What is Static IP Addressing? - From AutomationDirect

AutomationDirect presents: What Is? Networking Addition. What is Static IP Addressing? You may be setting up Ethernet communications between one PLC and an HMI, or you may be setting up a wide controls network with dozens of components on several subnets, all linked through routers and connected to the internet. In either case, every component involved is going to need an IP address in order to communicate, and for the components we want to talk to each other, their IP addresses need to be either directly compatible – part of the same subnet – or facilitated through a routing gateway. AutomationDirect offers a couple of really informative videos on the fundamentals of IP addresses and subnet masks if you need a quick review, and you can find links to those videos in the description below. Understanding all of this, the next question is: how do these components get their IP addresses? A device like a PLC can be manually assigned a static IP address or it can be set up to be addressed dynamically. This video concentrates on manually configuring a stand-alone controls network with static IP addresses, but first a few words about dynamic addressing. Dynamic Host Configuration Protocol is the network management system through which a server or router assigns and leases network parameters including IP addresses to client devices, or hosts, that request them. As an example, let’s look at a PLC being added to a communication network based around a DHCP server. The PLC, having been configured for dynamic IP addressing and finding itself with no assigned address, broadcasts a request for one. The server selects the next address it has available, from a defined range of addresses it has been allotted by a network administrator. It sends a reply offering the address to the PLC. The host – our PLC – then sends a message accepting the offer, requesting the address. The DHCP server then assigns the address - including default gateway and DNS addresses as set up by the network administrator - to the PLC, tying it to the PLCs unique Mac address. This assignment is called a lease, and as that implies, expires after a set period of time and has to be re-requested. This helps maintain efficiency in address assignment and keeps addresses from being wasted. That’s a quick and dirty run through of DHCP, and of course, there are myriad ins and outs to the protocol, not to mention to Default Gateways and DNS addresses that are just beyond the scope of this video – but there is one more thing worth a mention: If a Windows-based host broadcasts a DHCP Discover looking for a server or router and does not find one, it will assign itself its own IP address in this range, as well as a 16-bit subnet mask. This is known as Automatic Private IP Addressing (APIPA) and ensures that even without a DHCP-assigned address, the host will be able to communicate. Keep this 169.254 range in mind – we’ll be coming back to that in a couple of minutes. Ok, enough with DHCP. We’re here to talk about manually addressing a smaller industrial controls network with static addresses. First off, with all of the cool things that dynamic addressing brings to the table, why would you choose to use static IP addresses? There are a number of reasons why static addressing could be preferable, particularly for an industrial controls network. A simple reason might be that DHCP requires connection to a DHCP server or router. If you’re setting up a local controls network with a few PLCs and HMIs, there’s likely to be no need for such a router. Even beyond that, in a lot of circumstances it is considered best practice to use static IP addresses. In a controls network, reliability of communication is essential. Everyone from programmers and integrators to machine operators and maintenance techs need to know that they can quickly communicate with any component in the network. The PLCs need to be able to communicate with the HMIs and the supervisory systems as well as any networked remote IO, sensors, or other components. Replacing components or even network hardware must result in minimal downtime and should not require reconfiguration of a server or router, which often requires someone like a network administrator. Quite often, static is the way to go. In truth, if you’re setting up a private self-contained network, you can use any addressing scheme you want, just as long as all of the components that should be in communication are in the same subnet – which is to say that the IP addresses and subnet masks are compatible as laid out in those other videos mentioned earlier. However, it’s strongly recommended that even for a stand-alone network, you stick with subnetting standards. For private networks, certain address ranges have been defined by the Internet Engineering Task Force, depending on the number of devices being connected. A Class A private network uses an 8-bit network ID (subnet mask 255.0.0.0), uses addresses in the range 10.x.x.x and will facilitate up to 16.7 million devices. Frankly, it would be insane not to set up such a network with a DHCP server. Class B uses a 16-bit network ID (255.255.0.0), uses range 172.16.x.x to 172.31.x.x and allows up to 65,534 devices. Again, probably DHCP territory. Most small networks suitable for manual addressing fall into the definition of Class C - up to 254 devices with a 24-bit network ID – and should use the address range 192.168.x.x. Recall that your subnet mask is going to define the range of addresses available for assignment and the highest and lowest addresses in that range are reserved and cannot be assigned to a host. For example, with a 255.255.255.0 subnet mask, 192.168.x.0 will be the network address and 192.168.x.255 will be the broadcast address and neither can be assigned. Back to APIPA – the 169.254 address a DHCP-enabled device will assign itself in the absence of a server will not be allowed to pass through a router and should never be used as a static IP address. Again, if you are setting up a stand-alone network, not connected to any larger system, you technically could use this range but doing so would be… let’s say short-sighted. We strongly recommend you avoid it. With these caveats in mind, you should be ready to address manually the components in your private controls network. For more resources on network communication and all other aspects of automation and industrial control, visit www.automationdirect.com Click here for AutomationDirect communication products. Click here for more videos on networking. Click here to subscribe to our YouTube channel.