PLCs vs. PCs for industrial control

Since the dawn of the programmable logic controller (PLC), plants have come to expect reliability, stability, and simplicity from them. However, software controls running on personal computers (PCs) have made inroads to industrial automation and process control in recent years.

By Jack Smith, Senior Editor, Plant Engineering Magazine June 12, 2003

Key Concepts

The PLC has become greater than the sum of its card slots.

The line between PLCs and industrial computers running control software is becoming blurred.

PACs combine the best parts of the PLC with the best parts of the PC.

Sections: PLC legacy When PC control makes sense Programmable automation controllers — the next generation? Peaceful coexistence

Since the dawn of the programmable logic controller (PLC), plants have come to expect reliability, stability, and simplicity from them. However, software controls running on personal computers (PCs) have made inroads to industrial automation and process control in recent years.

So, which is best — computer controls, or PLCs — when it comes to controlling industrial equipment and processes? This squabble has been brewing for several years with no clear-cut answer. Perhaps the answers lie in the evolution and combination of technologies.

PLC legacy

The IEEE defines a programmable controller, or PLC, as a “solid-state control system with programming capability that performs functions similar to a relay logic system.” Relay logic — or ladder logic, as it is frequently called — got its name from many years of using relays to control industrial equipment. Other than entirely hard-wired systems, this is about as simple as it gets.

Relay logic was used to design motor circuits as well as entire relay control panels. Relay logic is called ladder logic because the program looks like a ladder when it is drawn pictorially — unlike a high-level programming language, such as C++ or BASIC, which uses alphanumeric characters.

PLC technology is entrenched in more than 30 yr of control fundamentals. Characteristics of this technology include:

Realtime determinism for process execution

Realtime determinism for reading and writing input/output (I/O) values

Ensured robustness and reliability in execution and continued operation

Protection of the control code and execution activities from outside influences

Functional isolation to ensure a dedicated control response.

The cost of processing power is decreasing, communications capabilities are improving, and open systems are becoming more accepted. Nano PLCs (Fig. 1) — those with up to 32 I/O points — are gaining networking capability, while micro PLCs — those with up to 129 I/O points — are becoming more sophisticated functionally. Because of this, the tendency to replace larger PLCs by networked configurations of smaller PLCs will gain ground, according to ARC Advisory Group. Reasons for this gain include reliability, simplified wiring, and scalability. Also, advances in software allow users to easily configure and use special functions that make PLCs do much more than logic control. Also according to ARC, there is a growing trend toward sophisticated motion and position control software as well as web-enabled functionality, such as web-based HMI, online upgrades, and remote troubleshooting.

ARC also reports that the PLC market will revive, despite the global economic slowdown. Users need automation to improve productivity, reduce costs, and increase flexibility to meet varying manufacturing demands. Signs of revival are also present in the high pace of technological change as fears of obsolescence drive upgrade plans. In addition, the potential of reduced total cost of ownership with open connectivity and programming platforms stimulates demand for higher levels of automation.

Manufacturers shrank the physical size of programmable controllers, reduced wiring requirements, distributed control processing more effectively, and added more extensive communications capabilities. With the addition of high-speed communications that do not interfere with PLC scan time, it is possible to off-load heavy calculations to a PC and use the results in the PLC process.

The PLC has evolved. It has survived the introduction of the PC into the control industry because it was able to incorporate quickly the technologies that allowed the PC to obtain its original control niche. These technologies include massive network connectivity, processing speed, local storage, and programming flexibility. Another reason for PLC survival is that the various control modules integrate without compatibility issues, reducing startup and system validation costs, while providing a tested and highly stable control environment out of the box.

When PC control makes sense

PC-based control provides the basic foundation of the existing control strategy without compromising functionality or precision. Also, there are the added benefits of flexibility, enhanced programming features, increased connectivity, and additional time and money savings when using PC-based control to create a fully integrated application environment.

Reasons to implement PC-based control include:

Networking the control system to higher-level applications such as ERP or EAM

Handling complex mathematical applications in recipe management or vision inspection systems

Connecting to bar code scanners, in-motion weigh scales, and other devices.

PC-based controls allow users to control multiple functions on a single platform, instead of with multiple PLCs. Also, they can be programmed in common languages, like C++ and Visual BASIC, run using Windows, and integrate with plant floor and enterprise-level ERP systems with standard interfaces.

The benchmark of PLC and PC-based control systems is determinism. The speed at which an application runs is often confused with the actual deterministic nature of the process. The true measurement of reliability and control accuracy is the determinism, or repeatable and timely nature of the code execution.

What is determinism? According to the IEEE, a process, model, or variable is said to be deterministic if its outcome, result, or value does not depend on chance. A deterministic model is one in which the results are determined through known relationships among the states and events, and in which a given input will always produce the same output.

Programmable automation controllers — the next generation?

PLCs have progressed from being just ladder-programmed controllers for discrete I/O. They incorporate ladder, process, motion, temperature, etc., within the processor. The PLC has become greater than the sum of its card slots.

For a while, software controls appeared to be gaining ground on PLC territory. The line between PLCs and industrial computers running control software is becoming increasingly blurred. The “next generation” PLC features capabilities that were heretofore found only on PCs. The new breed of PLC marries the reliability of a PLC with the open programming languages used by PCs. This blending of PC and PLC technology uses commercial computer chips and the Microsoft Windows operating system. With the installed PLC base, these advantages afford users with more sophisticated control without ripping out legacy systems to install PCs running software controls.

So what do we call the next generation PLC? As if we didn’t have enough acronyms, the ARC Advisory Group has coined the phrase “programmable automation controller” or PAC. The PAC concept identifies the convergence of control engines that address the multidiscipline needs of PLC users and the information requirements of manufacturing enterprises. PACs include the PLC mainstay functions and expanded control capabilities, as well as the object-based, open data formats, and network capabilities native to PC-based control.

One would be hard pressed to distinguish a PAC from a PLC by looking at the outside (Fig. 2). While there is no blatant distinction between the outward appearances of a PLC and a PAC, there are several key differences inside the box (see table, “Comparison of PLC, PC, and PAC attributes”). These differences include:

Multiple control disciplines — PACs allow the integration of multiple control disciplines —sequential, process, motion, and drives — into a single control system. The integration of motion and process control functionality eliminates the need for multiple controllers in many applications. However, traditional PLCs were designed for discrete applications, forcing many users to have a separate motion controller for motion applications, and a distributed control system for process applications. While these individual control solutions are ideal for their intended applications, integrating the various controllers for plantwide automation solutions is often a difficult, time-consuming, and costly task.

Flexible memory capacity — PACs have a modular, scaleable memory, allowing users to implement systems to meet current application needs, as well as easily expanding for future control demands. However, PLCs typically have a fixed memory capacity.

Rapid CPU processing — Control engines available today are fast, reliable, and accurate. Touting speeds up to three times faster than traditional PLCs, PACs can significantly increase throughput while maintaining high levels of accuracy.

Form factor flexibility — Typically, a user can select a PAC in a form factor most suitable for his or her application — chassis, DIN rail mounted, or PC type. Each form factor operates the same way, which significantly reduces training and spares costs and maintains flexibility to handle different applications.

Consistent development tools — The PAC pro- gramming environment provides a common “look and feel” from one form factor to the next. Using a common set of system tools simplifies application setup and maintenance, increasing productivity and lowering engineering costs.

PC features — Using commercial “off-the- shelf” chip technology inside the skin of the traditional PLC (industrially hardened, stable, “instant on” boot time, high reliability), PACs also incorporate “PC-like” functionality that users are demanding as a result of growing demands in the area of information management, supply chain integration, and regulatory compliance.

PACs (Fig. 3) combine the best parts of the PLC — the form factor, the rugged specs, and the reliability — with the best parts of the PC — the software functionality and flexibility, the floating-point processor, and RAM. The industry’s transition to the PAC is really a redefinition of the PLC — an evolution in nomenclature.

Peaceful coexistence

Using PLCs and PCs in automation systems is not an either/or situation. Both PLCs and PCs have their places and advantages in tomorrow’s factory as well as today’s. In many cases, they will continue to work hand-in-hand in optimized combinations based on application requirements. Hybrid applications that blend discrete and process aspects are becoming more commonplace, challenging the capabilities of each control approach.

So, what does all this mean? It means that PLCs have grown up. They have matured, evolved to the next generation. When the question is asked, “Which is better for industrial control, PLC or PC,” the logical answer is both. With PACs, the flexibility, multifunctionality, and speed of PCs running control software are combined with the reputation for robustness, reliability, and installed base of PLCs.

Do some plants continue to use PCs for industrial/process control? You bet! Do some continue to choose PLCs? Of course! Will they both be around to share the industrial control space? Count on it!

PLANT ENGINEERING magazine extends its appreciation to ARC Advisory Group, National Instruments, Omron Electronics, Phoenix Contact, and Rockwell Automation for the use of their materials in the preparation of this article.

Comparison of PLC, PC, and PAC attributes

Attributes
PC
PLC
PAC

(Source: Rockwell Automation)

Networking to higher-level platforms
X
X
X

Advanced control algorithms
X

X

Extensive database manipulation
X

X

HMI functionality in one platform?
X

X

Integrated custom control routines
X

X

Complex process simulation
X

X

Very fast CPU processing
X

X

Flexible memory capacity
X

X

Interfaces through multiple protocols
X
X
X

Wireless access
X
X
X

High MTBF, Low MTTR

X
X

Security, access levels
X
X
X

Auditability/traceability

X
X

Multiple control disciplines
X

X

Solution scalability
X

X

Backwards-compatible

X
X