HK1110381B - A method and apparatus for proving flow meters - Google Patents

Description

Method and apparatus for calibrating a flow meter

Technical Field

The present invention relates to the field of flow meters, and in particular, to calibrating flow meters.

Background

Coriolis (Coriolis) flow meters are "synthetic pulsers" defined by the American Petroleum Institute (API). This simply means that it takes a finite amount of time to calculate the flow that has passed through the meter. In a typical coriolis flow meter, it takes a brief but measurable amount of time to calculate the flow that has flowed through the meter. This results in a time delay between the actual flow and the measured flow. The result of the time delay is that the flow reading is offset from the actual flow or delayed by the time delay. In most applications, this delay does not cause a problem because at the end of the batch (batch), after a short wait, the correct value of flow through the meter is available.

Calibration (proving) is a method of calibration in place (calibration) in which a known volume flows through the meter and is compared to the meter measured flow. The prover may be stationary, e.g., permanently mounted adjacent to the flow meter, or onboard, so that the prover can calibrate multiple flow meters. A typical prover is a device having a pipe (104) with a known inner diameter. A ball (ball) or piston (102) slides into the conduit (104) and passes through two sensors (S1, S2) or detectors. The first sensor (S1) sends a signal to the computer of the prover to begin counting pulses from the calibrated flow meter. The pulse is generally proportional to the volumetric flow rate. The second sensor (S2) sends a signal to the prover to stop counting pulses from the calibrated flow meter. The volume inside the pipe between the two detectors is known and is often compensated for pressure and temperature. The total volume between the two detectors is compared to the number of pulses from the flow meter and the flow meter factor is determined therefrom. The flow meter factor is simply a correction factor applied to the output of the flow meter. The measurement time for a known volume through the meter is 0.5 to 60 seconds, depending on the prover volume and flow rate used. This procedure works well for many types of flow meters (impeller flow meters, PD flow meters, etc.), but there may be some problems associated with synthetic pulse devices due to the time delay in measuring the flow. Because of the time offset between the measured flow and the actual flow, the prover compares the flow volumes from different times.

Any change in flow rate during the calibration test will result in a difference between the measured flows at different times. One reason for the change in flow rate is when the prover actuates the ball or piston. Pressure changes may occur when the ball or piston is actuated, resulting in a change in flow rate. The prover measures a new flow rate and the coriolis flow meter measures a weighted average of the old flow rate and the new flow rate, causing an error between the two measurements. The prover often has a length of tubing through which the ball or piston must travel before crossing the first detector. This length of pipe is commonly referred to as "pre-run". The pre-run length is equivalent to a fixed volume. The prerun time depends on the flow rate. At high flow rates, the uptime may not be long enough to completely stabilize the flow rate. Even at low flow rates, the flow rate through the prover may not be completely stable. This is not a problem for many types of flow meters, but can lead to errors when calibrating a synthetic pulse device.

There is therefore a need for a system and method for calibrating a synthesized pulse device.

U.S. Pat. No.5072416 entitled "Method and apparatus for calibrating a flowing meter a master meter and a pro" discloses the following. A dual cycle prover is used to calibrate a flow meter requiring a long calibration run. A small volume or piston prover is coupled in series with a main flow meter and a flow meter under test, the main flow meter is then calibrated according to the small volume prover, and the flow meter under test is calibrated according to the main flow meter. The signal processor combines the calibration periods to associate the small volume prover with the flow meter under test. The density meter is connected to the flow path, and if the flow meter under test is a mass flow meter, the signal processor uses the density measurements to correlate the mass flow rate of the flow meter under test with the volumetric flow rate of the primary flow meter.

U.S. patent No.5774378 entitled "Self-equivalent sensors" discloses the following. The sensor provides a measurement and information about the validity of the measurement. The sensor includes a transducer for generating a data signal related to a value of a variable, and a transmitter for receiving the data signal and generating an output signal. The transmitter generates a first output signal related to the value of the variable. The transmitter also generates a second output signal based on a dynamic uncertainty analysis of the first output signal.

U.S. patent publication No.20020198668 entitled "System and method for a mass flow controller" discloses the following. A system and method for controlling a mass flow controller to have a constant control loop gain over a variety of different types of fluids and operating conditions, and for configuring the mass flow controller to operate at different fluids and/or operating conditions than those used in the production of the mass flow controller. Additionally, the system and method include providing control by reducing hysteresis in a solenoid actuated device by providing a non-operational signal to the solenoid actuated device.

Disclosure of Invention

A method and apparatus for determining a time delay between an actual flow and a measured flow in a flow meter is disclosed. The time delay is used to offset the flow rate measured by the flow meter to correspond to the actual flow rate measured by the prover. This allows for an accurate comparison between the flow measured by the flow meter and the flow provided by the prover.

One aspect of the invention includes a method comprising:

determining a time delay between an actual flow rate and a measured flow rate within the flowmeter;

providing the time delay to a calibration device for calibrating a flow rate of the flow meter.

Preferably, the method further comprises wherein the time delay is determined by introducing a disturbance in the flow and measuring how long the disturbance is detected by the flow meter.

Preferably, the method further comprises wherein the time delay is determined by calculating a delay due to electronics and filtering within the flow meter.

Preferably, the method further comprises wherein the flow meter is a coriolis flow meter.

Another aspect of the invention includes a method comprising:

providing a flow rate of a known quantity of material to a flow meter for a first period of time;

receiving measured flow data from the flow meter over a second time period;

receiving a delay time from the flow meter, wherein the delay time is a time difference between a measured flow rate and a provided flow rate within the flow meter;

the measured flow data from the flow meter is offset by the delay time.

Preferably, the method further comprises wherein the offset amount is a function of an amount of damping within the flow meter.

Preferably, the method further comprises wherein the offsetting is performed by delaying start and stop signals indicating the start and end of the first time period.

Preferably, the method further comprises wherein the flow meter is a coriolis flow meter.

Another aspect of the invention includes a method comprising:

setting a plurality of measurement parameters for the flow meter;

a time delay between an actual flow rate and a measured flow rate within the flow meter is determined when the flow meter is operating at the set measurement parameter.

Preferably, the method further comprises, wherein the plurality of measured parameters includes a degree of damping and a mode of operation.

Preferably, the method further comprises, wherein the delay time is provided to a prover used to calibrate the flow meter.

Preferably, the method further comprises wherein the flow meter is a coriolis flow meter.

Preferably, the method further comprises wherein the time delay is determined by introducing a disturbance in the flow and measuring how long the disturbance is detected by the flow meter.

Preferably, the method further comprises wherein the time delay is determined by looking up the time delay in a table mapping the plurality of measurement parameters to time delay.

Another aspect of the invention includes a prover comprising:

a pipe portion having a known diameter;

a first sensor located at a first position on the pipe portion;

a second sensor located at a second location on the pipe portion;

a device configured to move inside the pipe portion between the first and second positions;

the first and second sensors configured to transmit first and second signals, respectively, when the device passes through the first and second positions;

a flow computer configured to receive first and second signals from the first and second sensors;

the flow computer is further configured to receive a flow measurement from a flow meter to be calibrated;

the flow computer is further configured to receive a delay time from the flow meter to be calibrated and to offset the received flow measurement by the delay time relative to the first and second signals.

Preferably, the offset is performed within the prover by delaying the first and second signals from the first and second sensors.

Another aspect of the invention includes a coriolis flow meter comprising:

a conduit configured to contain a flowable material;

at least one driver configured to vibrate the catheter;

a sensor configured to measure a position of the vibrating conduit;

an electronic device configured to receive a measurement location of the conduit and convert the measurement location to a measurement material flow rate;

a storage area configured to store a delay time between at least one actual material flow through the conduit and a measured material flow through the conduit.

Preferably, the electronic device is configured to operate in at least two modes having different operating speeds, and different delay times for the two different modes are stored in the storage area.

Preferably, the electronic device is configured to operate with at least two degrees of damping, and a different delay time for the at least two degrees of damping is stored in the storage area.

Another aspect of the invention includes a coriolis flow meter comprising:

means for vibrating a conduit containing a flowing material;

means for measuring the phase of the vibrating conduit;

means for converting the measurement phase into a flow measurement;

means for storing a delay time representing a time delay between actual flow through the meter and flow measurement.

Drawings

FIG. 1A is a block diagram of a calibrator at the beginning of a measurement cycle;

FIG. 1B is a block diagram of the calibrator at time T1 during a measurement cycle;

FIG. 1C is a block diagram of the calibrator at time T2 during a measurement cycle;

FIG. 2 is a graph of a typical response obtained for a step change in flow rate in a Coriolis flowmeter;

FIG. 3 is a block diagram of the electronics for a typical Coriolis flowmeter;

FIG. 4 is a plot of flow rate versus time for flow measured using a Coriolis flowmeter and flow measured using a precision flow platform (flow stand) and high speed electronics; and

fig. 5 is a plot of flow rate versus time for flow measured using a coriolis flow meter with a damping change of 0.2.

Detailed Description

Fig. 1 through 5 and the following description depict specific examples to teach those skilled in the art how to make and use the best mode of the invention. For the purpose of teaching inventive principles, some conventional aspects have been simplified or omitted. Those skilled in the art will appreciate variations from these examples that fall within the scope of the invention. Those skilled in the art will appreciate that the features described below can be combined in various ways to form multiple variations of the invention. Accordingly, the present invention is not limited to the specific examples described below, but is defined only by the claims and their equivalents.

The prover typically begins a measurement cycle by beginning to flow material into the flow meter to be calibrated. Once the flow stabilizes, the prover releases a ball or piston which travels down a length of tubing of known diameter (see fig. 1A). After the ball or piston travels a pre-run length (if any), it hits the first sensor at time T1 (see fig. 1B). This triggers the flow computer inside the prover to begin counting pulses from the flow meter (i.e., measuring flow). When the ball or piston reaches the second sensor at time T2, the flow computer stops counting pulses from the flow meter (see fig. 1C). The pulses counted between times T1 and T2 are the amount of flow measured by the meter. The known volume in the conduit between the two sensors is the actual flow volume. The actual flow quantity is compared to the measured flow quantity to determine a flowmeter calibration. In an exemplary embodiment of the invention, a delay δ is added to the first and second sensor signals. In this embodiment, the sensor signal is not sent until after a delay time δ when the ball or piston reaches the first sensor. The prover does not begin counting pulses from the flow meter until after the sensor signal is sent. When the ball or piston reaches the second sensor, the sensor signal is not sent until after a delay time δ. The calibrator does not stop counting pulses until after the second sensor signal is emitted. The total time TT counted for the pulses remains the same as TT 1-T2 (T1+ δ) - (T2+ δ), while the measurements received from the meter have been offset by an amount δ.

In another exemplary embodiment of the invention, the delay time δ will be used by the flow computer within the prover. The flow computer will receive the first and second sensor signals without delay, but will not start or stop counting pulses from the flow meter until after the delay time δ has elapsed.

In each case, the actual flow from the prover is compared to the measured flow from the flow meter that is offset in time by an amount δ. When the equivalent delta corresponds to a time delay between the measured flow and the actual flow within the meter, then the prover compares the actual flow from the prover to the actual flow through the meter. In said another way, the prover will measure the volume at the same time.

The delay time δ will depend on the type or modification of electronics within the meter, flow rate, meter damping, vibration mode used by the meter, meter geometry, etc. The time δ may be different for different flowmeters and for different provers. In an exemplary embodiment of the invention, a time delay of the prover is set prior to calibrating the flow meter. In one exemplary embodiment of the invention, the user may select the type of flow meter being calibrated. The user is prompted to enter the delay time when a flow meter type having a non-zero delay is selected for testing. In another exemplary embodiment of the invention, the flow meter to be calibrated may interact with the prover and automatically provide the appropriate delay time to the prover.

The delay time of a coriolis flow meter may be calculated or may be measured. In order to calculate the delay, the cause of the delay must be clear. Fig. 2 is a typical response curve of a coriolis flowmeter measuring a step change in flow rate versus actual flow rate. The measurement response may be divided into two parts. The first portion of time t1 is a fixed processing delay during which there is no change in the measured output from the coriolis flow meter. The first time period t1 is typically due to an electronic device delay. The second part time t2 of the response curve is approximated as an RC delay function, which is an exponential function. The second time period t2 is generally due to filtering delay. The total delay (202) between the change in actual flow and the change in measured flow is the sum of times t1 and t 2.

Fig. 3 is a block diagram of the electronics of a coriolis flow meter. Core processor 302 receives analog signals from sensors (not shown) on the coriolis flowmeter. An analog-to-digital (A/D) converter (304) samples the analog signal and outputs a digital representation of the analog signal. The digital signal is processed within a Digital Signal Processor (DSP) (306). The signal is then filtered by a recursive damping filter 308. The user can vary the amount of damping used by the recursive damping filter 308. The signal is sent from the core processor to the transmitter (310) across the MODBUS (314) link. An output processor (312) within the transmitter (310) converts the signal into a user output signal. The fixed processing delays in the response curves include the processing time of the core processor (302) and the DSP (306), the inter-processor communication delays across the link 314, and delays due to the amount of damping used within the system. The main delays in the system are due to the signal filtering and sampling rate. Other electronic device configurations are possible. For example, the core processor and the transmitter may be combined into one unit.

Calculating the signal delay due to filtering is well known in the art. For example, the delay of an 8 th order elliptic filter with a cut-off frequency of 1500Hz, a sampling frequency of 48KHz, and a sampling rate of 12 can be decomposed into two parts, a sample delay and a group delay. The sampling delay of the filter can be calculated as: the sampling delay is 12(1/48KHz) 0.25 ms. The group delay can be calculated as:the total delay through the filter is the sum of the sample delay and the group delay, or 0.25+ 0.67-0.92 ms. For a multistage filter, calculating each stageThen the total delay is the sum of the delays of the stages. To get a complete system response, the filter response time must be added to the delay time. The filter response time may be calculated to reach 63% of the maximum response, to reach 100% of the maximum response, etc. For the filters from the above examples, the filter response time is 0.87ms (up to 63.2%) and 1.02ms (up to 100%). The total delay incurred due to the filtering of the exemplary 8 th order elliptic filter is therefore: the total delay is 1.79ms for 63.2% response and 1.94ms for 100% response. Calculating the delay through other stages of the electronic device is also well known in the art and depends on the operating speed of the electronic device, the amount of data to be transmitted, etc. Some coriolis flow meter electronics can operate in two modes. The electronic device has a normal mode operating at 20Hz and a "special" mode operating at 100 Hz. When in the special mode, the delay time is typically shorter due to the faster operating speed. For some coriolis flowmeters, the amount of damping employed within the filter may be varied. Any change in the amount of damping will affect the delay between the actual flow and the measured flow.

In one embodiment of the present invention, a coriolis flowmeter that can adjust damping will have different delay times for different amounts of damping. This different delay time can be calculated each time the damping is changed. In another embodiment, there will be a table in which the pre-calculated delay times for each different amount of damping can be obtained.

Another method of determining the delay time between a change in the amount of traffic and measuring the amount of traffic is to measure the delay. Fig. 4 is a plot of flow rate versus time for a flow measured using a coriolis flow meter and for a flow measured using a precision flow platform and high speed electronics. It can be seen that the flow measured by the high speed pressure transducer exhibits a substantially step change in flow rate. The measured flow rate (shown as analog output) from the coriolis flow meter remains constant for about 0.4 seconds. The measured flow took about 1.2 seconds to reach 63% of the maximum flow and about 1.8 seconds to reach 90% of the maximum flow. Fig. 4 is a plot of flow rate versus time for a measured flow when the coriolis flow meter is used to change the damping to 0.2. For a coriolis flow meter, the response time to reach 63% of the maximum flow rate has been reduced to about 0.5 seconds.

Measurements may be made for each degree of damping and for different modes of operation (i.e., normal and special modes). The delay times for the different modes and different damping factors may be compiled into a table and included in the coriolis flow meter. Using this delay time, the prover can offset the measured flow to correspond to the flow provided, allowing for more accurate calibration of the flow meter.

In the above description, the present invention is described using a coriolis flow meter. It is known to those skilled in the art that the present invention may be applied to other synthetic pulse devices and is not limited to coriolis flowmeters.

Claims (11)

1. A method of calibrating a coriolis flowmeter by determining a time delay (202) between an actual flow rate and a measured flow rate in the coriolis flowmeter, characterized by:

providing the time delay from the coriolis flow meter to a calibration device;

the coriolis flow meter is calibrated using the calibration device using the time delay.

2. The method of claim 1, wherein the time delay is determined by introducing a disturbance in the flow and measuring how long the disturbance is detected by the flow meter.

3. The method of claim 1, wherein the time delay is determined by calculating delays due to electronics and filtering within the meter electronics.

4. The method of claim 1, wherein calibrating the coriolis flow meter comprises:

providing a flow rate of a known quantity of material to the coriolis flow meter for a first period of time;

receiving measured flow data from the coriolis flow meter over a second time period;

receiving a delay time from the coriolis flow meter, wherein the delay time is a time difference between a measured flow rate and a provided flow rate within the coriolis flow meter;

offsetting the measured flow data from the coriolis flow meter by the time delay.

5. The method of claim 4, wherein the amount of offset is a function of an amount of damping within the flow meter.

6. The method of claim 4, wherein the shifting is performed by delaying start and stop signals indicating the start and end of the first time period.

7. A prover, comprising: a portion of pipe (104) having a known diameter; a first sensor (S1) located at a first position on the pipe portion; a second sensor (S2) located at a second position on the pipe portion; a device (102) configured to move inside the pipe portion (104) between the first and second positions; the first and second sensors configured to transmit first and second signals, respectively, when the device passes through the first and second positions; a flow computer configured to receive first and second signals from the first and second sensors, the flow computer further configured to receive a flow measurement from a coriolis flowmeter to be calibrated, characterized by:

the flow computer is further configured to receive a delay time from the coriolis flowmeter to be calibrated and to offset the received flow measurement by the delay time relative to the first and second signals.

8. The calibrator according to claim 7, wherein said shifting is performed by delaying said first and second signals from said first and second sensors.

9. The prover of claim 7 wherein the prover receives the delay time having a first value when the coriolis flow meter is operating in a first mode and a second value when the coriolis flow meter is operating in a second mode.

10. The prover of claim 9 wherein the coriolis flow meter operates at a first speed in the first mode and at a second speed in the second mode.

11. The prover of claim 9 wherein the coriolis flow meter has a first damping rate in the first mode and a second damping rate in the second mode.