Introduction

The possibility of inducing voltages in liquids moving through magnetic fields was known by Faraday in 1832, but the first flowmeter-like device was reported by Williams in 1930. The first real advance in the subject came from the medical field where Kolin (1936, 1941) introduced many ideas which are now standard practice. In 1941 Thürlemann gave the first general proof, essentially of Equation (12.2) as set out later in this chapter (see also Thürlemann (1955) and Shercliff (1962) for more background).

The industrial interest arose in the 1950s with:

1. • the Tobiflux meter (Tobi 1953) in Holland for rayon viscose, sand-and-water and acid slurries;

2. • Foxboro, to whom the patent was assigned in 1952;

3. • the first commercial instruments in 1954 (Balls and Brown 1959).

4. • nuclear reactor applications;

5. • the work which resulted in an essential book by J. A. Shercliff (1962).

In this chapter we shall concentrate on the application of the flowmeter to fluids that are of low conductivity, such as water-based liquids (Baker 1982; see also 1985). The flowmeter has also been proposed for use with non-conducting dielectric liquids and some designs have been built for this purpose. This is briefly discussed in the appendix section 12.A.9. It has also been used with liquid metals and this is briefly discussed in the appendix section 12.A.10.

Operating Principle

We start with the simple induction, which occurs when a conductor moves through a magnetic field. Figure 12.1 shows a copper wire cutting the flux of a permanent magnet. The wire is moving in a direction perpendicular to its length and perpendicular to the magnetic field with velocity V, and the result is a voltage generated between its ends of value BlV where l is its length and B is the magnetic flux density.

Figure 12.2 shows a diagram with the essential features of an electromagnetic flowmeter. The liquid flows in a circular cross-sectional tube. A magnetic field is created across the pipe, usually by coils excited by an alternating current. The tube itself must be made from nonmagnetic material so that the magnetic field can penetrate the tube.

Introduction

There are broadly two concepts of thermal flowmeter now available for gas mass flow measurement, one of which is also applicable to liquids. I shall follow the useful terminology in ISO 14511:2001 for the two types and encourage the reader to make use of the ISO document. The first is the capillary thermal mass flowmeter (CTMF), which has broad applications in the control of low flows of clean gases, but which can also be used with a bypass containing a laminar element to allow higher flow rates to be measured. The arrangement of heaters and coils between the various manufacturers differs, but the basic approach is the same, with heat added to the flowing stream and a temperature imbalance being used to obtain the flow rate.

The second is the full-bore thermal mass flowmeter (ITMF), which is available as both insertion probe and in-line type. It has a widely used counterpart in the hot-wire anemometer for measurement of local flow velocity, but, as the need for a gas mass flowmeter has become evident, it has been developed as a robust insertion probe for industrial usage and then as the sensing element in a spool piece flowmeter. It has been produced by an increasing number of manufacturers in recent years as a solution to the need for such a mass flowmeter.

However, the advent of the MEMS (micro-electro-mechanical sensors) or CMOS (complementary metal-oxide semiconductor) designs has opened a new set of flowmeters based on silicon chip technology. Their operation is similar to the CTMF, but usually on a smaller scale.

Capillary Thermal Mass Flowmeter – Gases

In various examples of the CTMF, the gas flows through a very small diameter tube that has heating and temperature-measuring sensors. For larger flows, this is used with a bypass laminar flow element in the main gas stream. In many cases, the mass flowmeter is in one unit with the control electronics and the gas control valve. However, in this chapter our interest is in the meter itself. These meters are designed for clean dry gases.

Description of Operation

In this design of thermal flowmeter, the gas flows through a very small tube (possibly with ID in the range of 0.2 to 0.9 mm), with a sufficient length to diameter ratio to ensure fully developed laminar flow, and on the outside of which there are heating and temperature-sensing windings.

Introduction

We resort to probes for flow measurement for three main reasons:

1. i. To provide a low-cost method of flow monitoring;

2. ii. To provide an in situ calibration;

3. iii. To obtain fine detail of the flow in the pipe (velocity profile, swirl and turbulence).

For our present purposes, we shall not consider (iii). It is relevant to fluid mechanics research, mainly uses hot-wire and laser Doppler anemometers, and is, therefore, outside the scope of this book. We will concentrate on work relating to bulk flow measurement. In situ calibration techniques are covered in Chapter 4. In this chapter, we shall consider the various devices that have been used and are, or have been, commercially available either as a local flow monitor or as the probe for in situ calibration. However, before considering these, we need to understand what is being measured.

i. Ideally, the probe measures the local velocity and obtains a representative value of the mean velocity from the shape of the velocity profile. Alternatively, it may measure the complete profile.

ii. In practice, the probe may be in error because:

1. • it was calibrated in an approximately uniform profile or on the pipe axis but will need to measure velocity profiles that result in a significant variation of velocity across the sensing head;

2. • it will alter the flow by its presence, may shed vortices and may oscillate as a result; and

3. • the flow pattern will continue to change as the probe head is inserted further and further into the pipe and as it approaches the opposite side of the pipe.

In some cases, these flows may be analysed theoretically, but in most cases one would expect to calibrate the probe, and this calibration will be different at the pipe axis and at the pipe wall.

iii. The probe will reduce the flow area of the duct by its presence, and the area will continually decrease as the probe is inserted further. This blockage will need to be allowed for. The blockage will also differ between incompressible and compressible fluids.

Our object in using a probe is to obtain the best measure of the local flow rate that can be obtained. We shall need to remember both the limitations of the probe and the uncertainty of the relationship between the reading and the mean velocity.

This chapter contains some personal comments on research, manufacturing variation, quality, existing and new flow measurement challenges, micro-engineering devices and new techniques for existing and new flow metering concepts.

I pose the following questions with brief explanation based on my knowledge of the flow measurement industry.

Is there an Opportunity to Develop New Designs in Collaboration with the Science Base?

There is a need and an opportunity for the flow metering industry on one hand, and the science base as it relates to instrumentation on the other, to exploit effective means for technology transfer and to bridge the gap between them, which in the process will raise the profile of the instrumentation sector and ensure that government is aware of its importance. Past experience of encouraging collaboration suggests that industry and the science base do not always appreciate the value of working together. Fundamental and unsolved issues remain in the technology which would benefit from collaboration between industry and the science base. Such collaboration should be mutually beneficial in making the other party aware of important developments.

This gap could also be reduced by facilitating exchanges of people between industry and the science base. Industrial collaboration is an essential part of ensuring that academic engineering and applied science research is focused on the real problems of industry, and the cross-fertilisation will be very likely to generate new ideas.

Over the past 15 years or so, I have found close collaboration with several companies in research and teaching extremely rewarding.

Is Manufacture of High Enough Quality?

The quality of an instrument, and thus its ultimate accuracy, will clearly be affected by the manufacturing process. The manufacturing variation may have predictable consequences for the flowmeter accuracy. Slight variation may require calibration of each instrument. Instruments off the same production process might have different characteristics which may require small adjustments to the product before final calibration. Some changes may cause finished instruments to have different random errors.

The production of instrumentation is a special case in which every finished product may be measured (calibrated) and also the effect of the production process on the final accuracy of the instrument may, in some cases, be predicted and used to specify the production requirements.

Introduction

Background

Spirals, screws and windmills have a long history of use for speed measurement. Robert Hook proposed a small windmill in 1681 for measuring air velocity, and later one for use as a ship's log (distance meter). A Captain Phipps, in 1773, employed the principle that a spiral, in turning, moves through the length of one turn of the spiral to create a ship's log. Many centuries earlier than this, it appears that a Roman architect, Vitruvius, suggested a more basic form of this device.

In 1870 Reinhard Woltmann developed a multi-bladed fan to measure river flows (Medlock 1986). The device was a forerunner of the long helical screw–type meter still called after him and used widely for pipe flows in the water industry. The first modern meters, of the type with which we are mainly concerned, were developed in the United States in 1938 (Watson and Furness 1977; cf. Furness 1982). These were attractive for fuel flow measurement in airborne applications. They consisted of a helically bladed rotor and simple bearings. Improved sleeve bearings were developed for longer life with hardened thrust balls or endstones to withstand the axial load. An alternative developed over several years and patented by Potter (1961) was to profile the hub of the rotor. The pressure difference caused by the hub shape, rather than the thrust on the bearings, may have held the rotor against the axial drag forces due to:

1. • the pressure balance across the rotor and/or:

2. • the spinning of the rotor on a film of the liquid, flowing upstream through the annular passage between the stationary axle and the moving rotor.

This allowed the rotor to run on a single journal bearing.

Qualitative Description of Operation

The turbine consists of a bladed rotor which turns due to the flow in the pipe. In most of the designs to be discussed, the rotor is designed to create the minimum disturbance as the oncoming flow passes round it. Ideally it cuts perfectly through the fluid in a helix so that every revolution of the helix represents one complete axial length of the screw and hence a calculable volume of the fluid. In practice drag forces slightly retard the rotation.

This is a book about flow measurement and flowmeters written for all in the industry who specify and apply, design and manufacture, research and develop, maintain and calibrate flowmeters. It provides a source of information both on the published research, design and performance of flowmeters, and also on the claims of flowmeter manufacturers. It will be of use to engineers, particularly mechanical and process engineers, and also to instrument companies’ marketing, manufacturing and management personnel as they seek to identify future products.

I have concentrated on the mechanical and fluid engineering aspects and have given only as much of the electrical engineering details as is necessary for a proper understanding of how and why the meters work. I am not an electrical engineer and so have not attempted detailed explanations of modern electrical signal processing. I am also aware of the speed with which developments in signal processing would render out of date any descriptions that I might give.

I make the assumption that the flowmeter engineer will automatically turn to the appropriate standard and I have, therefore, tried to minimise reproducing information which should be obtained from those excellent documents. I recommend that those involved in new developments should keep a watching brief on the regular conferences which carry much of the latest developments in the business, and are illustrated by the papers in the reference list.

I hope, therefore, that this book will provide a signpost to the essential information required by all involved in the development and use of flowmeters, from the field engineer to the chief executive of the entrepreneurial company which is developing its product range in this technology.

In this book, following introductory chapters on accuracy, flow, selection and calibration, I have attempted a clear explanation of each type of flowmeter so that the reader can easily understand the workings of the various meters. I have, then, attempted to bring together a significant amount of the published information which enlightens us on the performance and applications of flowmeters. The two sources for this are the open literature and the manufacturers’ brochures. I have also introduced, to a varying extent, the mathematics behind the meter operations, but to avoid disrupting the text, I have consigned some of this to the appendices at the end of the chapters.

Introduction

The previous chapters have been mainly concerned with devices, the designs of which have been defined by various standard specifications. In this chapter, we consider other mainly proprietary meters, which depend on the changing flow pattern and which essentially sense momentum of the flow. The output is in some cases given by a differential pressure measurement and in others by a position or force measurement. This means that the user must be aware of the possible effect of density on the results. Viscosity will also affect some readings.

The somewhat arbitrary order of consideration and the uneven amount of detail in this chapter reflect the available literature and my experience. Thus the variable area meter takes up a large proportion of this chapter.

The meters considered in this chapter are:

•variable area (VA) meters, which depend on gravity to oppose the movement of the float and consist of two main types:

1. • those with a tapered tube and a float with a fixed metering edge and

2. • those with an orifice with a fixed metering edge and a moving tapered plug;

• spring-loaded profiled plug in an orifice;

• target (drag plate) meter sometimes spring loaded;

• Venturi-type meters claiming a low loss, such as Dall, Epiflo, Gentile and Low-Loss;

• wedge meter;

• V-cone meters;

• meters using a bypass with an oscillating vane, a Pelton or a VA meter in it;

• slotted orifice meter;

• pipework features used as meters such as inlets and bends;

• averaging pitots under various names (Annubar, Torbar etc.); and

• laminar (viscous) flowmeters.

Variable Area Meter

The variable area (VA) meter is sometimes known as a Rotameter, but this is a trade name of a particular company and I shall, therefore, refer to it as the variable area meter. In the following sections on the VA meter, I have avoided history, theory or other items, which have been consigned to Appendix 8.A. One way to view the VA meter is as a variable orifice meter with a fixed pressure drop resulting from the weight of the float.

Introduction

The material in this chapter is based on the paper by Baker and Morris (1985) on positive displacement (PD) meters, which in turn has been updated with more recent industrial and published material and extended to gases. P. D. Baker's (1983) paper provided some additional useful information on these meters. The main types of liquid meter were given by Barnes (1982), Hendrix (1982), Henke (1955), and Gerrard (1979) (cf. Mankin 1955). The reader might refer to API (2005) and similar documents. Morton, Hutchings and Baker (2014a) provided additional references.

At least four of the meter designs to be discussed have been around for over 100 years. The nutating disc flowmeter for liquids was developed in 1850. The oscillating circular piston meter appeared in the late 19th century (Baker 1998).

The measurement of gas has depended, from an early date, on two types of positive displacement meter: the wet gas meter of high accuracy and credited to Samuel Clegg (1815), and the diaphragm meter of lower performance but greater range for which William Richards (1843) should take the credit and which has been widely adopted for the domestic gas metering market. The rotary PD gas meter appears to have shown good performance.

Background

The concept of carrying known volumes of fluid through a flowmeter is a short step from the use of a discrete measure such as a bucket or measuring flask. Thus in each of the designs described later, the flow enters a compartment that is as tightly sealed as is compatible with relative movement of adjacent components. A knowledge of how many of these compartments move through the flowmeter in one rotation of the shaft leads to a knowledge of the flow rate for a certain rotational speed. A simple theory is developed in Appendix 9.A.1. Clearly every leakage path will reduce or increase the amount carried and cause an uncertainty in the measurement. We can make an approximation of the leakage flows in the simple model. In addition, pressure and temperature will distort the volume and may cause small errors, and we may need to develop compensation for these.

Introduction and Some Early References

In the first edition of this handbook, I mentioned that flowmeters depending on magnetic resonance (MR) had been suggested, but that I was not aware of an instrument on the market. Nuclear magnetic resonance (NMR now, apparently preferably referred to as MR) essentially marks the fluid and measures transit times (Genthe 1974; King and Rollwitz 1983). Gol'dgammer, Terent'ev and Zalaliev (1990) showed that the magnetic resonance signal amplitude of a gas-liquid mixture in diamagnetic fluids depended on the liquid content regardless of the physicochemical properties of the medium and conditions of flow. This linearity was destroyed if paramagnetic centres were added to the fluid. De Jager, Hemminga and Sonneveld (1978) described their experimental data from water flow velocity measurements in the range of 0.5 to 5 mm/s by means of nuclear magnetic resonance using a sequence of inhomogeneous 180° pulses and a gradient in the stationary magnetic field. MR had been used in liquid-gas flows to measure the mean velocity of the liquid averaged over the liquid volume and also the average of the liquid fraction (Kruger, Birke and Weiss 1996). The combination provided the mass flow of the liquid. In tests using water as the liquid, the authors claimed to determine the mass flow to within ±5%.

Scott (1982), in “Developments in Flow Measurement–1”, comments that “Sensing of the flow relies on the application initially of a magnetic field to the nuclei of hydrogen or fluorine in the flow and the subsequent (downstream) removal of this influence. The nuclei enter the magnetic section with their individual magnetic orientations completely random and leave with their nuclear orientations aligned in the direction of the magnetic field. In the detector section which follows they approach a tagger coil and then a receiver coil. The former injects a short high intensity burst of radio frequency waves into the meter. This demagnetises a ‘window’ in the magnetised fluid and the window is discerned by the detector coil as a momentary drop in output signal. A knowledge of the distance between the tagger and the receiver, coupled with a time-of-flight measurement between them leads to the mean flow velocity.” The reader is referred to Scott's book for early references to this technique.

Introduction

Checking calibration is increasingly important and possibly mandatory. To achieve this there are, at least, four possible approaches:

1. 1. remove meter and recalibrate on a certified test stand;

2. 2. in situ calibration as discussed in Chapter 4. To obtain the required uncertainty, this may mean comparing the meter with a transfer standard in series with it in the line. This would require that installations of such a meter would be possible so as to check calibration (Chapter 4). But there may be major problems in so doing and limitations to the safety and accuracy of the process;

3. 3. use a clamp-on meter as transfer standard. This is an attractive option but, as indicated in Chapters 14 and 15, may not allow the level of uncertainty needed for many modern flowmeters;

4. 4. implement verification procedures.We shall review verification and then look at other possibilities.

Verification

This is, essentially, condition monitoring applied in a very precise manner to flowmeters. The procedure is, increasingly, a continuous one, which can trigger well-focussed maintenance.

It should be remembered that differential pressure devices, such as orifice meters with removable plates (Figure 5.16), have a form of in situ verification.

Probably the first modern meter to introduce an early version, simulation, was the electromagnetic flowmeter (EMFM). A signal simulating the flow signal was introduced at the electrodes and the ability of the amplifier to measure it correctly was checked. The idea of simulation possibly stemmed from the simplicity of Equation (12.2) and the sense that, if the performance of the amplifier which measures the induced voltage could be checked and the magnetic field measured, the meter's operation would be confirmed.

The pressure to achieve in situ calibration has caused flowmeter companies to develop simulation techniques and to extend them to all the modern types of flowmeter. The effect of this has been to explore the extent to which the new devices can be used as an interim verification between calibrations, allowing the period between calibration to be extended by testing and verifying all the major functions of flowmeters on site.

Introduction

The measurement of mass flow of multiphase fluids is likely to range across wide industry areas as diverse as food processing and subsea hydrocarbon extraction, but at present the technology appears mainly to be driven by the petrochemical industry.

The measurement also needs to take account of flows where the nature of the second component may or may not be known, and the proportions or indeed the existence of the second or even third component may not be known.

I am very conscious that others are more knowledgeable on multiphase flowmeters than I, and it is not appropriate to attempt to duplicate their published work on the subject. This is, therefore, a brief view of this important industrial area and a review of some published literature of which I am aware including some early references, together with pointers to key specialist books on the subject.

Some of the earliest data appear to be related to nuclear plants under emergency conditions. Extensive experiments were undertaken, and special designs and combinations of meter were tried in order to obtain a system which would signal fault conditions in the nuclear reactor duct flows (e.g. Baker 1977; Baker and Saunders 1977 in relation to sodium flows in fast breeder reactors). Some applications of traditional flowmeters to two/multiphase flows may be found in the chapters of this book specific to those meters.

Dykesteen (1992) identified the development of small oil wells in the North Sea where a platform to support a separator would be too costly as one reason for needing a compact and subsea multiphase meter. Subsea reservoir management needs to take place without the use of a separator, and where different satellite wells are operated under different licences, individual flow measurement will be required. The multiphase meter should, if possible, provide component fractions, component velocities and component densities.

An attempt was made (Baker 1991a) to draw together the references and the most useful data on the likely response of bulk flowmeters when placed in such flows (cf. Baker 1988b, 1989). Subsequently Whitaker (1993) reviewed further developments. In this chapter I have attempted to draw together more recent published work of which I am aware, but would particularly recommend to readers that they obtain one of the recent books specifically on multiphase flow measurement, for instance Falcone, Hewitt and Alimonti (2009), Schlumberger (2010) or Corneliussen et al. (2005).