A: There are a multitude of benefits, starting with scalability. Remote I/O allows for easy scalability and expansion of an automation system. It’s much easier to add or remove field devices without major modifications to the existing infrastructure. If you have to add a field device in a remote I/O setting, it's easy to add another I/O card to the segment that's in the field versus having to pull a wire all the way back to the control room.
Speaking of wiring, that’s one of the biggest benefits. It significantly reduces the amount of wiring, including the cost of installation. By placing I/O devices closer to the field or the process, remote I/O eliminates the need for extensive wiring between the process and the control room.
I think another big benefit is enhanced flexibility. Remote I/O allows for greater flexibility in system design. The layout with those devices can be placed remotely. The system’s components can be strategically positioned based on factors, such as environmental conditions or physical space constraints.
Also, with process plants, we see a lot of hazardous areas, where it's simply not possible to put a central process unit controller or DCS in a hazardous environment. You just can't protect those from explosions. But it's quite easy to put a remote I/O node in a hazardous area. We not only have couplers, but also I/O devices that are already approved for Zone 2. Of course, we can also work with devices in Zone 1 or Division 1 with intrinsic safety (IS). So, there are many different advantages to remote I/O.
Q: How common is remote I/O at this point? Is it gaining momentum in the process industries?
A: Generally, I think it’s become extremely common in the process industries, especially when you're dealing with such vast areas that are covered by automation systems, and there hasn't really been a great fieldbus standard in the process industries that's been widely adapted. There were some fits and starts, and Foundation Fieldbus was pretty big for a while, but there were some issues there, and maybe Ethernet Advanced Physical Layer (APL) will be that standard going forward. We'll see.
What I see in my work is that most greenfield installations and process plants are utilizing some sort of remote I/O infrastructure. Many customers we work with—the engineering firms and system integrators—are working on projects to migrate from centralized I/O to gain all the advantages we just discussed.
Q: How does diagnostics work in a remote I/O system?
A: Since you can transmit so much data via industrial Ethernet networks, there's just a lot of data there. You compare that to point-to-point 4-20 mA, even 4-20 mA with HART, and they’re somewhat limited in how much data you can transmit. We use EtherCAT, and in our ecosystem, it comes with a multitude of diagnostic capabilities. Information about communication errors can be looked at, such as lost frames, excessive jitter, issues with data synchronization and network health diagnostics. We can exmaine information about the overall health of the EtherCAT network, such as the number of devices connected, topology, bandwidth utilization and data integrity.
EtherCAT incorporates cyclical redundancy checks (CRC) to ensure data related to CRCs can help identify issues with data transmission and reception. And, because EtherCAT utilizes bandwidth so well, we can track six types of CRC errors. In the instance of a transient fault, shielding is grounding out in the cable, and EtherCAT can tell you what the failure mode is, which cable it is, and you know which slaves are affected. This makes troubleshooting and maintenance much easier.
We can also analyze network load. Diagnostics can provide insights into load distribution, and identify bottlenecks in the network, as well as potential areas for optimization. Another big benefit is being able to transmit all this data for event logging. EtherCAT devices can log events, such as errors, configuration changes and device-specific events. This helps with troubleshooting and historical analysis to determine what caused certain events.
Q: Obviously, when we go from something centralized to remote, the risk becomes something people start thinking about. What are the risks for remote I/O and how are they mitigated?
A: I’d say the biggest one would be when you have a lot of devices and you’re consolidating. Maybe you have one Cat 5e cable going back, and if a single cable is damaged or cut, you’re going to lose a host of devices.
There are ways to mitigate it. We recommend media or fieldbus redundancy. EtherCAT can natively support a redundant ring network, so if one of the cables gets damaged, it's not going to bring down the entire network. Not only does it keep the network up, but also, going back to scalability, a big advantage of remote I/O is having this redundant EtherCAT network.
If you have to add or remove a device, you can do it while the process is still running. You won’t have to shut down the process to add devices because communications aren't broken. This provides some additional resilience. Another benefit of EtherCAT is its redundancy concept isn’t bound by the same physical media. So, your primary route may be copper, while you have a secondary redundant ring using fiber-optics, which again, just gives you some added resilience.
The other big risk when installing or deploying a remote I/O system has to do with cybersecurity. With industrial Ethernet-type protocols, when you're dealing with Internet protocol (IP) addresses and switch-based networks, those devices are directly visible and accessible to the overall plant network and most likely beyond it to the corporate network.
There are different threat vectors now that can be introduced, not just in the IT network, but also the OT network. We're dealing with industrial Ethernet protocols, and that’s where you'll see EtherCAT has a major advantage over others. That's because EtherCAT is based on Ethernet, but it's not IP-based. Most cybersecurity attacks latch onto an IP address, which is how they propagate through the network. EtherCAT simply doesn't recognize them, and rejects them. This is a big way to mitigate cybersecurity risk when you're moving from 4-20 mA point-to-point to more networked environments.