Ethernet is the one of the most widely known ‘invisible’ technologies in use today. It can be seen in everyday use at home, in offices and now in industrial applications.
Essentially, Ethernet is a tool used to transmit data from one machine to another. There are however big differences in the industrial environment that make industrial Ethernet different to what is used elsewhere. Primarily, this means the industrial variety needs to be able to cope with different types of situations. This could mean operating in harsh environments, or in places with heavy vibration, or in areas with increased noise like factories.
Industrial Ethernet is a form of two-way communication. The best way to describe how this works is to look at this within an application. We’ll use a process plant as an example. Within a typical plant there will be many different processes that need to be controlled in the correct order and sequence.
Industrial Ethernet works using a method known as handshaking. This is when an operation has been completed and the controlling device (this could be an IPC, PLC or motion controller) sends a signal to the machine to stop the process.
But, even in high-end systems, this signal can get ‘lost’ for many different reasons. When handshaking is used the controlling device will know this loss has happened as it has not had a reply back from the machine to say it has stopped. The control system will then resend the message.
This is very different to office-based Ethernet where this level of feedback is not required. If a web page does not open, for example, the user will simply refresh the page. With industrial operations though, system crashes or slowdowns could end up with the machine becoming blocked or damaged, resulting in maintenance or time penalties.
The other type of two way communication incorporates collision detection. This is when messages collide within the network causing information to get lost. Industrial Ethernet stops message loss and machines from ‘going down’ by using delivery notes to tell the controller that the messages have been received. This is not needed within commercial Ethernet as the tasks are not as controlled.
There are some other main differences between standard Ethernet and industrial Ethernet, namely topology. This is the way that a network is arranged, and includes the nodes and connection lines within the network. In an industrial Ethernet system there is the ability to use a variety of connection types. This can include bus, tree, star, ring and hub & spoke topologies, allowing the system to be configured to match the process it will carry out.
In the white paper ‘Industrial Ethernet : A Control Engineer’s Guide’, Cisco discusses ring and hub-and-spoke topologies. In hub-and-spoke designs, for example, three layers of switches are usually installed. The first layer is often referred to as the access layer. These switches provide connections for end-point devices like PLCs, robots and HMIs.
Layer 2, called the distribution layer, provides connectivity between the access-layer switches. Finally, a third layer – called the core layer – provides connectivity to other networks or to the internet service provider (ISP) via routers.
The distribution layer may include switches with routing functions to provide inter-VLAN routing. Access-layer switches, on the other hand, generally provide only Layer 2 (data link) forwarding services. For performance, network equipment at each of these layers must be aware of the information contained within the Layer 2 through Layer 4 packet headers.
When it comes to network security, users need to be open to the fact that every network connected to the outside world is at risk of being compromised. Attacks can happen in many types; from well-known viruses or malware. The worst case is being attacked directly with the intention of damaging or stealing information about your company or machine. This means that when you have networks in place, security is key.
There are various ways that security can be put in place to make sure have the protection you need. Usually these are called firewalls.
With the increasing use of industrial Ethernet more work and time is being put into making sure that a network’s security is increased. Most products on the market now come with their own firewall and security systems to make sure potential damage will not reach your network. It can also help make sure there is no downtime on the machine due to breach of security.
Industrial Ethernet usage is growing consistently. Coupled with the advances in products and networks, system integrators can rely on Ethernet as their main communication protocol, allowing machine builds and operators to remotely control and monitor the machines from anywhere in the world.
What are your experiences of industrial Ethernet? Let us know in the comments below.