May 2010

SPECIAL: 9th Anniversary
ENGINEERING REVIEW - MAY 2010

ENGINEERING REVIEW - MAY 2010
Other Archives
INDUSTRIAL AUTOMATION SECTION TECHNO FOCUS

PC drive automated system: The gradual evolution

The idea of using a PC to drive automated systems was originally met with considerable skepticism. PCs, after all, crash, which could lead to production downtime and critical data loss.

Through the ’90s and the past decade, PCs became indispensable to every business, and naturally, many people and vendors started looking at using them to control machines and processes on the factory floor. Now, some automation vendors and systems integrators are saying PCs can be used for machine and process control in an increasing number of applications.

The idea of using a PC to drive automated systems was originally met with considerable skepticism. PCs, after all, crash, which could lead to production downtime and critical data loss. For example ; Siemens’ solution was its RTX real-time operating system, an OS that runs alongside Windows in the PC that keeps going even if Windows stops for some reason. Siemens has been demonstrating it in operation for years and is making headway in convincing the market that it’s viable. Some, however, maintain that PCs just aren’t cut out for real-time manufacturing automation control.

Very early on, some manufacturers noticed PCs weren’t always performing as quickly or reliably as their old-fashioned PLCs. Some users who switched from PLCs to PC control have gone back to PLCs. They’ve found that it’s more difficult to get support for PC control. One reason is that regular updates to the operating system or control software often requires updates to all the drivers. And costs of PLCs have fallen recently, as well.

Controllers have gotten more sophisticated, faster and able to handle larger programs today, today, manufacturers who want to automate can choose between PLCs, PCs and a hybrid of the two to control processes and machines, monitor factory floors and ensure safety. But how do they choose?

Strengths and weaknesses
PLCs were developed specifically for the manufacturing plant: dedicated to a limited set of tasks, they’re rugged, resistant to the rigours of the factory floor, such as vibration, heat and dust. A PLC’s main advantage, though, is not so much that it’s fast — today’s PCs have processing speeds that are orders of magnitude faster — but that its simple operating system (OS) means its main processing power is always available for the unit’s main task. The most popular PC operating system, demands priority on the central processing unit’s attention. In other words, if the machine that the computer  controls sends a signal at the same time that the OS sends a routine instruction to the CPU, the OS instruction will be processed first, forcing the machine’s signal to wait. If you’ve ever been frustrated by the hourglass symbol when you’re doing fairly simple work on your PC, imagine if that happens when a critical alarm signal from a manufacturing operation has to wait for the OS to refresh.

Worse, conflicts for computer processing cycles from different processes can cause the computer to crash, which can halt production lines or critical operations and even put people at risk if the system is controlling a safety-related process. Because PLCs have been around for more than 40 years, their use is well known in manufacturing settings; programming them is relatively simple, and in most manufacturing plants, electricians know how to set them up and troubleshoot them.

But their functions are limited, like their programming language. They can handle input and output, but complex tasks are difficult to program using ladder logic. Setting up different responses to a wide range of circumstances that might occur is far easier in one of the richer, more modern programming languages used in personal computers.

Another limitation is that PLCs are proprietary systems; while manufacturers can choose from a wide range of suppliers. This limits the manufacturer’s choice and flexibility. The personal computer, however, is (if anything) flexible. A multi-purpose device, it’s well suited to handling a wide range of tasks and integrating data in a way that’s easy to understand.

And while PLCs are well accepted and widely understood on the factory floor, PCs are equally well understood in the office. PCs are good at linking information from the factory to the front office. They’re often used for monitoring efficiency and displaying activity to management.

Today’s PCs use very powerful computers with extremely high data speeds compared to PLCs. They can handle images at high speeds and can be made compatible with many other computers, networks and even PLCs. And because they’re so widely known, there are a lot of people available who know how to program them.

But PCs are not typically rugged systems; they’re susceptible to heat, dust and vibration and have to be “hardened” or “ruggedized” for use on the factory floor. They’re also susceptible to computer viruses (something PLCs don’t have to worry about) and being hacked. And worst of all, they’re prone to crashing. This can lead to loss of data, which can be a huge problem and cause expensive delays when the PC is controlling critical manufacturing operations.

PLCs are good for interlocking operations, and fast decisions. They’re not so good with high-speed data collection or complex programs, PCs are better suited to monitoring, optimization and analysis — making sure you’re getting out of a process what you should be.

The hybrid approach
However, PCs and PLCs are not an either-or choice. Almost every systems integrator advocates using them side by side. PLCs are best suited, for example, in fault-tolerant applications, where you have two PLCs running side by side so that if one goes down, the second kicks in instantly.

PC in a PLC
Luckily, manufacturers don’t really need to choose between PLCs and PCs anymore. The evolution of the programmable automation controller (PAC) has removed the barriers between PCs and PLCs. PACs combine PCs and PLCs, adding the newest high-speed and flexible microprocessors to the rugged, reliable operation of PLCs and using newer, reliable, customized operating systems. Combining the two types of controllers means PACs are being used not only for process and machine control but also data acquisition, machine vision and remote monitoring. They can handle multiple communications protocols, including PC protocols from TCP/IP to OLE for process control (OPC), and PLC Fieldbus networks such as Profibus, RS-485 and others. They can also have USB ports and Ethernet connections.

PACs can use more sophisticated programming languages like structured text, which makes programming them for math functions more intuitive. And they can handle tasks that can’t be programmed with ladder logic.Simply put, choosing a side is not a necessity. With ever-evolving technologies and hybrid approaches, it’s just a matter of settling on an approach that fits the application.