A worldwide fieldbus standard will combine the best parts of a number of
today's systems, says Lane--and it may not be that far off.
Design News: What advantages do industrial PCs offer compared with proprietary control systems?
Lane: Until now control systems have been pretty much run by PLCs, which are developed by one manufacturer who then usually supplies all the associated input/output devices and software. When you get on the factory floor level and start using a certain manufacturer's PLC, that tends to become the standard throughout the factory because you've got training programs to teach operators and technicians that particular language. With industrial PCs, training the workforce is easier, you get flexibility for reprogramming the system to do different things, and you're not tied to one supplier. Also, it's much easier for the factory floor to communicate with the entire enterprise computer system. It's very difficult to tie a PLC system into a total company computer system.
Q: Why would a company want to change from a PLC to an industrial PC?
A: Because industrial PCs have come down so much in cost and they're much more flexible than a PLC. They also have open operating systems like Windows, which everybody understands, and open software programs that can be bought from many different suppliers.
Q: Do you think there will eventually be a worldwide fieldbus standard?
A: That's a real difficult question. ISA has Committee SP50 investigating what will be called the Foundation Fieldbus. That's supposed to be-come a worldwide standard. The specification has been completed for more than two years now but it's not being adopted in a very rapid fashion. Some of the other systems--such as Siemens' Profibus and Phoenix Contact's Interbus S--have established themselves in the meantime while we've been waiting for this world standard to be written. Whether the Foundation Fieldbus will eventually succeed or whether a different fieldbus system will become the champion in the long term is difficult to say at the moment. I think that other systems will evolve from the ones that exist today and then gain worldwide acceptance and that by the end of this year one major new standard will be established.
Q: What is "block I/O"?
A: The problem is that some of the I/O for fieldbuses isn't open--you can only get it from the manufacturer who effectively invented that particular fieldbus system. And then, traditionally, that I/O comes in blocks of usually 16 channels, with all 16 being the same--all analog or all digital and all input or all output. Which means you're paying for multiples of 16 whether you need them or not. What you also need is a higher granularity--that is, a lower channel count per module. The WAGO I/O System offers modules with as few as two channels, so you only buy what you need.
Q: What are some other drawbacks of block I/O?
A: Block I/O isn't open, in that when you buy the block I/O you buy it for the particular fieldbus you have chosen. WAGO's I/O is not only flexible in terms of number of channels, but it's also flexible in terms of which fieldbus it can be used with. Once the I/O is wired, you haven't yet decided which fieldbus system it's going to work with. The last part of the assembly is the bus coupler, which makes that set of I/O unique to a particular fieldbus.
Q: Which fieldbus standards does WAGO's I/O System work with?
A: We support all of the common fieldbuses, including Profibus, Interbus S, ModBus, LonWorks, and the CAN fieldbuses. We're discussing whether to support the Fieldbus Foundation standard. It's going to be very expensive to support, and until there's a demonstrated use of the system we won't do that development work.
Q: What should a design engineer consider when choosing a fieldbus?
A: He or she needs to ask, What is the operation and how critical is a failure? For instance if you're running a nuclear power plant or a petrochemical facility, and the control system goes down, that's pretty critical. If you're running an assembly operation, a failure is less critical. Some fieldbus systems retain more information when the system goes down, and can remember where the process is when the system comes back up. You pay for what you get--the systems that retain the most data are also the most expensive.