02 January 2013
Most modern industrial plant and machinery has its own control system, and this is usually connected into a site-wide control network to provide automation of the entire production process. Increasingly, the control network also interfaces with management IT systems to provide real-time or near real-time monitoring, reporting and control of every aspect of a company’s operations.While this ‘big picture’ seems clear, the details can be complicated. In essence there are three basic issues:* Getting data from where it is generated across the network to other locations where it is interpreted and used (note the plural ‘locations’ – a signal to say a storage tank is full might also be used to shut the inlet valve, update the material stores information, initiate the next stage in the production process and feed into the commercial office systems).* Getting the signal to the user locations in an appropriate timeframe (the inlet valve needs to close immediately, but updating stores is not so time critical).* Making sure the recipient can read the data – it may need translating into a different format or protocol before delivery. We will start by looking at the architecture, or hierarchy, of a typical control system. The first link in the chain is the field devices – sensors, actuators, motors, lights, switches, valves, contactors and measuring instruments. The operation of these is monitored by a nearby controller, often a PLC, with bi-directional data signals passing between device and PLC as necessary.Typically, there will be many PLCs dotted around a plant, each responsible for one machine or function. The PLCs also communicate to higher level controllers and ultimately to a central controller that collates all the information coming in from around the production plant. The controllers will often process raw data into sophisticated information. For example, the count of a switch may correspond to the stroke of a pump, from which flow rates and volumes can be calculated to give material usage and production output.The top level controller also communicates with the business systems’ IT network; the latter also collects information from the stores, various offices and dispatch. In a newer development, field staff may also be viewing or uploading electronic information from a tablet or smart phone.The communications network on the shop floor is often referred to as a fieldbus and offers real-time, distributed control. In the past, before such networks were widely used, devices were connected directly to their controllers, meaning large amounts of costly wiring. The chief benefit of fieldbuses is that a single cable can link all the devices in a daisy chain fashion, significantly reducing costs and increasing reliability. A single cable is easier to install and maintain and the system is easier to re-wire following changes, adaptations and reconfigurations that will inevitably happen.CompetitionThe concept of fieldbuses and supporting technologies began to emerge in the mid-to-late 1980s. As with all emerging technologies, competing camps soon sprang up, each developing slightly different solutions to fundamentally the same problems. The issue was that the different solutions were not compatible with one another and a lot of effort was wasted in the ‘fieldbus wars’ as the organisations involved fought for dominance. This was eventually resolved with the publication of the IEC 61158 standard.The standard has not achieved the ultimate goal of producing one single standard but, along with some natural market and technology forces, it has considerably reduced the number of options. Nowadays most major device manufacturers make products that are compatible with three or four of the most popular fieldbuses.Even today the various fieldbuses are not easily connected to each other. The design of their cabling, plugs and sockets varies, as do their bit timing, synchronisation, encoding/decoding, band rate and bus length. However, in recent years most fieldbus developers have introduced an Industrial Ethernet version of their offering.Ethernet is the IT version of fieldbus, and has succeeded where fieldbus failed. It is a protocol common to all computers, telephones and other business equipment that allows them to all be plugged into the same network and to communicate freely, even though they are made by different companies from different parts of the world. The obvious question to ask at this point is: why not dump all the industrial fieldbuses and swap to Ethernet? Until recently, the answer was that Ethernet fell short in two regards:* It was not industrially hardened. Early adopters found that commercial grade hardware could not survive the rigours of the shop floor and soon failed.* It was not deterministic. While this is not a problem for the delivery of emails, synchronising the operation of high-speed machinery makes this absolutely essential.These shortcomings led to the development of Industrial Ethernet which offers devices designed to cope with industrial environments and operate in real-time to overcome the determinism issues. Easy connectionsAn important advantage Industrial Ethernet offers is theoretically easier connectivity to company IT systems. In theory, this allows a company’s plant control system to feed data and information straight into other information systems, such as order processing, materials ordering, logistics and planning. Therefore a company can have complete real-time, up-to-the-minute information on all its operations, which can lead to efficiency gains, interdepartmental optimisation and far better forecasting and forward planning. Other advantages of Industrial Ethernet include reduced hardware cost (due to greater manufacturing volumes), faster signal transmission speeds and the use of standardised cables and connectors. Naturally, there can also be some disadvantages, such as the possible complexity of maintaining a combined IT/Industrial network. Some companies have struggled with a lack of familiarity with Ethernet among shop floor electricians, while IT technicians have shied away from the ‘dangerous’ world of the shop floor.Despite this, the general industry consensus is that the advantages of Industrial Ethernet far outweigh the disadvantages and it is rapidly becoming the technology of choice in many industrial applications.However, with Industrial Ethernet now being offered by most of the main fieldbus suppliers, the question of which protocol version to opt for still remains! Obviously sites that are already committed to, say, CC-Link are likely to be attracted to CC-Link IE, but if the installation is at a greenfield site the choice may not be straightforward and expert guidance may be needed.Some protocols began with specific application types in mind, such as motion control. Others have evolved over the past few years, adding more features such as hard real-time control. One thing is clear – the market today is crowded with many different types of Industrial Ethernet protocols, leading to almost as much confusion as in the fieldbus days. Added to this, there is still a misconception that ‘Ethernet’ is ‘Ethernet’, i.e. all protocols can talk to each other because they are all based on an Ethernet physical layer.With the fieldbus wars now well behind us, we can look back and see how unproductive they were. There is a danger that with Industrial Ethernet increasingly being accepted as the best technology we will descend into a new protocol war. To avoid this, users should make sure they are in a position to make informed decisions about what will work best for them. The various open network organisations are a good place to begin collecting this kind of information.
Print this page | E-mail this page
This isn't a paywall. It's a Freewall. We don't want to get in the way of what you came here for, so this will only take a few seconds.
Register Now