Will the need for Programmable Logic Controlers (PLCs) be eliminated by PC-based Controls?
In my opinion, the simple answer to this is no. While today's trends in industrial automation favor new technology, there are four reasons to believe that PC-based controls will not push out PLCs.
In my experience, the most reliable equipment that exists within my customer base are the PLCs themselves. I’d have to estimate that an overwhelming majority of service calls that I make are not due to PLC related failures. In fact, in one facility in particular, over the last 7 years that I’ve been their go-to Automation Engineer, there has only been one emergency call due to an outright failure of a PLC processor. An overwhelming majority of the remaining service calls during that time were due to PC-based HMI failures on the plant floor. In most instances these failures did not lead to downtime, but rather to an inconvenience to the operators on the floor as the lines in the facility can still run from other HMIs while the PLCs continue to run. Imagine the downtime that would have been lost over this timeframe had the plant floor been run by a PC that failed numerous times.
In many applications, PLCs are implemented as the controls on small machines or in remote unmanned stations. In the case of a small machine, it is simply not practical for a PC to be mounted somewhere on the machine itself. In many cases, these are in dirty environments that are not climate controlled and not well maintained. Even a good quality industrial PC will not hold up well under these conditions. PLCs on the other hand can withstand extreme temperatures, and still power up and run even in the worst environments.
With the current cost of many Micro Brick style PLCs, it is not cost effective to use an industrial PC with I/O in place of a Micro PLC. Take, for instance, the fact that these PLCs only cost a few hundred dollars on the low end. For many small machines, this is all that is needed for control. Try replacing this with an industrial PC with I/O and your cost immediately goes into several thousand dollars. Even if we look at a larger PLC based system, for example, where the cost of the processor goes into the thousands of dollars (not including field I/O). Again here the industrial PC that could, in theory, take the place of the PLC will cost several thousand dollars. When you factor in the I/O, it can work in the favor of the PLC. With the PLC having the capability of local in rack I/O, no additional adapters are required to control it. With the PC there is the need for a communication adapter to control remote I/O which adds to your overall cost of the system.
I have worked with a number of smart devices that have their own built in PLC like functions that could be considered PC or microprocessor based controls. These functions are rarely utilized; this seems to aim for just keeping things simple. Why over complicate your control system with different devices performing logic at different levels all throughout your plant, when it can be centrally controlled from your PLC based system? It certainly makes troubleshooting a control system a whole lot simpler when one can go online with a familiar PLC software and identify the problem. With all of these different devices having their own configuration software and programming languages this creates a lot more strain on the person troubleshooting it, especially if that happens to be the plant maintenance who may only be familiar with ladder logic with one type of PLC.
I could go on and on with reasons from my own experience on this topic, but I will leave it here with the four main points. I could see PC based controls becoming a larger threat to big expensive DCS systems, but that is another topic. In today's world, I cannot see the need for PLCs being eliminated.