GB2347747A - Method and means for measuring fluid flow - Google Patents

Abstract

When measuring the rate of flow of fluid, e.g. water, along a line (10) which is subject to turbulence, some of the fluid is withdrawn at a point (11) and passed through a heater (13) before being reintroduced into the flow at an injection point (14). The resultant temperature increase ( Š T) is mentioned at a point downstream of the injection point (14) and the flow rate (Q) is calculated using the equation Q = P<SB>1</SB> / C<SB>p</SB>. p . Š T, where P<SB>1</SB> is the power input, C<SB>p</SB> is the specific heat capacity of the fluid, and p is the density of the fluid.

Description

METHOD AND MEANS FOR MEASURING FLUID FLOW Field of the Invention This invention relates to a method and means for measuring fluidflow.

Many different types of flowmeters are available for many different applications, but depend on obtaining laminar or steady flow conditions over a reasonable length of the pipe or conduit through which the fluid in question in flowing. For water flowmeters, the requirement may be that a straight length of several metres is required, but this may not always be available when, for example, the rate of flow of water issuing from a bore hole is being measured.

It is accordingly an object of the present invention to provide a method of measuring fluid flow which can be used for measuring fluid flow under conditions of turbulence.

It is also an object of the present invention to provide an improved means for measuring fluid flow, which measuring means can be used for measuring turbulent fluid flow.

A further object of the invention is to provide a method of and means for measuring fluid flow which can be used for measuring either turbulent fluid flow or laminar fluid flow.

Summary of the Invention According to a first aspect of the present invention there is provided a method of measuring the rate of flow of fluid through a fluid flow line, which method involves introducing heat to the fluid at a measurable rate at a first predetermined point in said flow line and measuring the resultant temperature increase at a second predetermined point downstream of said first predetermined point.

According to a second aspect of the present invention there is provided means for measuring the rate of flow of fluid through a fluid flow line, such means including means for introducing heat to the fluid at a measurable rate at a first predetermined point in said flow line, and means for measuring the resultant temperature increase at a second predetermined point downstream of said first predetermined point.

The introduction of heat to the fluid is preferably effected by withdrawing some of the fluid from the flow line at an intake point upstream of said first predetermined point, passing the fluid through a heater and returning the heated fluid to the flow line at said first predetermined point.

The rate of addition of heat to the fluid is preferably measured by measuring the rate at which heat is supplied to the fluid by the heater.

The resultant temperature increase is preferably measured using two temperature to current transducers connected in series.

It is important that uniform heating of the fluid across the cross-section of flow is obtained. This can be achieved either by carrying out the measurement at a position in the fluid flow line in which turbulent flow conditions are obtained or by providing a heating arrangement in which the heat which is added is introduced across the cross-section of flow.

Brief Description of the Drawings Figure 1 shows a circuit for measuring the temperature difference between two points in a fluid flow line, one upstream of the point at which heat is added and the other downstream thereof, Figure 2 shows a voltage measuring circuit, and Figure 3 shows a typical layout for carrying out a flow measurement utilising the circuit of Figure 1.

Description of the Preferred Embodiment As shown in Figure 1, two AD590 temperature to current transducers T1 and T2 are connected in series between voltages V+ and V-relative to a quality ground (represented by a dagger symbol). The transducers T1 and T2 are housed in two separate probes, each in good thermal contact with the fluid flowing through the flow line upstream and downstream of the point at which controlled addition of heat is to be effected. The fluid is travelling through the flow line under conditions of turbulent flow produced, for example, by the presence of a bend just after the point at which heat is added so that, by the time that the fluid reaches the location of transducer T2, the temperature of the fluid is constant across the cross section of the flow line, i. e. the temperature difference which is measured is directly related to the quantity of heat which is added and the quantity of fluid flowing through the flow line per unit time.

The probes within which the transducers T1 and T2 are contained may also include pressure transducers if measurements of the pressure of the fluid are required in addition to temperature measurements.

Each AD590 transducer T1, T2 regulates a current that is linearly proportional to absolute temperate in degrees Kelvin with a constant of proportionality that is nominally 1 pLA/ K.

The choice of a current-based transducer system is very beneficial as it overcomes cable losses experienced by voltagebased systems. This thus allows any length of transducer cable to be considered, provided that sufficient potential difference, i. e. voltage, is available to drive the currents required.

Another benefit arising from the use of a current-based system is that it is far less susceptible to electrical interference, which can be a problem with long leads around motors and generators or other electrically noisy environments.

The voltages V+ and V-are stabilised using a circuit containing four resistances r and a connection to a reference voltage Vref but other methods of stabilisation may be employed.

Stabilisation is necessary to remove secondary errors in the characteristics of the transducers T1 and T2, which exhibit a small variation of current with voltage.

The junction between the two transducers T1 and T2 is connected to op-amp A1, wired as a current to voltage converter.

The op-amp A1 may altematively be integrated into a complex semiconductor circuit, such as an analogue to digital converter. The resulting output from the op-amp A1 is an output voltage Vout that is proportional to the difference in currents in the transducers T1 and T2 and hence to AT, i. e. the difference in temperature in the fluid between the two points at which the transducers T1 and T2 are located. Vout can then be measured by any voltmeter including, for example, an analogue to digital converter which is linked to a computer so that further computations can be effected if required.

A resistance R is connected across the op-amp A1. The value of this resistance R determines the sensitivity of the system.

For example, with the nominal 1pA/ K., a 1MQ resistor will give a sensitivity of 1 mV of Vout per mK of AT. If the two probes are at the same temperature, there will be no current difference and Vout will be zero.

A major benefit of the particular circuit shown in Figure 1 is that the op-amp A1 forces its-ve input to quality ground voltage. The two cables joining the measurement unit containing the two probes can, therefore, be shielded at ground, as indicated at S1 and S2 making the provision of shielding easy and safe. In addition, there will be no voltage between the sensor connections to the opamp A1 and the cable sheath and, since there is no voltage, there will be no leakage current. This safeguards the accuracy of the measurement of the difference current. There will also be no problem due to leakage from V+ and V-to the shield, because these voltages are stabilised.

All AD590 transducers are not, however, perfectly matched.

Their sensitivity varies slightly from 1 sLAJ K, from one component to the next. This is overcome by calibration, which therefore becomes a preferred part of the invention. The method of calibration is as follows : Place the two transducers T1 and T2 in a controlled"constant temperature"bath so that they are both at the same temperature and so that any output Vous must then correspond to some mismatch of the transducers T1 and T2.

This error is recorded as a function of temperature over the range of temperatures which, in use, are likely to be encountered and subsequently used to correct the readings which are obtained. Any suitable method may be used to measure the temperature of the"constant temperature"bath.

This procedure can also be used to monitor the long-term stability of a pair of transducers.

The temperature measurement is, however, preferably effected using the circuitry shown in Figure 2. This includes a shunt resistor SR1, SR2 in series with each transducer T1, T2, with a differential amplifier A2 or A3 to measure the voltage across each shunt resistor SR1, SR2. This voltage is proportional to Temperature. The provision of the differential amplifier A3 is optional, but it helps to maintain symmetry and hence reduce errors.

The shunt resistors SR1 and SR2 need to be low value resistances so as not to compromise the shielding advantages referred to above. For 1 00Q the voltage is only about 30 mV.

As well as possible asymmetries in the transducers T1 and T2, there are other potential sources of errors in the electronics.

Long-term stability and temperature dependence of active components are the main practical problems, while noise introduces a practical limit to the sensitivity of measurement.

The main source of amplifier error is due to the temperature dependence of offset in amplifier A1. This should, therefore, be a high-gain chopper amplifier whose offset is negligible. Amplifiers A1, A2 and A3 should all have negligible input currents.

There are two types of noise, i. e. interference and inherent thermal noise in the components. The quality of cable screening is important. Filters can be added to the transducer cable connections or in the cables themselves to reduce susceptibility to radiated interference, and the cables need to be connected in such way as to avoid earth loop pick-up from the device being monitored. The main source of electronic noise is in the AD590 transducers themselves.

Signal averaging can reduce the effects of both types of noise but there is a limit to what can be achieved when the parameters are themselves changing. For steady parameters and a measurement time of about one second, the practical limit of sensitivity is of the order of 0.5 mK.

Figure 3 shows a layout utilising the circuit of Figure 1 for measuring the rate of flow of water through a pipe 10 having a pair of right angle bends such that the flow through the pipe is turbulent and thorough mixing across the cross-section of the pipe is obtained. Water is withdrawn from the pipe 10 at an intake point 11 by means of a pump 12, passed through a heater 13 and returned to the pipe 10 at an injection point 14 downstream of the intake point 11. In the particular arrangement shown in Figure 3, the injection point 14 is located intermediate the two right angle bends so that thorough mixing of the heated water with the main flow is obtained. In the arrangement shown by way of example, the intake point 11 is on the high pressure side of a pump 15.

The temperature of the water prior to the injection point 14 is measured at any one or more of a plurality of measurement points 20 to 28 and temperature measurements are taken at one or more of a plurality of measurement points 30 to 38 located on either side of an open gate valve 39. Transducer T1 will thus be located at any one of points 30 to 38 whilst transducer T2 will be located at any one of points 20 to 28. In order to avoid any errors as a result of variations in ambient conditions, measurements will normally be taken with each of the probes T1, T2 located at more than one of the possible points.

The rate at which power is supplied to the heater 13 is measured by means of a power meter 16.

The rate of flow is calculated as set out below : Power input = Mass of fluid x Specific Heat x Differential Temperature, i. e. P, = Q. Cp. p. AT, where P, = Power input (watts), Q = Flow Rate, Cp = Specific Heat Capacity of Fluid, p = Density of Fluid, and AT = Temperature Rise.

Thus Q = Pi/Cp. p. AT.

For water, p = 1 and, if the Power Input is in Kilowatts, Cp = 4.186.

In a typical example, the Power Input, Pi, is 9kW, and the temperature differential, AT, is 0.1 C. so that Q=9/4. 186x0. 1, i. e. Q = 21. 5 kg/sec.

When carrying out measurements to determine the temperature differential obtained as a result of the addition of heat at a predetermined rate, the addition of heat may be effected cyclically to compensate for any ambient effects. For example, heat may be supplied for fifteen seconds and then the heater switched off and kept off for say one minute before the heater 13 is turned on for another fifteen seconds and so on. A plot will be obtained of the temperatures at the two measurement points and of the differential temperature measurement obtained as described above with reference to the circuit of Figure 1.

Claims (9)

Claims :

1. A method of measuring the rate of flow of fluid through a fluid flow line, which method involves introducing heat to the fluid at a first predetermined point in said flow line and measuring the resultant temperature increase at a second predetermined point downstream of said first predetermined point.

2. Means for measuring the rate of flow of fluid through a fluid flow line, such means including means for introducing heat to the fluid at a measurable rate at a first predetermined point in said flow line, and means for measuring the resultant temperature increase at a second predetermined point downstream of said first predetermined point.

3. A method as claimed in Claim 1 or measuring means as claimed in Claim 2, in which the introduction of heat to the fluid is effected by withdrawing some of the fluid from the flow line at an intake point upstream of said first predetermined point and retuming the heated fluid to the flow line at said first predetermined point.

4. The invention of Claim 3, in which the rate of addition of heat to the fluid is measured by measuring the rate at which heat is supplied to the fluid by the heater.

5. The invention of Claim 4, in which the flow rate (Q) is calculated using the equation Q = P1/Cp. p. AT, where P, is the power input, Cp is the specific heat capacity of the fluid, p is the density of the fluid, and AT is the resultant temperature increase.

6. A method as claimed in Claim 1 or measuring means as claimed in Claim 2, in which the resultant temperature increase is measured using two temperature to current transducers connected in series.

7. The invention of Claim 6, in which calibration of the transducers is effected substantially as hereinbefore described.

8. A method of measuring the rate of fluid through a fluid flow line, particularly when the fluid is flowing under conditions of turbulence, said method being carried out substantially as hereinbefore described with reference to and as shown in Figure 3 of the accompanying drawings.

9. Means for measuring the rate of flow of fluid through a fluid flow line, particularly when the fluid is flowing under conditions of turbulence, said measuring means being constructed and arrange to operate substantially as hereinbefore described with reference to and as shown in Figure 3 of the accompanying drawings.