Industrial automation arguably dates back to Henry Ford's installation of a moving assembly line for the Model T in 1913; it is the use of various control systems to operate industrial equipment such as machinery, manufacturing processes and material handling equipment, with little or no human intervention.
Although human engineers and technicians are still ultimately responsible, in an automated factory the routine operation of a piece of equipment is under the command of a computer program running on an embedded microcontroller, industrial PC or programmable logic controller (PLC).
Figure 1: Industrial automation requires huge numbers of embedded microcontrollers (source: TI)
Automating industrial processes has a number of benefits: it saves energy and materials; it improves the quality, accuracy and precision of industrial processes; it allows operation in hazardous environments (in nuclear plants, for example); and it vastly saves on labor.
The results are impressive. ln 1909 it required 303 employee-hours to assemble one car; in 1929 the time had been reduced to 92 employee-hours; and in 2008 the Jeep plant in Toledo, OH needed just 13.6 employee-hours to assemble a vehicle that's orders of magnitude more complicated than the Model T.
Industrial robots now perform many tasks that were formerly done by humans, such as welding, pick-and-place and assembly, and machine vision systems have replaced quality control inspectors.
The Connected Factory and the Internet Of Things
The next stage after automating individual industrial processes is to make sure that they all work together smoothly – and provide data to their human masters, of course! The modern automated factory therefore relies on an industrial network using one of the numerous automation protocols such as Ethernet, Fieldbus, or HART Protocol to provide connectivity at the factory level.
Figure 2: A few of the many industrial communications standards (source: TI)
The Internet of Things (IoT) raises the bar to the next level by enabling integration of industrial machinery and systems with the Internet. The most recent version of the Internet communications protocol, IPv6, with its 128-bit addressing, means that potentially every single device can have its own IP address, heralding practically unlimited connectivity and massive data collection and analysis in the cloud.
In the smart factory of the future, the levels of the automation “pyramid” – from device to enterprise level – will be interlinked, and manufacturing-related data will become available in real time for making business decisions.
Galvanic Isolation and Industrial Automation
Adding electronic control and connecting multiple systems together via a network has many benefits, but there are issues and challenges, too. One of these is the challenge of combining high-voltage, high-current machines such as industrial robots and CNC machines with low-voltage, low-current data acquisition systems and networked communications. We're going to talk about an important technique used to battle this problem –galvanic isolation.
Galvanic isolation is the technique of isolating functional sections of electrical systems to prevent current flow between them; no direct (i.e., resistive) conduction path is permitted. Although there's no resistive path between sections, power or information is still transferred by capacitive, inductive, optical, or other techniques.
An isolation device passes a signal, either analog or digital, from input to output across an isolation barrier; to be effective, the isolation barrier must have high breakdown voltage and low leakage. Isolation devices are a key component in hooking up the connected factory, and are found everywhere in the signal chain: from individually- isolated sensor and data acquisition systems to the high-speed communications backbone that runs through the whole factory.
Why is galvanic isolation needed in industrial automation?
Safety - Protecting users of electrical equipment from potentially lethal voltages and currents is a key requirement in any electrical design. There are a number of regulatory standards, such as UL60950-1, that govern safety in electrical equipment. A safe design includes several levels of protection, including insulation, grounding, and isolation; in an offline power supply design, for example, a transformer provides inductive isolation between the AC input and the rest of the circuit. In general, galvanic isolation may be used wherever potentially dangerous voltages are present.
Ground Differences & Ground Loops- Unlike the simple schematics we drew in school, as practicing engineers we soon learn that ground is most certainly not the same at different points in a system, especially when those systems are widely separated - between different parts of an industrial plant, say. This can lead to errors or even failure in a digital network because any difference in the ground reference between transmitter and receiver reduces the margin to correctly identify a logic '0' or '1'. In analog circuits, a DC ground difference adds an offset error term and AC changes can affect the harmonic content of the signal. Galvanic isolation removes the effect of such ground differences, and breaks ground loops.
Common-Mode Voltages – in many cases we need to extract a small signal riding on top of a larger common-mode voltage: an in-phase signal or voltage that appears simultaneously on both input terminals. In some cases, this can offset the signal being measured by an amount that exceeds the full scale range of the instrument input, or exceed the overvoltage limits, causing damage. An isolated data-acquisition system can block the common-mode voltage and allow the signal of interest to be measured.
Regulatory Standards
A number of regulatory standards govern isolation for industrial applications, including IEC 60204; UL508; UL60947, and CSA 14-10. In addition, IEC 61010-1 and VDE 410/411 cover industrial control.
UL1577 strictly covers only optical isolators, but its definition and testing requirements for isolation barrier voltage often appears in specifications for other isolation devices.
Overview Of Isolation Technology And Devices
In general, an isolation device comprises a high voltage isolation component or barrier, a transmitter to couple a signal into one side of the isolation barrier, and a receiver to convert the signal on the other side of the barrier back into a digital or analog signal. Refer to TI’s ‘Digital Isolator Design Guide’ for more details.
There are three main techniques used to pass signals and power across an isolation barrier.
Capacitive Isolation: A capacitive isolator uses high-voltage capacitors, tpyically made of silicon dioxide (SiO2) to serve as the isolation component. Figure 3 shows the architecture of ISO7810 single channel digital isolator, which has an isolation barrier of 5.7kVrms and can handle data rates up to 100Mbps.
Figure 3: On-Off Keying architecture used in TI's ISO78xx 5.7kVrms reinforced digital isolator (source: TI)
The ISO7810 uses On-Off Keying (OOK) architecture: the incoming digital bit stream is modulated with an internal spread spectrum oscillator clock to generate OOK signaling, such that one of the input states is represented by transmission of a carrier frequency, and the other state by no transmission. Figure 4 shows a representative OOK signal.
Figure 4: Representative signal in OOK architecture (Source: TI)
This modulated signal is coupled to the isolation barrier and appears in an attenuated form on the receive side. The receive path consists of a pre-amplifier to gain up the incoming signal followed by an envelope detector that serves as a demodulator to regenerate the original digital pattern. The TX and RX signal conditioning circuits are used to improve the common mode rejection of the channel resulting in better Common Mode Transient Immunity (CMTI) and minimizing radiated emissions.
Since capacitors block DC, an analog signal must be converted to a form more suited for passing across the isolation barrier. The input of the ISO122 isolation amplifier, for example, consists of an opamp followed by a modulator to pass the signal across the capacitive barrier; the output side demodulates the digital signal to extract the original waveform.
Inductive Isolation: inductive coupling using a transformer is widely used to provide isolated power, since an isolated design requires separate power supplies on each side of the isolation barrier. The DCP01B family, for example, is a family of 1W, unregulated, isolated DC/DC converters with a 3kV isolation barrier. It contains an embedded transformer, provides up to 85% conversion efficiency, and is ideal for low power applications such as sensors and field transmitters. For designs requiring more power, TI offers a range of isolated power supply reference designs.
Optical Isolation: an opto-isolator, also called an opto-coupler, transfers electrical signals between two isolated circuits by using light. It commonly consists of an LED and a phototransistor in the same package.
Optically isolated couplers require high current pulses and suffer from the effects of LED aging. In addition, they can be too slow for high-speed digital communications, and are therefore being replaced by capacitive designs.
Isolation is often incorporated into widely-used industrial automation functions such as transceivers, amplifiers and ADCs. TI's isolation product portfolio includes:
- RS-485 transceivers
- CAN transceivers
- I2C transceivers
- Gate drivers
- Analog-to-digital converters
- DC/DC modules
- Amplifiers
Conclusion
There's no doubt that we'll be hearing a lot more about Industrial Automation and the Internet Of Things in future.
Over the next decade, industrial IoT is expected to revolutionize the industrial sector: it's already begun to transform major industries such as manufacturing and transportation. As IoT integration spreads to different industries, the industrial IoT market has immense growth potential – one forecast has it being worth almost $320 billion by 2020.
Whatever the final number turns out to be, isolation technologies will be playing a key role.