displaced through it displaces a known volume, as stated above. The normal ones Modes of the electromagnetic field within the channel 28, which is designed as an upright circular cylinder, are shown in FIG divided into two groups: transverse modes of electric field strength (TE), for which the electric field is along the cylinder axis (ζ) of the chamber 28 is equal to 0, and transverse modes of the magnetic field (TM) for which the magnetic field is longitudinal the cylinder axis is equal to 0. Furthermore, these modes are denoted by three integer indices that represent the TE-

130062/0716

Modes in the form of the cylindrical coordinates r, θ and ζ as follows are determined:

£, = number of whole periods of E related to Θ; m = number of half periods of E _Q related to r; η = number of half periods of E related to z.

The same group of indices applies to the TM modes, for which the integers I ₁ m and η are defined accordingly, in relation to the components of the magnetic field,

ie H and H _Q.
r ü

The solutions of the resonance frequencies of the normal modes result from the geometrical dimensions and root expressions of the Bessel functions by the following general relationship:

f _ _{(C / D)}

Here D is the diameter, L is the length and c is the speed of light in the medium filling the void volume; χ is given by:

χ = mth root of J '(x) = 0 for TE modes;

x. = mth root of J ₀ (x) = 0 for TM modes. Around. *

The numerical values of these Bessel root terms corresponding to the lower TE and TM modes are in the following table specified:

1300S2 / 0716

Transverse mode of the transverse mode of the

electric field (TE) magnetic field (TM)

TE X _n TM x,

11 1.84118 01 2.40483 21 3.05424 11 3.83171 01 3.83171 21 5.13562 31 4.21009 02 5.52008 41 5.31755 31 6.38016

12 5.33144 12 7.01559

A graph of the dependence of the size (fD) on the

2
Quantity (D / L) from equation (1) gives a straight line

2
with the point of intersection (cx „/ π) and with the slope (cn / 2). This diagram is shown in FIG. 13 for the lower modes of the chamber designed as a vertical cylinder for the values η up to 2. 13 graphically shows the respective resonance frequency values as a function of

2 speed of the geometric parameters (D / L) as well as the Change in the resonance frequencies with a change in the linear quantity L. It is also useful to change the number of to be able to predict expected resonances in a given frequency range for any fixed value D / L, as well like the values of L, at which two different, degenerate oscillation forms or modes occur which are mutually exclusive interfere with or get into interference with each other,

The mode diagram of Fig. 13 is used to represent the modes of the electromagnetic field, the expected resonance frequencies of the selected modes and the size of the diameter D and the length L of the chamber 28 to be determined. In one embodiment of the chamber shown in FIG the diameter D is 304.8 mm (12 inches),

130062/07 1

and the change in length L is from 254 mm (10 inches)

2 to 762 mm (30 inches). The corresponding values of (D / L) are therefore in the range from 1.44 to 0.16. If the observed The frequency range is between 500 and 1000 MHz,

so (fD) ^{2 changes} between 2.32 χ 10 ²⁰ and 9.29 χ 10 ²⁰

2 2

Hz cm. The number of expected resonance modes when the length L changes in the range from 500 to 1000 MHz is then given by the number of lines within the rectangle of the diagram in FIG. It can be seen from this that if L is 254 mm, there are only 3 modes that occur between 500 and 1000 MHz. These are namely the modes TM ₀₁₀ at about 755 MHz, TE ₁₁₁ at 830 MHz and TM _Q1 .. at 960 MHz. For L = 762 mm (30 inches) there are eight modes, namely TE ₁₁₁ , TE ₁₁₂ , TM ₀₁₀ , TM ₀₁₁ , TM

^, TE ₃₁₁ , TE ₃₁₃ and TE ₁₁₃ , which are not all shown in Fig. For reasons of clarity. In any intermediate position, the number of modes with their resonance frequencies f can be determined from the intersection of the vertical line that corresponds to the desired

Value of (D / L) corresponds, with the lines indicating the different Specify fashions to be obtained. Since the measurements of the resonance frequency of two .different modes Define both D and L unambiguously, the other modes can either be used for redundancy checking or else be to assess the effects of disturbances caused by deviations from the geometric Ideal form, and any correction terms that may be present to calculate higher order.

Qualitatively it can be stated that the electrical and magnetic Field components of the electromagnetic field and a defined relationship to the resonance frequency Have sizes D and L of the chamber 28 if it has a regular geometric shape, in particular diameter and Length of the chamber 28 as shown in FIG. Ttfie still explained, the relationship between the resonance

130062/0716

Α0 (?

frequency and the quantities D and L for a particular vibration type or mode through a mathematical one Expression. The chamber 28 with an upright or perpendicular circular cylinder shape has two unknowns Dimensions that determine their volume, namely their diameter D and their length L. It is therefore necessary to to indicate two mathematical relationships that are solved simultaneously for the unknown values D and L. can be; it is therefore necessary, as stated above, to generate electromagnetic fields in the chamber 28, which have two separate modes, and to determine the resonance frequencies of these two modes, whereby the Values of D and L and consequently the volume of the chamber 28 can be calculated .. As will be explained below the excitation modes are chosen so that high quality factors occur at the resonance frequencies, thereby reducing the effects of disturbances in the measurement of D and L or errors caused by them can be kept to a minimum.

The line of the oscillation mode TM ₀₁₀ has the slope 0, as shown in FIG. 13, and is therefore independent of the length L of the chamber and only depends on its diameter D. This special property can therefore be used to identify this type of vibration in experimental measurements. The vibration type TM ₀₁₀ can therefore be used to make an independent measurement of the quantity D. It can also be seen that the change in the resonance frequency for a given vibration type as a function of the quantity L is determined solely by the last index η. Consequently, the vibration modes TE ₁₁₁ , TM _n11 and TE ₉₁₁ shift by the same amount

ill vJ If Z I I

wear while L changes while modes TE ₁₁₂ , TM ₀₁₂ and TE ₂₁₂ are shifted together by twice the amount. The tuning properties of these oscillation modes can therefore be used to determine the relative change in the length L.

130062/0716

/ 0}

to be determined with very high accuracy as soon as the absolute diameter of the cavity is determined.

The figure of merit, Q, of the chamber 28 at resonance in a given normal vibrational mode is twofold significant. First it determines the width of the resonance frequency and therefore limits the accuracy that can ultimately be measured experimentally on the basis of this resonance frequency. But more important is that the quality factor Q is a measure of the magnitude of the expected deviation of the resonance frequency from the ideal results according to equation (1). The higher the quality factor Q is for a given vibration mode, the more precisely it is the theoretical relationships used to determine the geometric dimensions of the cavity. The modes of vibration to excite the chamber 28 are chosen so that the highest quality factor Q results in this way a to obtain a more precise determination of the dimensions and consequently the volume of the chamber 28.

For a volume that is completely conductive walls is limited, the eigenvalue Q of the quality factor of the chamber in the unloaded state is infinite, and the frequencies of the resonance modes found are exact. In walls with finite conductivity, however, the electrical ones penetrate and magnetic fields within the chamber 28 into these walls at a depth known as the skin depth δ and is given by the following equation:

δ = [λρ / 12Οττ ² μ J ^1/2 cm (2)

Here μ is the permeability of the wall material, λ the Wavelength in free space in cm and ρ the resistivity (direct current) of the walls in ohms cm. There is a finite depth of penetration the apparent dimensions of the chamber as seen by the electromagnetic fields

130062/0716

AOt

If larger than the actual geometric dimensions ^{, the release effect on the walls caused by the ohms 1} see losses disturbs the resonance frequency of the normal modes and shifts it to a lower value. A measure of this disturbance is the value of the ratio of the penetration depth to the free wavelength, ie (δ / λ), which can be calculated using equation (2) by using the known values of resistivity in the case of direct voltage and the permeability of the wall material. For a cavity made of copper (μ = 1, ρ = 1.72 χ 10 Ohm cm) the penetration depth is equal to 3.8 χ 10 ~ λ ' cm. At 1000 MHz (χ = 30 cm) the ratio (δ / λ) is about 7 χ 10. This correction of the resonance frequency caused by the finite conductivity is

obviously negligible for the present application. For stainless steel of type 300 (μ = 1, ρ = 72 X 10 ~ ohm cm) the ratio of (δ / λ) at 1000 MHz is 4.5 χ 10, which is no longer negligible if an overall accuracy of 0 .01% of the absolute volume measurement is aimed for. In a preferred embodiment, the walls of the chamber 28 are coated with chrome (μ = 1 / P = 13 χ 10 Ohm era) and the corresponding value of (δ / λ) is equal to 2 χ 1 " ⁵ at 1000 MHz. Based on theoretical calculations the correction of the resonance frequency of the normal modes in the chamber 28 would therefore have to be negligible in the case of an upright circular cylinder, and the relationship according to equation (1) would have to be sufficient to give a total volume determination with an accuracy of more than 0.01%.

In practice, however, the theoretical value of the penetration depth for a given material is never reached, for several reasons including wall imperfections, material contamination, and surface contamination. Consequently, the actual need

130062/0718

Values of the penetration depth are determined experimentally, by checking the Q factor of chamber 28 at selected preferred Excitation modes is measured, and if necessary, the appropriate correction must be made due to this effect the volume determination can be made.

The value Q _{Q of} the chamber 28 is obtained by calculating the ratio of the stored energy to the losses dissipated on the walls per cycle of electromagnetic oscillation, according to the formula:

2jH ² dv

Q ₀ * , (3)

⁰ ÄJH ^Z ds

Here, H is the magnetic field vector of the normal vibration mode and 6 is the penetration depth. The Q factors for the normal modes and for the geometric shape of the chamber 28 for an upright cylinder cavity are as follows Given equations:

For TE modes:

in the

For TM modes:

r 2 ₊ «2.2.1 / 2

for n> 0 (5)

for π · 0

In it are R = (D / L), P = nir / 2 and all other symbols like previously defined.

130062/0716

/ ΙΛΟ

By computing the right hand sides of the above equations for a given mode and cavity geometry and dividing by the experimentally determined value Q, an effective ratio δ / λ can be obtained. A direct comparison of the result obtained with the result according to equation (2) then gives a quantitative one Assessment of the size of the disturbance when determining the volume the chamber 28 is to be expected.

The chamber 28 is coupled to an external measuring circuit shown in Fig. 10 and the microwave energy for the chamber 28 generates in order to build up an electromagnetic field therein and the reflected power therefrom refer to. The chamber 28 is a reflection type cavity which is just a coupling device in the form of an antenna 70 requires. As will be explained later in detail, the resonance characteristic, particularly the resonance frequency of a given normal mode of vibration, determined by the power reflected from antenna 70 as a function the incident microwave frequency is measured. The use of a reflection type chamber 28 only yields minimal interference compared to the behavior of an ideal chamber and enables the use of a directional coupler 1106, which allows the reflected power to be sampled without disturbing the incident power. As in 10, the power output of a tuned generator 1100 is applied to the antenna 70, while the reflected power passes from the coupler 1106 to a crystal detector 1110.

The presence of the coupling device in the form of the antenna 70 results in additional losses in terms of dissipation the electromagnetic energy stored in chamber 28 at resonance, d. H. the amount of the Measurement taken power. This is usually defined as an equivalent coupling, Q or Q, as opposed to that

130062/0718

/IN THE

Factor Q of the cavity in the unloaded state, which is also is referred to as Q. In addition, the interaction of the cavity with the remaining circuit parts in FIG via antenna 70 for a detailed description of the "cavity coupling system" therewith the appropriate relationships for the invention can be derived.

It can be shown that for a reflection-type chamber 28, i.e., a chamber, with performance over a single one Antenna 70 introduced and taken from this, the ratio of the reflected power to the incident power near the resonance is given by the equation:

i f ♦

Here ν is the frequency, ν is the resonance frequency, Q is the Factor Q without loading the resonance cavity and Q is the coupling factor which is proportional to the power loss through the Coupling device is.

At the resonance frequency ν is the ratio of the reflected Power versus incident power defined as the reflection coefficient for a given normal vibration mode, i.e .:

<Vo> v .v _o -fi ■ We - <> o) ² / «c ⁺ " ο) '(7)

For ß = 0 the cavity is considered to be 100% coupled so that Q = Q. This condition corresponds to the case of a low frequency equivalent circuit in which the load impedance is matched to the generator impedance.

130062/0718

Combining equations (6) and (7) and eliminating Q gives the factor Q in unloaded State in normal vibration mode depending on the measurable parameters;

(8th)

^Q o "V (P _r / P ₀ ) -β / l ' (P _r / P ₀ ) 3 ^1/2 / (v.

Thus, by measuring β and the frequency width at any arbitrary power level (P / P _n ) on the response curve of the resonant cavity, the quantity Q can be determined. If a half power point is defined such that:

Pj / 2 '(1 ⁺ / 3) / 2, (9)

so the frequency difference results according to the two Points for half power as "half the width" of the resonance curve and is given by:

Av »2 (V ₁ , j - ν),

V ₁ / o is the frequency corresponding to the points for half power. For this case, equation (8) can be further simplified:

Q ₀ * (2V ₀ ZAv) Z (I _{+ iff} 1/2, _{m (io)}

The value of Q therefore changes between (ν _ο / Δν) and (2ν_ / Δν) ι depending on the degree of coupling.

A graph of the response curve with the parameters defined is shown in FIG. The center frequency v _{o of} the response curve, which corresponds to a minimum of the reflected power, is the resonance frequency of the normal mode.

The in Fig. 10 naturally causes a change in the resonance frequency, i.e. a dragging of the

130062 / Ö716

Frequency, due to the interaction between the chamber 28 and the elements of the circuit according to FIG. 10. In the circuit shown in FIG circuit arrangement shown, in which the frequency of the tunable Generator 1100 changed by a factor of about 2 the directional coupler 1106 receives the output of this generator and applies a relatively small portion of it to the antenna 7. The one to the auxiliary or exit opening of the directional coupler 1106 applied power is about 80% the output power of the tunable generator 1100. The chamber is connected to the main entrance port and the crystal detector 1110 is connected to the third port of the directional coupler 1106 coupled to measure the microwave power reflected from the chamber 28. The application of the directional coupler 1106 leads to a damping factor of 100 or a damping between the tunable generator 1100 and the load formed from the antenna 70. As a result, the pierceable generator 1100 is relatively good of the antenna 70 decoupled, so that it is ensured that no interaction occurs which would disrupt the supply in the chamber. As a result of the decoupling between the antenna 70 and the generator 1100 by means of the directional coupler 1106, the strength of the Frequency tracking by the quality factor Q of the chamber 28, the coupling coefficient ß and the standing wave ratio between the crystal detector 1110 and a cavity coupling system certainly. Assuming that tunable generator 1100 is completely decoupled from cavity 28, result in a cavity 28 with the quality factor Q = 5000, a coupling coefficient β of 0.5 and a standing wave ratio

—5 2 a frequency pull in the order of 1.3 χ 10 and 0.0013%, respectively. This deviation is an order of magnitude better than necessary in order to achieve the desired overall measurement accuracy of the Chamber volume of better than 0.01% can be achieved.

An examination of the cavity 28 and the associated resonant frequency measurement system 10 has shown that disturbances due to geometrical deformations of the chamber 28,

130062/0718

no matter whether it is a lack of roundness of the cylinder shape
of the cavity 28 or small local surface irregularities and deformations can be compensated by the measures explained below
to be hit. If the machining tolerance of the chamber 28 ensures that its diameter D is maintained within 0.3 mm (12 mils), the diameter D and volume of the chamber 28 can be determined with an accuracy within
Of the order of 10 or less can be determined, and
consequently, this deformation can be used for what is considered here
Volume measurement methods are neglected. The shift in the chamber resonance frequencies for any normal mode due to a slight inward or outward curvature on the inner wall of the chamber 28 is also considered to be negligible if the diameter of this deformation is significantly smaller than the cutoff wavelength of the electromagnetic field in the chamber 28, so that this deformation no electromagnetic field is coupled to other structures. Under these conditions, the frequency tightening is approximately proportional to the ratio of the cube of the deformation diameter to the νοίμπιεη of the chamber 28. In practice, the frequency tightening for inward or outward curvatures in the walls of the chamber 28 is negligible. For example, a 25.4 mm diameter hole in a chamber 28 that is about 305 mm (12 ") in diameter and about 508 mm (20") long results in a resonance frequency change of only 7 parts 10.

As for the effects of the inlet 62 and the openings to the As far as the reception of the transducers 51 and 57 in the piston 14 is concerned, they can have openings or holes during the volume measurement and calibration with metal plugs to eliminate these sources of error when determining the resonance frequency completely ruled out.

In order to meet the accuracy requirement of ± 0.01% when measuring the

130062/0718

To meet displacement volume in the chamber 28, the resonance frequency measuring circuit shown in Fig. 10 must be able to measure the resonance frequencies of the normal modes within the chamber 28 with an accuracy of 1:10. The circuit shown in Fig. 10 is designed to reduce systematic errors which can affect the measurement of the resonance frequencies f, such errors being due to impedance mismatches between tunable generator 1100 and antenna 70, changes and fluctuations in microwave power, interference signals and drift phenomena can be caused by components. As shown in FIG. 10, the microwave power source has the form of the tunable generator 1100, which can be, for example, the VS80A type from Exscan. The tunable generator 1100 can, for example, be operated at a fixed frequency (CW) or automatically swept over a range of 50 kHz and 300 MHz, at a rate between 0.05 Hz and 30 kHz. Furthermore, the output signal of the tunable generator 1100 can be adjusted via its fine adjustment knob 1100a in order to effect frequency changes directly by the user over the specified range. The output signal of the tunable generator 1100 is applied via a coaxial cable to a directional coupler 1102, which acts as a transformer, whereby part of the energy applied to the directional coupler 1102 is transferred to a frequency counter 1108, which in one embodiment of the invention is a counter of type no. 192OA made by Fluke Corporation. As will be explained later, the counter 1108 indicates the frequency at which a standing wave is established in the chamber 28. The output signal of the directional coupler 1102 is in turn applied to a similar coaxial cable and arrives at a second coupler 1104 and via the directional coupler 1106 to the microwave antenna 70. As shown in FIG. 10 and in detail in FIG. 2A, it is the microwave antenna 70 around a simple metal loop 70a , which is isolated from the head part 60 of the chamber 28 by means of an insulator 70b.

130062/0716

As is well known in the art, the energy reflected by the antenna 70 from the chamber 28 decreases significantly when a standing wave occurs compared to the energy reflected at other frequencies. This is known as the resonance condition, and the associated frequency is called the resonance frequency. When changing the frequency at the output of the tunable generator 1100, a resonance frequency is selected at which a standing wave occurs in the chamber 28, which depends on the shape and the dimensions of the chamber 28. The frequency at which a standing wave occurs determines - as will be explained later - the chamber dimensions or its diameter D and length L and consequently the volume of the chamber 28. The coupler is used to determine the power drop in the resonance frequency. 1106 ■ connected to the crystal detector 1 ¹ UTO which converts the light reflected from the antenna 70 microwave power into a DC signal. The crystal detector 1110, which can be, for example, type No. 423A (NEG) from Hewlett Packard, applies its DC voltage output signal to the Y input of the oscilloscope 1112. This oscilloscope 1112 can be, for example, the · P 922R from Tectronix. The X input of the oscilloscope 1112 is fed by the tunable generator 1100, so that when this generator is set to tune mode, the response of the chamber 28 to the reflected power depends on the input signal frequency and is displayed on the oscilloscope 1112. As can be seen in the enlarged display 1112a, the power reflected by the antenna 70 has a minimum 1113 at the resonance frequency, as shown in FIG. The frequency at which the minimum 1113 occurs is displayed on the counter 1108. The second coupler 1104 applies the microwave power to a crystal detector 1114, which emits a corresponding DC voltage signal, which is amplified by an operational amplifier 1115 and applied to the tunable generator 1100 in order to regulate the level of the output of this generator

130062/0716

this an essentially constant power when passing through the frequency at which a standing wave occurs, records.

The circuit shown in Fig. 10 operates in the manner described below to measure the resonance frequency. First, the tunable generator 1112 is set on large Durohötlmmboroicrh GinqoHlcll t so that essentially all normal resonance modes can be displayed simultaneously on the screen of the oscilloscope 1112, while the resonance response for any particular vibration mode can be displayed individually by the tuning width and the Center frequency of the tuning band suitable to be selected. The oscillation modes can be clearly identified by their resonance frequencies for a certain I ·: I fC'1 t'l I UIi (J <l "i Kd IIkmiuI t · II uii'J 1 J) d« m h.-i 1 b dri Ktiinnir) 2b (JiMm *. ".:. '' Π and equation (1) is applied; instead, the piston can also be moved, and the magnitude of the change in the resonance frequencies as a function of the piston position is determined by the values from Fig. 13 compared.

The resonance frequency I of a given vibration mode becomes measured by first showing the response curve on the oscilloscope screen is displayed and then the tunable generator 1100 is set to operate at fixed frequency (CW) and the frequency fine adjustment knob 1100a is turned until the to the. The voltage displayed on the oscilloscope 1112 has a minimum. the The indication on the frequency counter 1108 when this minimum is reached is then the resonance frequency of the normal cavity oscillation mode. It is useful if the second beam of the oscilloscope 1112 is used to determine the position of this Minimum better to define by putting the system on tuning mode is adjusted and the vertical position of the second beam is adjusted so that this is the ground just touched the resonance curve. If the generator 1100 does so

130062/0716

is set that it works with a fixed frequency in CW mode, the resonance frequency f of the chamber 28 is determined by the frequency setting of this generator, according to the condition that the two beams coincide. In a preferred embodiment, the frequency of the counter is read while the two beams coincide exactly, whereby the frequency drift of the generator 1100 is eliminated so that the measurement can be made with a much better accuracy than the tunable generator 1100 itself is stable. Furthermore, since the measurement is based solely on the fact that a minimum is set in the response behavior of the chamber, it is independent of non-linearities of the detector and of fluctuations in the incident microwave power over time. Repeated measurements with a certain vibration mode show that the resonance frequency of the chamber with an accuracy
can be.

an accuracy of better than ± 3 kHz. or 5:10 determined

A first embodiment of a method for accurately measuring the volume of a portion of the chamber 28 will now be described in detail. In the method according to the invention, the resonance frequency of the chamber 28 is first measured for two different excitation modes. As an example, consider an upright cylindrical chamber 28 and _{measure the resonance frequencies in the TM 010} "and TE ₁₁₁ modes. The modes are non-degenerative and their resonance frequencies can be measured with an accuracy of 1:10 or better using normal technique The mode TM ₀ "(parallel surface) depends only on the mean diameter (D) of the chamber 28 and is independent of the length (L) of the cavity, while the mode of oscillation TE ₁₁₁ depends on both (D) and ( L) depends. The volume of the chamber 28 can therefore be determined from the measurement of the two frequencies. For the case TM _{010 the following} results:

X ₁ = Re s on an ζ we ile length - ^ - or D = - (11)

1 X ₀₁ π

130062/0716

Here D is the diameter and X _Q1 the first Bessel root term or J _Q (X) = 0.

In the case

surrendered:

X ₉ = resonance wavelength =

f ^X 11) ₊ 1

{-Ίφ-J ⁺ τ

(12)

Here D = diameter, L = length or height of the cavity and X ₁₁ = first root term of J ₁ (x) = 0.

The combination of the two results gives:

__ ^X 01 λ, L = 1/2

D - - l

^λ 1 ²

-1/2

(13)

Depending on the resonance frequency, the results can be written will be as:

^X 01 ^C

2f.

1 -

^f 2 ^X 01

-1/2

(14)

Here, X ₁₁ = 1.8412, X _Q1 = 2.4048 and c = the speed of light in the medium with which the cavity is filled (in the present case air).

The following results for the total volume of the chamber:

Volume = ttD

-L = ^Λ 1 ^λ 2 ^Χ 01

C ³ X ^{2
c Χ} 01

1 -

11

Oil

-1/2

11 01

-1/2

(15)

130062/0716

It can be seen that the volume is primarily proportional to λ or 1 / f. Therefore:

or 3 (ä |] (16)

As a result, the frequencies can be measured with an accuracy of 1:10, as well as by counter 1108, see above that the theoretical accuracy of the volume is on the order of 3:10.

The procedure can be applied to the volume change. to measure in an upright circular cylinder 12, which is caused by the direct displacement of the piston 14 located therein. The volume of the chamber 28 can be easily measured both before and after the movement of the piston, as can the magnitude of the change in volume. In the arrangement shown in FIG. 10, when the piston 14 moves from the X position to the Y position, the resonance frequency of the oscillation mode TM _{Q1O remains} constant (or changes solely due to irregularities in the diameter of the cylinder), while the resonance frequency of the TE ₁₁₁ shifts proportionally to the shift AL. In the position X, the resonance frequencies f .. and f ″ are measured and used in equation (2) to obtain the first volume V ₁ . The piston 14 is then moved to the second position Y, and a second pair of resonance frequencies f ¹ , f ₂ 'for the modes TM ₀₁₀ and TE ₁₁₁ are recorded in order to calculate the second volume V2 according to equation (15). Finally, the displacement volume AV is calculated by subtracting the first determined volume V ₁ in position X from the last determined volume V- in position Y. By continuously monitoring the resonance frequency of the TM ₀₁₀ mode, changes in the diameter of the chamber 28 (due to machining inaccuracies) between L 1 and L 1 as a function of L can also be measured. The change in volume can be measured in the same way

130062/0716

AXA

by continuously adjusting the resonance frequency of the TE ₁₁ . -Vibration mode is monitored.

Disruptive influences in the above relationships are the dielectric ones Properties of the air, the presence of the coupling lines 30 and 32 and the other gas inlet 62, Surface irregularities, the finite electrical conductivity of the wall material of the chamber 28 and degenerative waveforms in the case of intersecting modes. As long as the irregularities are small compared to λ (on the order of 30 cm or larger), the disturbances are in the first place Approximation proportional to the change in volume. The deformations are thus averaged out by the method, so that the measurement is precisely proportional to the actual volume of the chamber 28.

The coupling lines 30 and 32 and the gas inlet 62 are dimensioned so that they are significantly smaller than the cutoff wavelength of the microwaves, so that they should disturb the resonance frequency no more than 1:10 ui Approach can be corrected.

should bother at most like 1:10 and at least initially

The disturbance due to the finite electrical conductivity the wall material of the chamber 28 should be of the same order of magnitude, which can be achieved if the walls made of a highly conductive metal such as copper, silver, or gold Formed or coated with aluminum with careful polishing within reasonable limits. E.g. the theoretical penetration depth for copper at 300 MHz is 3.8 x 10 cm. The disturbance of the volume lies in the Order of magnitude of the relationship between the penetration depth and the linear dimension of the resonance cavity, which is at an upright circular cylinder with a radius of 50 cm gives about 7.6 x 10. The actual depth of penetration can can be estimated from the cavity losses that are in direct

130062/0716

Relation to the quality factor Q of the chamber, which is generally experimental with an accuracy of about
1% can be measured. As a result, a first order correction can be made, eliminating the uncertainty
less than a few parts is reduced to 10.

The change in resonance frequency between vacuum and air in the cavity is given by:

^<F Vakuuir / ^f Air> = (ε) ^1/2 (17)

Here ε is the dielectric constant for air at microwave frequencies, which for dry air under normal conditions STP has the value STP = 536.5 χ 10 ~ ⁶ . The change in frequency during the transition from vacuum to air is therefore of the magnitude

_ λ ·

order of 2.7 x 10. Since ε is known exactly for dry air at microwave frequencies and as a function of pressure and temperature, this shift can be corrected with an accuracy of at least 1:10. To correct the pressure and temperature dependency of ε to an accuracy of better than 0.1%, the following relationship can be applied
will:

1atm] - < ^ρ / ^7δο > / 0 +0.00341 (t-20) J.

—4
Since the frequency change is initially only 2.7 χ 10, an overall accuracy of the resonance frequency determination in the order of magnitude of 10 can be expected if the barometer pressure is monitored better than 0.1% (corresponding to 0.0013 bar).

The contribution of water vapor (relative humidity) to the
Dielectric constant of air can be given:

Water vapor ^{X 10} "'= ⁵ ' ⁰⁰ ^ ψ ^) ^ ' Ο»)

130062/0716

Therein, T is the one via a precision temperature measuring device measured temperature in degrees Kelvin, and P is the partial pressure of water in millibars. For T = 20 ° C (293 ° K) the saturation vapor pressure is (100% relative humidity) 23 millibars. For this extreme case, the following applies:

(^ ¹ > water vapor ^{x 10} ~ ^{6 = 100} '

which is about a third of the value for dry air. This effect, in turn, can be corrected as a first approximation by measuring the relative humidity, so

_ · 7

that an accuracy of the order of 10 in determining the vacuum resonance frequency of the cavity can be achieved.

The frequency of the two modes TE ₁₁₁ and TM ₀₁₀ is not degenerative with other TRM resonances. Unexpected deyimeraLivo Kfiekte. auJ <jj and modes randomly crossing from a LcIi can be avoided by appropriately choosing the dimensions of the volume. The conditions for mode crossing between D / L> 0 and D / L = 3 are:

D / L = 0.45; D / L = 1 and D / L = 2.14.

At D / L = 0.45 the modes TM ₀₁₀ and TE _{112 are} degenerative; with D / L = 1 the modes TM ₀₁₀ and TE _{111 are} degenerative; at D / L = 2.14 the modes TE ₁₁₁ and TM _{110 are} degenerative. By choosing the ratios D / L different from these values, interactions with perturbing modes are avoided, and (1, iii Refionanzvei hai i en de :; Ji <> h 1 raiuiioo i:; L is well-defined so that the formulas for calculation of the resonance frequencies from the dimensions of the chamber 28 are strictly valid.

As an illustrative example, it should be assumed that work is being carried out in the range 1 <D / L <2.14, and that this is due to the net shift

130062/0716

of the piston 14 is to be displaced to a volume of 2.2652 χ 10 ⁵ cm ³ (8 cubic feet). Then the following values result that can be used:

D = 104.88 cm

L ₁ = 52.44 cm = end position Y of the piston L ₂ = 78.66 cm = initial position X of the piston.

So the shifted net volume is

π ₍₂₆ , 22) = 2.2652 χ 10 ⁵ cm ³ .

It can also be seen that the ratio D / L changes from 1.33 for the initial position to 2 for the end position, what is well within the desired work area. In this case the following applies:

f ₁ = resonance frequency of TM _O1Q mode = 219.0 MHz f ₂ (i) = initial value of TE ₁₁₁ -MOdUS = 253.9 MHz (D / L = 1.33) f ₂ (f) = final value of TE ₁₁₁ -MOdUs = 331.5 MHz (D / L = 2).

Similarly, for the case of a total volume of 56.6337 x 10 ³ cm ³ (2 cubic feet), the frequencies are:

f ₁ = 347.6 MHz
f ₂ (i) = 403.0 MHz
f ₂ (f) = 526.2 MHz.

The dependence of the frequency on the increase in L can be specified become:

Af

- Γ 2_ ^ 1

LO, 3739) (LVD> + 1Y

This gives for D / L = 1.33:

(Af / f) = 1.1 (AL / L). (20)

130062/0716

and for L / D = 2 we get:
(Af / f) = 1.5 (AL / L).

As can be seen, the inaccuracy in the measurement of L is approximately equal to that in the frequency measurement.
Consequently, a very high precision can be achieved in the determination
of the volume shifted here.

It is also useful to estimate the quality factor Q of the resonance modes, because the measurement precision when measuring the resonance frequencies depends largely on the sharpness of the resonance points. For the TM ₀₁₀ mode:

ςή- = 0.22 for D / L = 1.33,

Q- = 0.19 for D / L = 2.
λ

Here, δ is the skin or penetration depth due to the relationship δ = | _ (λρ) / 120ττ μΐ '; ρ is the specific resistance of the Wall material of the chamber 28, λ is the wavelength and μ is the permeability of the wall material.

If the chamber 28 is made of copper, then ρ = 1.7 χ 10 applies, μ = 1 and δ = 4.43 χ 10 cm at 219 MHz. Hence:

Q = 6.8 x 10 ~ ⁴ for D / L = 1.33 and
Q = 5.9 χ 10 ⁴ for D / L = 2.

The width changes depending on the coupling effect the resonance curve at the points for half power between (2f / Q) and (f / Q), where f is the resonance frequency. Consequently is the width of the resonance curve for the values of Q:

at f _o = 219 MHz: 3.2 <Af <6.4 kHz for Q = 6.8 x 10 ⁴

3.7 <Af <7.4 kHz for Q = 5.9 χ 10 ⁴

130062/0716

_2

Since f usually has an accuracy of 10 to f or can be better determined, an accuracy on the order of 10 can be expected in determining f. This in turn gives an accuracy of the measurement of the diameter D of the chamber 28 of the same order of magnitude.

The same applies to the mode

£ = 0.28 for D / L = 1.33;

q £ = 0.27 for D / L ■ = 2

This gives:

Q = 7.5 χ 10 ~ ⁴ at 253.9 MHz and Q = 5.5 χ 10 ~ ⁴ at 331.5 MHz.

As before, the following results for the widths of the resonance curves:

3.4 <f <6.8 kHz at 253.9 MHz and 6.0 <f <12.0 kHz at 331.5 MHz.

If it is further assumed that f can be determined with an accuracy of 10 Af, then for the worst case (12.0 kHz): Af _Q / f _o - χ 1O ~ 4. ⁷ From the relation previously derived for D / L - 2 one obtains:

Af

Therefore, (AL / L) can be determined to be (4 χ 10 ~) / 1.5-2.7 χ 10

So it is possible to disturb the resonance frequency of a cavity mode due to the presence of the gas inlet and outlet 62 on the head part 60 of the cylinder 12. From the theorem of adiabatic invariance and knowledge of the The field configuration inside the cavity can determine the frequency

130062/0716

-ra-

Detuning can easily be determined due to holes. When the dimensions of holes are significantly smaller than the cut-off frequency (which strictly applies to the case considered here), the frequency shift is proportional to the Ratio of the cube of the hole diameter to the volume of the chamber 28.

_{The change in the resonance frequency at TM 010} caused by the opening 62 in the middle of the plate 60 is given, for example, by:

(Af / f _o ) = (d ³ ) / 8D ² L (X ₀₁ JJ ₁ ² (X ₀₁ ) (21)

Here d is the diameter of the hole, J ₁ (X ₀₁ ) the value of the Bessel function J ₁ at X ₀₁ and Af the frequency shift ..

Inserting the numerical values with D = 104.88 cm, L = 52.44 cm, X ₀₁ = 2.40483 and J ^ (X ₀₁ ) = 0.2695 results:

(Af / f _o ) = 3.35 χ 10 " ⁷ d ³ .

Obviously, for d is on the order of 2 cm or less the frequency shift is only about 2 χ 10. As a result, the displacement is very small, and this effect can be by a suitable calibration procedure carried out initially e.g. covering the opening 62 with a suitable metal plug, as a systematic error with regard to the precision of the method can be practically eliminated.

The coupling of the microwave energy into the chamber 28 for the two modes TM ₀₁₀ and TE ₁₁₁ is best achieved by arranging a coaxial feed line which ends in the antenna 70 at a point which is approximately halfway off center of the head piece plate 60 of the cylinder 12, the loop being oriented along a radius. The magnetic field at this point has about 90% of the maximum field intensity inside the cavity for both modes. Consequential

130062/0716

Both modes are borrowed with the same high coupling strength stimulated. By relocating the coupling to the head part 60, the coupling is furthermore established by the movement of the piston 14 unaffected.

There now follows a description of a second preferred embodiment of the method for measuring the displacement volume in the chamber 28 and for using this precisely determined volume to calibrate the pulse train from the optical linear encoder 26. In the same way as described above, the piston 14 is from the first in 10 brought into a second position, designated _{L 1, which is designated L 2} , with a shift by AL. The cylinder 12 is rigid as such, so that it is possible to carry out the calibration only occasionally in order to ensure that no "yatiMnatisehon r.iuiq / t'1 UlndrriinqfMi mif tt'c> t on, i. B. Changes in the dimensions of the cylinder 12, misalignment and malfunction of the optical linear encoder 26 or deformation of the piston 14. In order to optimize the absolute measurement accuracy of the microwave volume calibration, the mechanical shape of the cylinder 12 must be as perfect as possible, i.e. a completely enclosed rectangular circular cylinder so that any sources of systematic errors that would otherwise interfere with the microwave measurements are eliminated or reduced.

Reference is now made to Figure 12A. Certain mechanical changes are planned there. First of all, the spatial Gap between the piston 14 and the walls of the chamber 28 well closed so that no microwave energy can be lost through this gap. As mentioned in the beginning, Simultaneously filed application of the applicant with the title "piston seal" is explained, the gap has a considerable size, due to the nature of the seal between the piston 14 and the wall of the chamber 2.8. A Lid 11 is formed from a suitable metal material such as stainless steel and has a number of spring-like ones

130062/0716

Fingers 15 shown in detail in Fig. 12B and between the piston 14 and the inner periphery of the chamber 28 are arranged; when the cover 11 is placed over the piston 14, the fingers stand in the gap between the piston 14 and the wall of the chamber 28, these fingers in close contact with the piston 14 and the wall are. The fingers 15 act as a short circuit through which the electromagnetic field is reflected, which would otherwise be through would pass through the gap. In one embodiment, the spring-like fingers 15 are made of beryllium copper. the Pressure and temperature sensors 51, 57, 48 and 68 have been removed and replaced by suitable smooth metal plugs, the shape of which is such that a substantially flush surface with the inner walls of the chamber 28 is achieved. The fluid inlet opening 62 in the head portion 60 is also covered by a metal plate to be substantially smooth To create surface on the top of the head part 60. Furthermore, the peripheral wall of the chamber 28 is provided with a suitable one Solvent cleaned to remove any residual oil that has leaked from the piston seal can be. If it is noted that the desired calibration is that of a displacement volume AV and not that of the absolute Volume of the chamber 28, it can be seen that the above mechanical changes reduce the accuracy of the calibration process not affect. As soon as the measurements have been carried out for the chamber 28 modified in this way, the same measurements can be carried out be performed immediately after the chamber 28 is brought to its normal working condition, and a set of suitable calibration factors can be generated to relate the two sets of measurements to one another set. The results of the second group of measurements can then be used as basic data, with which subsequent Verifications of the absolute calibration can be compared without the whole process of changing and redoing Assembling the chamber 28 to have to repeat.

130062/0716

In summary, the volume measurement and calibration process take place,

in
by bringing the piston 14 / the first position L ₁ in FIG. 10, which is achieved by manually turning the rotating part 19 of the servo motor 20. In the first position, the antenna 70 is _{subjected to electromagnetic energy of the two modes TE 111} and TE ₁₁₂ , these modes being selected in such a way that the interference explained above is minimized. The frequencies f .. and f ~ at which resonance occurs for each mode are determined by observing counter 1108. Then the piston 14 is moved over the distance AL into the second position L ₂ , where the antenna 70 is again subjected to electromagnetic energy of the two modes mentioned and the corresponding frequencies at which resonance occurs are determined for the two modes. The output signal of the optical linear encoder 26 is applied to a counter which counts its pulses while the piston 14 moves over the distance AL. The diameters D ₁ and D ^ of the chamber 28 in the two positions L ₁ and L ₂ are calculated. The route AL is then calculated using the previously calculated values of D ₁ and D _? . The calculated value of AL is divided by the number of pulses from the linear encoder 26 during the movement of the piston 14 over the distance AL in order to obtain a length calibration factor which is based on the measurements of D ₁ and D ". The volume AV, which corresponds to a volume that is _{delimited by the planes through the points L 1} and L ₇ and by the inner circumference of the chamber 28, can be _{specified mathematically as a function of the diameters D 1} and D ₂ and of the distance AL . When the output signal of the optical linear encoder 26 with respect to a given volume is to be calibrated, for example, 28 1 (1 cubic foot), then this value is substituted into the equation, and this is solved for the calculated values of D ₁ and D _2, the value of AL, which corresponds to the movement of the piston 14 in order to suck this volume via the flow meter 38. The calculated value of AL is multiplied by the previously calculated length calibration factor to obtain the number of pulses

130062/0718

to be obtained, which are output by the optical linear encoder 26 when the piston 14 extends over the distance AL moved in order to suck said volume into the chamber 28. As explained above, the output from the linear encoder 26 becomes The count rate obtained is used to calculate the calibration factor, which occurs in step 1062 of FIG. 9N. In particular the calibration factor is the reciprocal of the count rates that are determined in this way for the desired volumes, which a correction of the calculation of the percentage error in the reading of the flowmeter for an accurate measurement of the volume of the chamber 28, as explained above.

First of all it is necessary to measure the frequencies at. which standing waves occur in the positions L ₁ and L ₂ . To calculate the diameters D ₁ and D ₂ , the speed of light is required, which. changes with changing environmental conditions such as temperature, pressure and relative humidity. However, the corrections for changes in the speed of light are assumed to be minor, and the calculation of the speed of light is typically done once or twice in the course of the process of calibrating the optical linear encoder 26.

The speed of light in vacuum Co is 2.997925 x 10 cm / s. The corresponding value c for air is obtained by dividing Co by the refractive index of air at the observation wavelength. For the microwave range (f <30 GHz) the refractive index η is related to the atmosphere parameters as follows:

<nl) XlO ⁶ «-27 * 6 _(P.

Here, P is the total pressure in millibars, T is the temperature in ⁰ K and e is the partial vapor pressure of water in millibars. The speed of light is given by:

130062/0716

c = Co / n = Co / [l ₊ I2xi (P ₊ «10e, _{x 1C} f ⁶ ) (23)

Temperature and barometric pressure can be read directly on a thermometer or barometer near the flow meter calibrator 10 can be read. The partial vapor pressure of water can be derived from the relative humidity of the air via a sling moisture meter using the

Moisture meter formula is obtained, or else simple by using a standard table such as Smithsonian Physical Table No. 640.

In order to calculate the value of Al, the mean diameter value of the chamber 28 must be determined, in particular the values of the diameters D- and D2 at the points L ₁ , L2 must be determined. The calculation of D ^ and D ₂ is done with great care, since the resulting uncertainty of the volume is twice as great as the uncertainty of this measurement. The piston 14 is thus brought into the first position L ₁ , in which the frequencies f .. and f ₂ , at which standing waves occur in resonance, are measured for the two different modes. According to a preferred embodiment of the method, the frequencies f .. and f ₂ of two different modes of the same electrical characteristic are measured simultaneously as a function of the piston position L, and the resolution according to the averaged diameter D is carried out using the appropriate theoretical relationship.

In a preferred embodiment, after which the chamber 28 has the form of an upright circular cylinder, the modes used for this purpose are the modes TE ₁₁₁ and TE ₁₁ -. It has been found that the quality factor Q has a high value when excited with these modes, as a result of which errors in the measurement of the resonance frequency are reduced. The mean diameter of the cylinder at any given point L is given by:

D (L) - ₍₂₄₎

f "4f ₂ ² (L) -f _χ ² (L) H ¹⁷²

- JrTA '-

Therein f ~ and f .. are the resonance frequencies of the oscillation modes TE ₁ .. and TE ₁₁₂ , and c is the speed of light in air, which was calculated using equation (23). By using two different modes of electromagnetic excitation, the various disturbances such as different penetration depth, frequency shift due to the antenna 70, divergence of the inner circumference of the chamber 28 or its deviation from a perfect circular cylinder shape, can be compensated so that the absolute value D is obtained with great accuracy can be. If the frequencies are carefully measured at the counter 1108 (FIG. 10), an absolute accuracy of the values D as a function of L in the order of magnitude of 1:10, which corresponds to 0.00254 mm to 304.8 mm in diameter (or 0 This accuracy is on the same order of magnitude as changes in volume of chamber 28 due to thermal expansion and contraction in a temperature stabilized room where the temperature remains constant within ± 0.55 ° C (1 ° F).

In order to confirm these measurements and to show a quantitative possibility of estimating the magnitude of the expected disturbances of the measuring system, the diameter can be determined independently of this by measuring the resonance frequencies in the chamber 28 in the _{TM 010 vibration mode.} In such an excitation mode, the resonance frequency is independent of the length L, and therefore it should not change when the piston 14 is displaced in the case of a perfect cylindrical shape. However, the TM _Q10 mode is disturbed by other effects that must be taken into account in order to achieve the same accuracy as for the two aforementioned modes. For mode TM _Q the mean diameter is given by:

D (L) = 0.7654799c / f (25)

Where f is the resonance frequency of mode T.

130062/0718

As soon as the mean diameter D of the chamber 28 is determined as a function of L with the desired accuracy, the value AL is derived and placed in relation to the observed number of pulses from the optical linear encoder 26 to determine the length calibration factor as length per pulse interval or Number of pulses per unit of length. The piston is brought into position L and the resonance frequencies f ₁ and f n for the selected modes TE ₁₁₁ and TE ₁₁₂ are measured. The piston is then moved to the next position L _{? By rotating the rotatable part 19.} and then the resonance frequencies of the same modes are measured again, counting the number of pulses from the optical encoder during the movement of the piston 14 from the first to the second position. The number of pulses is divided by AL = L ₁ -L to get the desired length calibration factor. The distance AL = L ₁ -L ₂ should be large enough that the calibration accuracy is not affected by the accuracy of the pulse count rate (+1), and the calibration should be carried out over a number of distances AL to ensure that no non-linear Effects occur in this measurement.

For the TE ₁₁₁ mode, the change in distance AL is given by the following expression:

_f 2 0.5860671C

₂ -l / 2

Γ 2 f o, 5860671C] j

- L ^f l ~ l D (L ₁ ) JJ

(26)

Therein f "and D (L") are the resonance frequency of this mode TE ₁₁₁ or the previously determined mean diameter in the piston position L ", and f .. and D (L ₁ ) are the corresponding values in the position L ₁ .

For the TE ₁₁₂ mode, the change in distance AL is given by:

130062/0716

tL β (L ₂ " ¹ ! ^{5 = C}

-1/2, -

0.586067Ic

y> ~ ^{λ / χ} r ( 121

'ι - f ² | Of5860671C, j J Ll I ^D < ^L 1> JJ

wherein the various quantities are defined in the same way as above.

The same relationships can be found for each excitation mode

can be specified, and multiple modes can be used

to check the consistency and absolute accuracy of these measurements.

The exact displacement calibration volume AV between the piston positions L ₁ and L ₂ is given by:

Δν = V ₂ -V ₁ (π / 4) Γθ ₂ ² ^) +. (D ₂ ² -D ₁ ² ) L ^ | (28)

Here D ₂ and D _{1 are} the averaged diameter of the cylinder cross-section at piston positions L ₁ and L ₂ , and L = L ₁ -L ₂ . Obviously, from this equation (28), if the values D ~ and D _{1 are known} , and if a given absolute value AV of the displacement volume is assumed for the calibration, e.g. B. 28 1 (1 cubic foot), the corresponding value of AL, ie the distance the piston 14 must be moved to draw this volume into the chamber 28 of the flow meter calibrator 10, can be easily calculated. The purpose of the calibration is to obtain the number of pulses delivered by the optical linear encoder 26 for any displacement volume AV, and this number is obtained by multiplying the determined value of AL for a given volume by the length calibration factor to obtain the equivalent number of pulses output from the optical linear encoder 26.

130062/0716

- 4-27 -.

The choice of modes TE ₁₁₁ and TE _11? for exciting the cavity 28 is based on experiments in which the quality factor Q was determined for all modes for a number of normal resonance modes. For this quality factor measurement, the ratio of the reflected power (Pr) to the incident power (Po) at resonance must be measured, as well as the frequency width of the response curve at the points of half the power level P ₁ , "= (Po + Pr) / 2. It is desirable that the DC voltage output by the crystal detector 1110 depends linearly on the incident microwave power. This condition can be met in that the crystal detector 1110 is operated in the area with the quadratic dependence of the characteristic curve, corresponding to a DC voltage level of typically less than 20 mV. If necessary, the linearity of the response can be checked by using a step-wise attenuator on the tunable oscillator 1100. As soon as this is done, the coupling coefficient (Pr / Po) can be read directly on the screen of the oscilloscope 1112 as the corresponding voltage ratio. Then the level for half the power can be calculated as an equivalent voltage. The half width of the resonance curve in FIG. 14 is exactly the difference between the two frequency settings on the tunable generator 1100 which correspond to half the power level on either side of the resonance frequency observed on the oscilloscope 1112. The Q factor of the resonance is calculated using equation (10). From this determination of the quality factor Q it has been found that for the vibration mode TE ₁₁₁ the quality factor Q over the displacement range of the piston 14 is approximately 6000-7000, while it is approximately 8000-10000 in the _{TE 112 mode.} As indicated above, the figure of merit is a measure of the magnitude of the expected shift in the resonance frequency with respect to the ideal results according to equation (1). When using the above-mentioned modes, the resonance frequency can therefore be increased

130062/0716

Accuracy can be measured, and the disturbances due to surface defects or the like and frequency shifts due to penetration depth effects can be minimized. It is therefore assumed that the application of the modes TE ₁₁₁ and TE ₁₁₂ enables a more precise determination of the resonance frequencies and consequently of the mean diameter D or the volume displacement between the two piston positions.

The flow meter calibration device described above is capable of measuring fluid, in particular gas, flows through a measuring device with high precision. What is achieved by the invention in particular is that the volume of the cylinder into which the fluid or gas is sucked can be measured with extreme accuracy in order to be compared with the output signal of the encoder which detects the movement of the piston in the cylinder, whereby the volume sucked into the cylinder can be displayed with a correspondingly high level of accuracy. This reference or calibration volume is compared to the output of the flowmeter being checked to give an indication of its registration accuracy and the percentage error in the reading from the flowmeter with respect to the actual or calibration volume indicated by the optical encoder of the flowmeter calibration system to specify. The flowmeter calibration system of the invention is computer controlled, which carry out a number of tests and Pr ^- allows Ufungen in which to correct the display of the measured flow volume, the parameters of temperature and pressure of flow meter and calibrator are taken into account, which is also possible is to run repeated tests under different conditions. In particular, different fluid volumes can be sucked through the flow meter by means of the flow meter calibration device by entering corresponding counting factors into a counter of the computer and the selected counting

130062 / 071S

Λ »

rate is counted down to zero to end the testing process. The invention also provides a method created by which the volume of the cylinder into which the fluid is drawn, for a given piston determination can be determined with high precision. This exact measurement takes place via the frequencies at which standing Waves occur in the two piston positions, giving a precise indication of the fluid volume and the output signal of the optical encoder which detects the piston movement.

1300S2 / Q718

Blank page

Claims (27)

■ Patomanwplte Dipt -Ing. Dipl -Chem. Dipl.-Ing. E. Prince - Dr. G. Hauser - G. Leiser 3112227 Einsbergetsltar.se 19 8 Munich 60 ROCKWELL INTERNATIONAL CORPORATION March 27, 1981 Grant Street Pittsburgh, Pennsylvania 15219 / V.St.A. Our mark: R 1014 PATENT CLAIMS 1. Procedure for calibrating a flow meter calibrator, which is a chamber with the geometric shape of an upright circular cylinder of diameter D and length L, a piston rectilinearly movable within the chamber to convey a fluid between a test subject To move the flow meter and the chamber, and comprises a measuring device, which on the relative movement of the piston is responsive and generates a signal indicative of the amount of piston movement, each increment unit the signal of an increase in movement of the piston and consequently corresponds to the volume passed through the flowmeter being tested through the following process steps: a) the piston is inside the chamber in a, first Position taken; b) a first and a second electromagnetic field are generated within the chamber, the electromagnetic Fields of different normal vibrational 130062/0716 have modes which are chosen so that each vibration mode within the chamber under resonance conditions their electric and magnetic field components are related to the diameter and length of the chamber; c) the frequencies at which the two electromagnetic fields are in resonance condition within the chamber, are determined as the first and second frequency; d) the piston is brought into a second position which is displaced with respect to the first position while the signal derived from the measuring device is accumulated; e) a third and a fourth electromagnetic field are generated inside the chamber, these being electromagnetic Fields have different normal oscillation modes, which are chosen so that at Resonance condition in the chamber its electric and magnetic field components in relation to diameter and length of the chamber stand; f) the frequencies at which resonance conditions for the third and for the fourth electromagnetic field within the chamber prevail are measured as the third or fourth resonance frequency; g) on the basis of the first and the second resonance frequency v / earths diameter and length and from this first volume, which is limited by the chamber and the piston in its first position, determined and on the basis of the third and fourth resonance frequencies are the diameter and length and a second one Volume determined, which is limited by the chamber and the piston in its second position; and h) the piston displacement volume is expressed as the difference between the second and the first volume are determined, and the cumulative signal from the measuring device is in Correlation to the \ displacement volume set in order to to obtain an indication of the incremental value of the fluid, which is determined by an incremental unit of the signal of the measuring device is pictured. 130062/0716 2. The method according to claim 1, characterized in that the measuring device has an output signal in the form of a pulse train supplies, each pulse representing a distance of further movement of the piston, and that in step d) the delivered pulses are counted during the movement of the piston from its first to its second position. 3. The method according to claim 2, characterized in that a length calibration factor is also determined in step h), namely as an increase in length or distance of further movement, around which the piston is moved in order to generate a single pulse by means of the measuring device. 4. The method according to claim 3, characterized in that in method step h) a first diameter in the first and determining a second diameter of the chamber in the second piston position based on the Measurements of the first, second, third and fourth resonance frequencies, respectively, for a given volume and the determined second diameter the piston displacement is determined, which is necessary to the given fluid volume between the tested flowmeter and the chamber, and that on the Based on the length calibration factor and the piston displacement, the number of pulses is determined, which by means of the measuring device must be generated to the given volume between the tested flow meter and to move the chamber. 5. The method according to any one of the preceding claims, characterized in that the oscillation modes of the electromagnetic Fields in the chamber are chosen so that disturbances are reduced that would otherwise be caused by type, shape and surface of the chamber. 130062/0716 6. The method according to claim 5, characterized in that the first and second electromagnetic oscillation modes are the modes TM ₀₁₀ and TE ₁₁₁ , respectively. 7. The method according to claim 5, characterized in that the first and second electromagnetic oscillation modes are the TE ₁₁₁ and TE ₁₁₂ modes. 8. The method according to any one of the preceding claims, characterized in that the method steps are repeated, whereby the first position is changed for the successive groups of process steps, so that the piston is displaced through different chamber sections. 9. Device for calibrating a flow meter calibration device, characterized in that the flow meter calibration device contains: a) a chamber that is rigid and undeformable and its inner circumference as an upright or right-angled circular cylinder of determinable length and determinable Diameter is designed; b) a piston, which can perform a rectilinear movement within the chamber, to a fluid between a move the tested flowmeter and chamber; c) a drive device for driving the piston at a substantially constant speed during the Checking the flow meter; and d) a precision measuring device which responds to the relative movement of the piston within the chamber and a pulse sequence is generated, each pulse of which is a further movement distance of the piston and, consequently, the volume of fluid passed by that under test Flow meter is passed through, wherein 130062/0716 the calibration device comprises: - A device for moving the piston from a first to a second position; - A device for generating an electromagnetic field of variable frequency for building up a first and second electromagnetic field in the first and second position, these electromagnetic Fields have different normal oscillation modes, which are chosen so that under resonance conditions in the chamber the electric and magnetic field components of the first and second fields The diameter and length of the chamber are related; - means for determining the first and the second frequency at which the conditions for standing waves are resonated within the cylinder in the first and second piston positions, respectively; and a device for counting the number of pulses which the measuring device emits during the piston is moved from its first to its second position, thereby providing an indication of the increase in volume of the chamber, which is represented by a pulse, can be derived from the measured resonance frequencies certain displacement volume between the first and the second piston position and on the counted number of pulses is based. 10. The device according to claim 9, characterized in that the first and second oscillation modes are selected so that disturbances which can be caused by the type, shape and surface of the chamber are reduced. 11. Procedure for calibrating a flow meter calibrator, which has a chamber of regular geometric shape and a piston which is rectilinear within the chamber is movable to move a fluid between a tested flow meter and the chamber, 130062/0718 - D ~ characterized by the following process steps: a) Inside the chamber there are electromagnetic fields with first and second, mutually different, normal Oscillation modes generated, these modes are selected so that in the resonance condition within the Chamber the electric and magnetic field components to the dimensions of a regular geometric Shape having chambers are related; b) reflected energy is extracted from the chamber; c) a first frequency is determined at which the resonance conditions are satisfied for the first normal mode of vibration within the chamber; d) a second frequency is determined at which the resonance conditions the second mode of vibration are satisfied within the chamber; e) on the basis of the determined resonance frequencies, the dimensions and consequently the volume of the K ammer t immt; f) the piston is passed through the chamber and a quantity indicative of the extent of piston movement is generated; g) the quantity indicative of piston movement is based on an accurate determination of the chamber volume calibrated. 12. The method according to any one of claims 1 to 8 or 11, characterized in that the modes TE ₁₁₂ and TM ₀₁₀ are used as vibration modes. 13. The method according to any one of claims 1 to 8 or 11, characterized in that the modes and TE ₁₁₁ are used as vibration modes. 14. Device for measuring the volume of the chamber of a flow meter calibrator, comprising a piston rectilinearly movable within the chamber to carry a fluid 130062/0716 to move between a flow meter to be tested and the chamber, characterized by: a) a generator device for generating electromagnetic Variable frequency microwellononoryia; b) an antenna coupled to the generator device for emitting electromagnetic waves in the chamber; c) a detector device for detecting those leaving the chamber electromagnetic energy; d) a resonance condition detector which is connected to the Energy detector device is coupled and indicates the occurrence of minimum levels of energy, which leaves the chamber, corresponding to the establishment of resonance condition for standing waves within the. Chamber; and e) means for determining the frequencies at which resonance conditions for standing waves are within the Chamber are made, whereby the volume of the chamber can be precisely determined. 15. The device according to claim 14, characterized in that the generator device means for changing the frequency the electromagnetic microwave energy until the conditions for a standing wave within the chamber are made includes. 16. The device according to claim 15, characterized in that the generator device for automatically changing the frequency of the emitted electromagnetic microwave energy is set up. 17. The device according to claim 15, characterized in that the generator device comprises operating means to set the frequency of the emitted electromagnetic energy or To change the shaft via the operator. 130082/0716 18. Device according to one of claims 14 to 17 _f, characterized in that the energy detector device and the generator device are each coupled to the antenna and to the resonance condition detector device, whereby an indication of the level of the energy leaving the chamber as a function of the frequency the electromagnetic energy generated by the generator device is obtained. 19. Device according to one of claims 14 to 18 / characterized characterized in that the frequency detector device is coupled to the output of the generator device and Means which indicate the frequency of the electromagnetic energy. 20. The device according to claim 19, characterized in that the detector device comprises a counter. 21. Device according to one of claims 14 to 20, characterized in that means are provided which are attached to the Output of the generator device are coupled and include the level of energy and this energy level output indicative feedback signal to the generator device, whereby the power level of the generator device is stabilized. 22. a method of accurately measuring the volume of a flowmeter calibrator; which contains a chamber with a regular geometric shape, marked through the following process steps: a) An electromagnetic field is generated in the chamber with a normal vibration mode selected in this way is that under resonance conditions within the chamber the electric and magnetic field components the electromagnetic field in relation 130062/0716 related to the dimensions of the regular shaped chamber; b) reflected energy is taken from this chamber; c) the frequency of the electromagnetic energy is determined at which the resonance condition for the normal mode of vibration is established within the chamber; and d) based on the determined resonance frequency the dimensions and from this the volume of the chamber is determined. 23. The method according to claim 22, characterized in that the chamber has the geometric shape of an upright or has vertical circular cylinder and the dimensions to be measured, the diameter and the axial length of the Chamber are, with as dimensions of the chamber its length and its diameter and from this the volume of the as upright Circular cylinder formed chamber can be determined. 24. The method according to claim 23, characterized in that the normal vibration mode with which the electromagnetic Field is generated, is chosen so that interference due to the type, shape and surface of the chamber can be reduced to a minimum. 25. The method according to claim 24, characterized in that the selected oscillation mode is the mode TE ₁₁ -. 26. The method according to claim 24, characterized in that the selected oscillation mode is the TE ₁₁₂ mode. 27. The method according to claim 24, characterized in that the selected oscillation mode is the mode TM _01Q \ Lst. 130062/0716