“Ethernet-APL is the latest step in Ethernet’s transition down the ISA-95 enterprise-control integration stack from Level 3’s manufacturing operations management to Level 0’s production processes," says Andrew Kravitz, product management director instrument connectivity at Emerson. "Many users have been trying to work with Foundation Fieldbus and other protocols, but they’re often too complex. This is where Ethernet-APL can help because it’s simpler to implement, much like familiar 4-20 mA, and can provide power and intrinsic safety along with communications using two-wire, twisted, shielded-pair cable.”
However, even though everyone uses Ethernet in their daily lives, Kravitz reports it poses some added challenges in industrial settings. “It’s easy enough to set up Ethernet and use web browsers to manage individual components and gain access to their data and diagnostics. However, this process doesn’t scale well for configuring and maintaining hundreds or thousands of devices in industrial settings,” explains Kravitz. “We already use Emerson’s AMS Device Manager software for bulk configurations and diagnostics, and we believe that Ethernet-APL users will leverage the same tools for automating maintenance work practices faster and without having to do as much retraining.”
Kravitz reports that traditional 4-20 mA networking has relatively higher latencies because it must go through the usual I/O infrastructure, while Ethernet-APL is faster because it can talk directly to sensors, instruments and other device-level items. These reduced touchpoints and simpler network path also let Ethernet-APL capitalize on its 10 Mbps, compared to fieldbuses like HART that runs far slower at 1.2 kbps. For instance, processes that can benefit from quicker data delivery include radar gauges that take 15 minutes to provide an echo curve with HART can relay one in less than 5 seconds with Ethernet-APL.
Decorate the Ethernet backbone
To transition to Ethernet-APL once switches and other components that support it are available, Kravitz recommends that users develop a thorough plan that lets Ethernet-APL convey both the traditional process signals they’re used to originating via 4-20 mA, as well as the rich datasets from devices that have been traditionally underutilized in smart process devices. However, if a process application or facility doesn’t already have a regular Ethernet network, it will be necessary to install one before Ethernet-APL devices are deployed in brownfield applications.
“The key is to have one solution that supports both Ethernet-APL and traditional protocols and signals simultaneously,” adds Kravitz. “Pepperl+Fuchs, Phoenix Contact, R. Stahl and Softing are testing Ethernet-APL switches, and we’re updating our CHARMS modules with a distributed carrier backplane that allows an Ethernet-APL I/O channel alongside the other traditional I/O signals it supports. This Ethernet-APL capable CHARMs solution will be available in the upcoming DeltaV version 16 release timeframe. Today, we have a DeltaV PK controller than can communicate with Ethernet-APL devices through available Ethernet-APL switches. Beyond this, we’re allowing devices to communicate with more than the DCS they usually talk to. This will let intelligent field devices more effectively communicate their own health and maintenance information. Ethernet-APL is better at pulling in this kind of intelligence, which helps users know when to act before faults happen.”
Despite its speed, Kravitz adds that one drawback to Ethernet-APL is that it hasn’t integrated enough cybersecurity yet because it’s not using secured protocols. “Ethernet-APL is flexible, but it also must make its connections secure because it’s going to be talking to all kinds of devices and systems,” says Kravitz. “However, HART-IP is the only protocol available today that’s natively secure. Profinet and EtherNet/IP aren’t encrypted in field devices today, so their communications aren’t as secure. The good news is Ethernet-APL’s four standards development organizations (SDO) are working to add cybersecurity to their respective protocols. For example, while OPC UA is encrypted, it is only utilized at the controller-to-cloud level today, and isn’t used by the sensors and other field devices at Level 0 yet. Ethernet-APL lets controls live alongside monitoring and optimization data, but their connections and communications have to be secure, otherwise those field devices should be isolated, which will cause Ethernet-APL to lose its flexibility and ability to interact with multiple systems.”
Despite its speed, Kravitz adds that one drawback to Ethernet-APL is that it hasn’t integrated enough cybersecurity yet because it’s not using secured protocols. “Ethernet-APL is flexible, but it also must make its connections secure because it’s going to be talking to all kinds of devices and systems,” says Kravitz. “However, HART-IP is the only protocol available today that’s natively secure. Profinet and EtherNet/IP aren’t encrypted in field devices today, so their communications aren’t as secure. The good news is Ethernet-APL’s four standards development organizations (SDO) are working to add cybersecurity to their respective protocols. For example, while OPC UA is encrypted, it is only utilized at the controller-to-cloud level today, and isn’t used by the sensors and other field devices at Level 0 yet. Ethernet-APL lets controls live alongside monitoring and optimization data, but their connections and communications have to be secure, otherwise those field devices should be isolated, which will cause Ethernet-APL to lose its flexibility and ability to interact with multiple systems.”
