Selecting an automation controller
Consider these criteria when selecting an automation controller for motion, robotics, machines, and other applications.
An automation controller does more than replace relays as the programmable logic controller (PLC) was originally envisioned. Now a controller can be capable of integrating logic, motion, robotics, and communicating with other machines and management systems. Performance can range from simple devices to multicore processors.
The distinctions between a traditional PLC, programmable automation controller (PAC) and industrial personal computer (IPC) are largely irrelevant-the processing power is still available. Due to the acceptance of the International Electrotechnical Commission (IEC) 61131-3 programming, control software has gained a degree of standardization. Robust real-time operating systems running in the background obviate the need for Microsoft Windows-based operating systems, so the term “IPC-based control” would more accurately be termed “Intel-based” or “AMD-based” to reflect using powerful mainstream processors.
Since today’s automation controllers do more than logic, “PLC” is probably an obsolete term. Because all automation controllers are programmable, the “P” in PAC also seems to be redundant. Controllers are fundamentally computers, and may run multiple operating systems (real-time, Microsoft Windows, and Linux) on the same processor. An IPC may be used for control, data acquisition, and emerging tasks such as edge computing.
It’s become popular to coordinate all of the machine’s functionalities in the same software environment, in the same program, running on the same processor. This allows the machine functions to be synchronized and the modular code structure makes for an organized, cohesive approach. Still, there is a place for less integrated control such as straightforward applications that are not intended to scale up.
The application’s complexity and performance requirements dictate the controller’s specifications. The following are criteria from platforms that may or may not need to be checked off depending on the application.
There is a fundamental need for logic control, which is why we continue to call automation controllers “PLCs.” PLCopen is an organization that maintains and expands the scope of the IEC 61131-3 programming standard and manages a large base of knowledge, training, and libraries. The group’s activities extend far beyond logic to include motion, safety, OPC Unified Architecture (UA), XML, and more.
Depending on the required complexity and synchronization of the motions, dozens or even hundreds of axes of motion can be controlled by an automation controller. A separate motion or robot controller with a dedicated motion network is no longer required due to Moore’s Law and industry standards.
Hardwired network safety is still preferred in North America. Network safety running on the same network controlling the machine has become a proven and useful control functionality. Network safety can be implemented from a redundant core on the control processor, to a separate safety controller, and then to a safe input/output (I/O) in small systems. Network safety also extends to motion safety and robotics functionalities that allow machines to run in safe mode rather than shutting down-providing exceptional operational efficiencies.
It is possible for the same automation controller to integrate multiple delta, selective compliance assembly robot arms, articulated and/or gantry robots along with all the other machine functions. In addition, it’s also possible to perform the kinematics in an IEC 61131-3-compliant environment. Dedicated robot controllers continue to serve valuable functions because of what is built into the robot’s system-ranging from palletization algorithms to assembly patterns.
Monitoring a machine’s health is key to a predictive maintenance plan and reducing unplanned downtime. A controller can be combined with various off-the-shelf transducers such as temperature probes and accelerometers to monitor actual conditions. Machine monitoring can also help detect anomalies before catastrophic failures occur. Energy monitoring can also be applied to compressed air usage, fluctuations, natural gas use in heaters and dryers, and water usage for process and cleaning.
An automation controller can be a web and OPC UA server and client. They have functions to collect Industrial Internet of Things (IIoT) data, and to receive instructions back from the cloud or edge to optimize the process. Automation controllers typically send data to the manufacturing execution system (MES), enterprise resource planning (ERP), overall equipment effectiveness (OEE), trusted platform module (TPM), and product lifecycle management (PLM) software. In an IIoT environment, it’s equally important also to receive useful analytics.
Before, when a new component-such as a drive-was replaced, it was necessary to manually determine and load the correct firmware version for the device. Today, automation controllers may automatically read the device and alert a technician to make the necessary adjustments without intervention.
Today, even low-cost controllers have one or more Ethernet communications ports to communicate with human-machine interface (HMI) panels, management systems, programming, and other non-time-critical tasks. It is also common for controllers to support a type of industrial Ethernet for a deterministic machine network such as EtherNet/IP, EtherCAT, Powerlink, Profinet, and others. Unfortunately, there is still not one, universally recognized industrial Ethernet standard to provide the high-speed, deterministic communications suitable for machine control.
There is great anticipation, however, in the development of Time-Sensitive Networking (TSN), which along with OPC UA and OPC UA publish-subscribe (Pub-Sub), will bring a certain level of determinism to the Institute of Electrical and Electronics Engineering (IEEE) 802 family of Ethernet standards. A testbed has been established in the Industrial Internet Consortium, with multiple industrial automation suppliers participating in “plugfests” demonstrating the viability of TSN in machine-to-machine communications.
TSN is important because for IIoT to work, there will need to be communications interoperability among different control platforms in use at a facility or enterprise-wide and the cloud. If serial interfaces are required, they should be specified due to the decrease in popularity.
The following are the three most common automation controller form factors.
- IP20, cabinet mounted: This is the traditional PLC form factor that has a separate HMI, and typically uses integral and/or backplane/rail mounted I/O, and/or remotely-mounted I/O modules.
- IP65/67/69K sealed, pedestal or panel-front mounted: This format integrates the HMI and controller and is increasingly popular with swing-arm mounting for its ergonomic advantages.
This format can also incorporate PC functions to run various Windows applications in addition to control, such as HMI, although there is an increasing trend for a web-based HMI.
There are sealing requirements that correspond to environmental conditions and cleaning practices. For example, stainless steel is often used, especially in hygienic environments.
Pedestal mounting tends to be more expensive than panel mounting, stainless steel bezels, and higher levels of sealing protection for comparable controllers.
Some prefer separate panel-mounted PLCs and HMIs to avoid having to replace both components if one or the other is damaged. However, this is no longer a concern, as integrated units are available in which the controller is mounted to, but removable, from the HMI. This makes it easier to switch to a larger screen, or swap in more powerful control hardware without changing the screen.
- IP20, cabinet-mounted industrial PC, with separate HMI: Just like the integrated form factor, this form is capable of serving as a controller with real-time operating systems, various computer operating systems, and web services. The controller may be separate and the industrial computer is dedicated to non-control tasks such as edge, fog, or cloud computing. Historian, serialization, and vision inspection are also common applications.
While software development environments are often tied to hardware—nano, micro, mid-range, and large PLCs—it is also possible to work in development environments that are independent from the hardware. This means that a project is coded, and then the control hardware can be selected or changed. This flexibility extends to motor and drive types. A low-end stepper or variable frequency drive (VFD)-based machine can share the same program as a high-end servo machine. The need for scalability is most valuable when a family of machines is being designed and allows key software elements to be reused.
There are a lot of choices from low-end to multi-core processors, often with overlapping performance characteristics.
Therefore, it is recommended to work with the technology provider’s technical support and sales engineering team to select the optimum price/performance for the anticipated application’s requirements because of their product knowledge.
Ideally, the processors should be scalable so the control software is compatible across the controller product line.
Automation technology providers invest in significant stockpiles of critical components to ensure product availability, and migration for drop-in replacements.
Also, determine whether fanless operation will be required, and the expected ambient temperatures where the controller will be mounted. Other heat dissipation options include fans, air conditioning, heat sinks, and water cooling.
Solid-state memory has become very popular in automation controllers, removable media like C-Fast cards, and permanently installed memory devices in the more cost-sensitive applications. The advantages to removable memory are that it can easily be replaced, it is easy to make and store backups, and it is easy to expand memory capacity.
However, be careful using industrial memory cards and ensure that the media meets the necessary specifications for the application. Different memory types have different service lives, which are dependent on read and write cycles. This is also a topic to discuss with an automation supplier.
John Kowal is the director of business development at B&R Industrial Automation Corp. Edited by Emily Guenther, associate content manager, Control Engineering, CFE Media, firstname.lastname@example.org.
KEYWORD: programmable logic controller (PLC)
The three most common form factors for automation controllers
Possible application requirements for automation controllers.
What boxes need to be checked off to meet your application’s automation controller requirements?
Online: See the PLC Digital Report at www.controleng.com/DigitalReports.