In the words of the Profibus Trade Organization (PTO), "Openness and performance has catapulted Profibus to the number one position in the hotly contested fieldbus market as indicated by market research and by the rapidly increasing number of Profibus ASICs sold."

To bolster this view, PTO brings out some strong graphs and charts, surveys and commentary. Among these is a study from research company Consultic Marketing in Germany. The results show a 41 percent market share for Profibus in Europe and 43 percent in Germany. Profibus, it should be noted, is an extension of an architecture developed in Germany by Siemens. However, this study was done in 1996 and while the trend line — from 20 percent market share in 1992 to the 40s in 1996 is impressive growth, the market dynamics makes a two year old survey seem somewhat obsolete. As one automation consultant says, "Two years in this business is a millennium."

In addition, other surveys and studies along the same lines have produced different results. The Interbus users group posted the following on their web site: "ARC (Automation Research Corporation) in the United States found, at the end of 1996, Profibus had a ratio of 732,000 nodes worldwide which corresponds to 18.3 percent of the market. However, the ARC survey concludes that Interbus, with a global market share of 37.4 percent, is the clear world market leader. With an installed base of what is now 4 million nodes worldwide, this is a share which represents 1.5 million Interbus devices."

So, which research is right? Interbus-S competes with Profibus in most automation markets and has a

history going back more years. In fact, the Interbus association recorded its 1,000,000 chip (ASIC) in 1995 compared to 2 million ASICs for Profibus in 1996. Ah, statistics.


Why should we care?

"Solutions for automation demand open and continuous concepts. Technically optimized solutions should be viewed with international competitive and price pressures in mind. To meet these challenges, a fieldbus system is required that is both powerful and low cost," quotes PTO.

Most of the players, regardless of architecture, will agree with that. However, in a field that has been the exclusive realm of proprietary devices and architectures for decades, going open isn't easy. And it hasn't been pretty watching multiple device buses battle for the thorny crown of "standard for the world." Each of the dozen major and dozens of minor bus designs can claim some benefits. After all, when they were developed — and many are outgrowths of formerly proprietary buses — they were developed to give their vendor a competitive advantage in the market. But in an open world, they also have the baggage of being optimized for a particular vendor's hardware, PLC (programmable logic controller), software system or other element in the automation network.

The technical details are best left to those with the hands-on expertise necessary to define the benefits desired and the compromises that are acceptable. Most of the market leaders — which besides Profibus include CAN/DeviceNet, Interbus-S, and Foundation fieldbus — can rightly claim to offer an optimized network for smart automation devices. PTO counters that its fieldbus specification, now that it is gaining adherents and manufacturing experience, is also cheaper to make and install. When you consider the thousands of devices found in a typical automation system, cheaper with comparable reliability and data throughput means a lot.

In the way of technical specifications, PTO offers, "Profibus is the open fieldbus system that makes short response times possible and is also capable of transferring large data volumes. In a single message, up to 244 bytes of pure user data can be transported. The combination of speed and large data volumes opens up solutions that will not fail regarding bit counts or because of required response times. Because it is capable of intrinsically safe transmission according to IEC 1158-2, Profibus meets the preconditions for use in the process industry as one of the first fieldbus systems. Reduced installation costs, simple expandability and distribution of the intelligence are the primary motivation for migrating from conventional wiring to serial bus systems. This makes sense only if everyone can connect devices to the bus which means the system has to be open and standardized. In the case of fieldbus systems that are not open, the adaptation cost eliminates any cost savings. This is why Profibus was specified and implemented as a open system from the start."

Profibus is standardized under the European Fieldbus Standard EN 50 170 as well as IEC 1158-2. It boasts an installed base of more than 2 million nodes in over 200,000 applications as of mid-1998. PTO itself is comprised of user groups and committees in 20 countries in Europe, North and South America, Africa, and the Pacific Rim. More than 675 members make this the biggest fieldbus interest group worldwide.


Big and Bigger?

But is larger necessarily an indication of better? In the eyes of some, no. Profibus has the distinction of being one of the first large scale automation intelligent device networks that was implemented in silicon and that gave it a head start over other architectures such as the Foundation fieldbus. Getting the chip sets into the market quickly, at a reasonable cost, allowed PTO to capture a share of open systems mind-share as well as market share. However, there are other voices shouting for your attention.

"The usage of Profibus is considerably different from building products with ASICs," counters long-time Profibus-user Rob Hulsebos. Hulsebos, based in Eindenoven, Netherlands, first came into contact with industrial networks in 1984 and has developed several modules for the Philips P8 PLC for all modern fieldbus systems. He is also the author of several books about fieldbus technology, published more than 30 magazine articles about this subject, and gives seminars. As a user, he sees Profibus — and its competitors — in a different light than the device vendors, even those he's worked for. "Considerable confusion has been caused in the past by the different Profibus-versions: FDL, FMS, DP, and PA," explains Hulsebos. "When buying equipment, the buyer should take care that the correct protocol-version is delivered. Many vendors automatically assume DP when not specifically mentioned otherwise. In addition, a large distinction should be made between �systems' and �standards.' Profibus is a standard: a few inches of paper. And while it is up to the vendors to make all the products that are needed to build a network, it is up to the user to get it to work."

Many so-called "closed" networks are more system-like, since the vendor makes sure that all equipment works together. For the user it is often a matter of "plug and play". It is also in the interest of the vendor to make sure that all other equipment in the market is strictly conforming, since any deviation may give rise to interoperability problems. Even when such problems can not really be blamed on the network itself, for many users the distinction is non-existent. It is here where many networks that can not be named "open" are still successful contenders in the fieldbus market — for example Interbus-S.

Hulsebos continues his critique by saying, "The Profibus/FMS protocol is specifically intended for distributed intelligence applications. Actually FMS is based on MMS from MAP (Manufacturing Automation Protocol), although much simplified. Profibus/DP is not suited for distributed applications, since it is a master/slave protocol, meaning that the slaves can not communicate with each other. Although FMS is conceptually sound and no problems have been found in the protocol specification, it suffers because it is seen as too complex by many users. This causes major problems during commissioning, since many a user does not know how all the parameters must be set. It is one of the drawbacks of many an "open" fieldbus, since openness usually means flexibility — at the expense of requiring many configuration parameters that the user must set before use."


A Speeding Bus

In most of the debate, Profibus makes sure the issue of bus speed is brought up. That makes sense when you realize that speed is one of the selling points of the design. However, there is speed and then there is speed. "Claiming that Profibus — running on 12 MB — is a fast system, is true," notes Hulsebos. "But the reason for running at such high speeds is to counter the inefficiency of the Profibus-protocol. The original Profibus/FMS protocol, running at a maximum speed of 500 KB, proved far too slow for use with remote I/O. After two years, the Profibus/DP protocol emerged with a maximum speed of 1.5 MB, which was increased to 12 MB.

"The bitrate of the network itself is not important for end-users," he claims. "The only useful comparison between different fieldbus systems can be made when the I/O cycle-time is compared. CAN and Interbus-S have far more efficient protocols. Of course, when Profibus is set to run at 12 MB, it will be faster. Still, increasing the speed of a network is not without its problems. It makes the network more susceptible to noise and other electronic disturbances, and the quality of the cabling is more critical. More important is the fact that the faster the network, the smaller the maximum distance it may cover. Thus, quoting the maximum speed of any network doesn't say anything!"

There are several organizations around the world trying to determine which, if any, fieldbus architecture should be given its blessing as a open standard. According to Hulsebos, "In Europe, EN 50 170 standardizes three fieldbus systems: Profibus (originally from Germany), FIP from France, and PNet from Denmark. The document was developed simply by taking the three national norms, translating them into English, and stapling them together. Since these three systems are incompatible with each other — and describing them together in EN 50 170 does not change this fact — EN 50 170 brings nothing to the user. It only creates more confusion, since unknowing users assume that having the three systems in a single standard automatically enforces compatibility. Unfortunately, it doesn't."

So much for "open standards." With the demand for added automation throughout every facility, the intelligent device network is fast becoming a management as well as technical concern. The additional demand for Year 2000 compliance at every level of computer system, including embedded and device level chips and networks, makes adding yet another layer of computer functionality suspect. But we can't stop and wait to see what happens. Those that do court catastrophe. Choosing the right system, from the network to the mainframe is critical. Technical considerations must define the choices, but business concerns must drive the choices.


The Choices are Yours

Is Profibus the best choice for an automation intelligent networking architecture? Many companies have made, and are in the process of making, that decision. But the competitors are drawing new lines and focusing on benefits as well as differences to attract the technical and business specialists who must see the long range implications of any installation of this magnitude.

The Profibus positions is summarized in this way: "Future fieldbus systems will have to carry out more and more system functions. The trend is to distribute more and more intelligence in addition to I/O components and this demands a fieldbus that is able to efficiently transport large data volumes such as system files in addition to I/O data. When planning such systems, counting bytes is a thing of the past. Profibus with its enormous speed and its high capacity for data volumes offers the foundation to answer such challenges. In addition to the performance data, openness and interoperability means that to invest in Profibus today means to be a part of tomorrow."

And Hulsebos? His summary is, "Although Profibus is not without its history, it is still one of the most capable fieldbus systems on the market. The combination of Profibus/AS-Interface is a very good one, which makes it suitable for the lowest three application-areas of industrial networking: sensor/actuator, remote I/O, and inter-PLC communication. No other open fieldbus offers this versatility." Just watch out for the hype and don't let it get in the way of your sense of the strategic business issues that automation is supposed to make easier.

Web Site © Copyright 2020 by Lionheart Publishing, Inc.
All rights reserved.


Lionheart Publishing, Inc.
2555 Cumberland Parkway, Suite 299, Atlanta, GA 30339 USA
Phone: +44 23 8110 3411 |
E-mail:
Web: www.lionheartpub.com


Web Design by Premier Web Designs
E-mail: [email protected]