Taking flow measurement to the next level

By Alan Jermyn   |   October 25, 2019

Avnet Abacus helps engineers to develop smart flow meters for monitoring water and gas consumption. Alan Jermyn, VP Marketing, looks at what’s driving demand for smart meters, how they work and how they’re built.

One of the key aspects of “smart cities” is the management of valuable resources such as water, gas and electricity. It’s becoming more critical by the day because, according to the World Health Organisation, over half of us now live in cities and that figure will rise to over 70% by 2050. Smart meters are therefore an essential part of smart cities. They allow both consumers and utility companies to monitor and manage resources, delivering cost-savings for consumers and, in the case of electricity and gas meters, reducing emissions to assist governments in meeting their CO2 emissions goals.

The European Commission predicts that there will be 200 million smart meters for electricity in the EU by 2020 and 45 million for gas. When it comes to water, a report by information company IHS Technology suggests that Europe is the fastest-growing market for two-way smart water meters. Here, one in four water meters will be “smart” by 2020. Smart water meters have been shown to reduce energy consumption in water/waste water operations by up to 30% and to cut water losses by up to 15%. In gas metering, smart meters are also useful for leak detection. Here, safety benefits are the primary concern but financial benefits are an added attraction.

From an engineering standpoint, measuring electricity consumption is simple – current sensing technology is mature, you know the voltage and can apply power factor correction. For fluids, both liquids and gases, it’s a little more complicated and until relatively recently most meters have been mechanical types which have to be read manually where they’re installed. Smart meters not only need to be more accurate and robust than these mechanical devices, they also need to be able to be interrogated remotely so that the costs of taking regular readings can be reduced. What’s more, many smart meters are being installed where a grid power supply is not available, so there’s a requirement for them to have very low power consumption so that they’ll operate for long periods from small batteries.

In the last few years a number of semiconductor companies have revolutionised the market for remote metering. They have designed dedicated integrated circuits for ultrasonic meters that can be used to measure the flow of liquid, gas and heat.  These devices eliminate mechanical flow meters and provide both greater accuracy and greater reliability. Utility companies breathed a sigh of relief at this technology advance and ultrasonic meters have become very popular.

There are two distinct approaches to ultrasonic flow measurement: time-of-flight (TOF) and Doppler Effect. Some smart meters are configurable to use either.
 

The time-of-flight (TOF) approach

The principles underpinning utrasonic TOF measurement can be traced back to a book entitled ‘The Theory of Sound’, written by Nobel Laureate physicist, J.W.S. (Lord) Rayleigh, back in 1877. The book describes how sound behaves when flowing through solids and fluids.

TOF flow measurement in pipes involves applying two (or more) piezoelectric transducers to a pipe at a specified linear distance apart.  Piezoelectric transducers of a particular size and type generate an ultrasonic signal from an applied alternating electrical voltage. Conversely, when they are subjected to an ultrasonic signal, they produce an alternating electric voltage in response.

In TOF measurement, ultrasonic signals are transmitted from one transducer received by the other. If there is no flow in the pipe, the transit time of the signals is the same in both directions. When there is a flow of gas or liquid in the pipe, the signal transit time is reduced when it travels in the direction of flow and extended in the other. The time difference, or phase shift, is directly proportional to the rate of flow.
 

Using the Doppler Effect

Proposed by Austrian physicist, Christian Doppler, in 1842, the Doppler Effect is the observed change in frequency of a wave as it moves relative to an observer. It appears to increase in frequency as it approaches the observer and decrease in frequency as it moves away. The effect can be observed for any type of wave, including ultrasonic ones.

With an appropriate arrangement of transducers, transmitted and received ultrasonic signals can be compared in the frequency domain, rather than in the time domain used in TOF measurement. When there is no flow in a pipe, the frequencies are the same. Any flow of water or gas produces a Doppler Effect and the frequency difference observed between the transmitter and receiver increases with flow rate. This frequency difference can then be converted into a DC level and then, using an ADC, into a digital signal for subsequent processing, transmission and display.

The Doppler Effect may be particularly valuable when measuring low flow rates, below 1 litre per hour for water, and the measuring installation can sometimes be smaller than that needed for TOF measurement. However, the most appropriate technology depends on the final application.
 

How to make a smart flow meter

As illustrated in Figure 1, the main elements of an ultrasonic water or gas flow meter are piezoelectric transducers with appropriate housing and connectivity, a transducer interface with host microcontroller to compute the flow rate, a power source that is independent of the grid and, in a growing number of applications, some form of low power wireless communications to communicate measurement data to the outside world.

Figure 1: The building blocks of a smart water or gas flow meter
 

Piezoelectric transducer construction

A typical piezoelectric transducer for flow measurement is shown in Figure 1. The piezoelectric elements are held within a housing, which is hermetically sealed at the point at which the signal cable emerges from the assembly. The cable and connector assembly is usually of a custom design, tailored to the specific application.

Figure 2: Simplified construction of a piezoelectric transducer assembly for water and gas flow meters
 

The analogue front end and digital processing

A number of semiconductor companies manufacture analogue front ends and microcontrollers that are suited to flow measurement applications. One popular device for TOF meters, due to its low power consumption and high degree of functional integration, including temperature measurement, is the Maxim MAX35103. It’s a single-chip, time-to-digital converter with 20ps measurement accuracy and capable of 2Hz to 16Hz measurement frequency in event timing mode. Current consumption for TOF measurement is 5.5µA. A typical application circuit is shown in Figure 2.

Figure 3: The MAX35103 is a highly accurate time-to-digital converter for use in water and gas time-of-flight flow meters
 

Adding wireless communications

Most new smart meters now have integrated wireless communications. Numerous wireless protocols are deployed with none yet achieving the status of a de-facto standard, at least not in Europe. SIGFOX, ZigBee, Wireless M-Bus, and 868 MHz ISM and cellular radios are all being used and there’s growing interest in Narrowband IoT (NB-IoT) – a protocol that’s designed to transmit short bursts of limited amounts of data over existing 2G, 3G or 4G cellular radio networks. What all these radios have in common is the need for an effective antenna. The more efficient the antenna, the lower the transmitter power needed to achieve an acceptable signal path. The fundamental choices are PCB antennas made from copper traces, chip antennas or external whip antennas. Whip antennas are the largest and may not be physically or economically the best option in every application but they do offer the highest performance. Figure 4 gives a brief summary of the pros and cons of each antenna type.

Figure 4: Size, performance and cost are the key factors in antenna selection for wireless smart meters
 

Powering smart meters

As mentioned earlier, remote smart meters need to be able to operate for a long time on small, grid-independent power sources. Ideally, they should run for 10 to 20 years between battery changes. It’s not just the total capacity of the primary battery that needs to be considered. The profile of how power will be drawn matters too, as do the environmental conditions under which the battery is required to operate. Lithium-based primary batteries with low self discharge are the preferred power sources for most metering applications but this is a broad category of batteries – there are several different chemistries to choose from. Experience suggests the lithium thionyl chloride batteries are best suited to most metering applications, millions have already been proven in the field since the 1990s, but it is vital to ensure that battery vendors can provide comprehensive and verifiable test results for the products on offer and that there is full traceability, right back to the raw materials used. Caveat emptor definitely applies to the battery industry, but you can find out more about test results and approvals in our standard li-ion / polymer batteries brochure.

The radios in smart meters create a demand for infrequent, short bursts of power, typically at currents of up 2 Amps, in order to transmit metering data to the wireless network. A secondary, rechargeable battery can be used to deliver the required power pulses but that creates a maintenance issue. Electrical double layer capacitors – known as either EDLCs or ‘supercapacitors’, charged from the primary battery, have proved to be a popular solution, but it’s one with some limitations. Many supercapacitors, particularly early generations, suffer from self-discharge, gradually losing charge over time. In some applications, the effect may not be of concern, but where it is, a new generation of lithium ion capacitors from Taiyo Yuden promises a solution (read our post on hybrid-capacitors here). Figure 5 (right) compares the self-discharge curve of a 40F lithium ion capacitor with that of a similar (50F) supercapacitor, or EDLC, and you can see that this particular limitation of supercapacitors is all but eliminated.

Summary

Building a cost-effective and reliable flow meter for liquids or gases involves a lot of component choices. Understanding the design trade-offs, ensuring that each component is the best for the application and, most importantly, that the individual parts work well together, is at the heart of creating a successful metering device.

If you have a question about flow metering applications or need some advice on component selection, our pan-European team of field applications engineers are on hand to help. Click the Ask an Expert button to the right of this post to get in touch.

SOME USEFUL RESOURCES

MAX35103: Reduced Power Time-to-Digital Converter

Batteries for Smart Gas Meters

Lithium Ion Capacitors: An Effective EDLC Replacement

Want more like this? Sign up for our newsletters

About Author

Alan Jermyn

Alan Jermyn has occupied senior management roles at the Abacus Group for a number of years. Previou...

View Bio