Interview to mark 25 years of PC-based control technology
Interview to mark 25 years of PC-based control technology
The pioneer of Industrial PCs
Beckhoff presented the first PC-based machine controller as far back as 1986 – in other words, a quarter of a century ago – and, in doing so, was a pioneer in opening up the way for IT technologies into industrial environments. To mark the anniversary, elektro Automation interviewed the founder of the company and primary strategist, Hans Beckhoff, about the latest developments and future potential of Industrial PC technology.

elektro Automation: 25 years of PC-based Control. Mr. Beckhoff, congratulations are easy with the benefit of hindsight, but at the start, puzzled looks were probably in the majority. How did you view the situation at that time?

Beckhoff: That’s right, we delivered the first PC-based controller as far back as 1986. We then presented it to the general public in 1990 at the Hanover Fair. In the run up to the debut, we tested the new technology for four years together with many customers from the woodworking and window industries, among others. This mid-sized cooperation made a considerable contribution to the successful development of PC Control. Our customers recognized the advantages of PC-based control technology at an early stage and they trusted us! As a result, the PC characteristics in our control technology were quickly accepted. Suddenly it was possible to input data directly to the machine from a floppy disk; that was a gigantic advancement which was simply not possible with the standard control technology of the time. In addition to that, the PC made very high computing performance available, which for example, enabled optimization programs directly in the machine controller for cutting materials to size; that was also a first at the time. Our performance was much higher than anything achievable with conventional control technology back then.

We expanded our PC Control concept consistently and quickly in the years that followed: multi-axis NC functions and a complete, integrated software PLC enabled the PC-based Control of highly complex machines and plants. As mentioned previously, PC Control technology had matured to the extent that Beckhoff presented the concept to a wider public for the first time at the Hanover Fair in 1990. The professional world was surprised, skeptical and impressed all at once. The high performance and value for the price quickly convinced many customers, however. Starting with this Hanover Fair, PC control has become generally accepted worldwide as a standard technology in automation and has enabled the Beckhoff workforce to grow from 40 employees at the time to 1950 people worldwide today.

The pioneer: Hans Beckhoff, Managing Director of Beckhoff Automation GmbH in Verl

The pioneer: Hans Beckhoff, Managing Director of Beckhoff Automation GmbH in Verl

 

elektro Automation: How did the idea to develop PC-based control technology actually come about back then?

Beckhoff: We established our company in 1980, starting with microprocessor-based 1-, 2- and 3-axis Motion Controllers, combined with standard PLCs. As the machine manufacturers’ tasks became more and more complex, the single-processor system was no longer adequate and we developed multi-processor systems complete with their own graphics card and bus system. Beckhoff thus had an outstanding control platform by around 1984 and was also very commercially successful with it; further economic development was secured as a result. Then we saw a possible paradigm shift, however. We realized that our motion hardware could be replaced by software on a standard computer! The actual initiator of the software-based PC control concept, however, was a customer for whom we were to integrate a hard disk into our microprocessor controller for data storage. At that time, the idea of developing our own hard disk controller was a very daunting one. Instead, we decided to ‘hide’ one of the XT PCs from IBM, which at that time had just been introduced, in the control cabinet and to couple it to our microprocessor controller via the serial interface. The result was a microprocessor-based machine controller on an 8-bit processor with a 2 MHz clock frequency combined with a hard disk controller in the form of a 16-bit PC with an 8 MHz clock frequency and, even then, a color display. This discrepancy in performance data led us to the realization that the roles of the PC and the machine controller could be reversed. That’s how the idea came about to port our software know-how from the microprocessor world to the PC. The PC architecture proved to be astoundingly capable for real-time- performance and industrially compatible, so that we already had the first PC-based machine controller just six months later.

Grundstruktur der PC-basierten Steuerungstechnik

Basic structure of PC-based control technology

elektro Automation: Didn’t the image of the “unreliable office PC” play any part then?

Beckhoff: That was continually discussed, of course, but even our first PC controllers in 1986 proved to be suitable for harsh daily industrial production. On top of that, the PC had unbelievable advantages compared to the normal controller. Apart from the high computing performance, the integration of IT communication, outstanding HMIs and the openness of the architecture, the main advantage was the available memory space. Even 32 KB were almost unaffordable for a PLC at that time, whereas the PC provided 640 KB, a good half of which could be used as PLC memory. Machine manufacturers viewed that as a genuine liberation. In addition, the high computing performance and the short cycle times meant that they could realize not only a 3-axis machine, but also a 12-axis machine. In those days that was something of a sensation, because 12-axis controllers were actually unaffordable, but in our control system they were “just” a piece of software. Technologically, we were so far ahead that many new customers were also ready to accept the perceived risk of using the PC.

The all-in-one principle of PC Control: a PC takes over all essential automation functions

The all-in-one principle of PC Control: a PC takes over all essential automation functions

 

elektro Automation: Did porting the software to the PC mean the end of the development of your microprocessor-based control technology?

Beckhoff: Naturally, we continue to maintain our product series; that is to say, the further development of the microprocessor-based controller continued actively until 1989, although only to a very small extent. We still supply it today, by the way. However, we made a relatively hard paradigm shift starting from 1985; that is to say, we developed PC software and concentrated our hardware development on plug-in cards for the PC. Such a paradigm shift (in other words, questioning an existing product and starting a new product line) has taken place several times in the history of our company. For example, we stopped the further development of the fieldbus modules developed between 1988 and 1994 after the idea of the modular bus terminal matured in 1993. Sometimes the good just has to make way for the better. But of course we still supply the previously developed fieldbus modules today. PC technology and the associated fieldbus technology are in any case characterized by long life cycles – this characteristic is good for all users, although we didn’t see it that way at the start. There have been no fundamental changes to the successful concept of PC control since then: The abstraction of the control functionality from the device hardware and the consistent use of ‘mainstream’ IT technologies converging with the principles of automation technology lead, on the one hand, to continuously improving performance and functionality and, on the other, to cost reductions. The PC Control principle is very simple: a powerful Industrial PC, an equally high-performance fieldbus to which peripheral devices for sensor and actuator systems are connected, and control software with real-time capability for motion and logic – nothing more is required. Beckhoff has developed further important products over the last 25 years that could be considered milestones, since they each represent a smaller or larger revolution and have contributed in no small way to the current state of PC-based automation.

 

elektro Automation: A great deal has happened, from PC Control to the Bus Terminal and Industrial Ethernet. In your view, what have been the most important milestones during 25 years of PC-based Control?

Beckhoff: The first real milestone was naturally the presentation of PC-based control technology in1986, followed by the introduction of the first fieldbus system, our own fiber optic bus Lightbus in 1989. It was way ahead of its time, with a data transfer rate of 2.5 Mbaud, ring bus structure and hard determinism, and even then it could process 1000 inputs or outputs per millisecond, or it could control axes in real-time. We learned a great deal about communication during its development and several operating principles of Lightbus can still be found today in EtherCAT. The next milestone, likewise in 1989, was the first freely programmable, Step5-compatible software S1000, which served as a work platform for the next six years. That was followed in 1996 by the TwinCAT software generation developed for Windows-based systems.

Historic C4000 PC controller versus modern C6915 Industrial PC (left), the most compact Beckhoff IPC today, measuring just 48 x 164 x 116 mm (without mounting plate)

Historic C4000 PC controller versus modern C6915 Industrial PC (left), the most compact Beckhoff IPC today, measuring just 48 x 164 x 116 mm (without mounting plate)

 

Our first proprietary-developed PC motherboard in 1990 is also very important for me. As a universal, all-in-one board with FPU coprocessor (at that time, still a separate chip), integrated graphics, ROM disk, communication and fieldbus interfaces, it was a single-board PC and thus a single-board machine controller. That is standard today, but back then it was a little ahead of its time.

The next big step was without doubt the Bus Terminal in 1995, with which we took up the idea of the electrical terminal block and turned it into an electronic terminal block. From our point of view we brought a development onto the market that was not only evolutionary, but also revolutionary because, in our opinion, this has continued to affect automation designs to this day worldwide. The Control Panel Link (CP-Link) from 1998 is also an important development. This simple-to-handle option to connect remote control panels to the PC over long distances via two coaxial cables has been of great benefit to PC control technology with regard to robustness in practical use.
We presented what was in our view a further technical revolution in 2003: EtherCAT – Ethernet for Control Automation Technology. EtherCAT again demonstrates the powerful approach of uniting IT technology and automation with each other, in other words – using the convergence of technologies! Beckhoff has bundled its entire experience in real-time, communication and I/O into EtherCAT and made it available for global use. Today, EtherCAT is an IEC-standardized global standard; the EtherCAT Technology Group (ETG) user community encompasses 1600 industrial members and continues to grow. EtherCAT helps shift the limits of automation technology every day; it also provides the basis for high performance machine controllers, which then automate efficient, resource-saving machines and plants.

The next milestone followed in 2005 with TwinSAFE. TwinSAFE – that is to say, safety-relevant communication in unsafe industrial communication environments – is the basis for the integration of functional safety and control technology in a system architecture. We consider this development to be tremendously important and we will continue to push safety technologies forward in the near future.

Two newer Beckhoff technologies are particularly pleasing for me personally: XFC (eXtreme Fast Control) and Scientific Automation. XFC enables an ultra-fast control technology with which the cycle time can be reduced to 100 µs, thus improving the limits of control technology by an order of magnitude. Scientific Automation enables the integration of measurement technology and engineering science into control technology, opening up new areas of use and the further development of PC-based automation technology.

In 2010 we presented TwinCAT3 – once again a significant step for our users and for us, since we have integrated our engineering tool even more deeply into the IT world, with Visual Studio as the framework and C as a fully integrated programming language.

 

elektro Automation: Apart from these development highlights, are there any further aspects that particularly distinguish Beckhoff, the company?

Beckhoff: We are most obviously a technology company. For that reason in particular, however, it is important for us to cooperate closely with our customers. That’s why we have always pushed the expansion of our support, sales and marketing networks forward along with the actual technological development. We exhibited at a trade fair for the first time in 1990; today more than 500 employees work in sales and support in 30 Beckhoff companies worldwide. A further characteristic is our philosophy of mastering the fundamental technologies of automation in the smallest detail in-house. For example, our PC Control products are naturally based on proprietary-developed motherboards with our own BIOS; our real-time software is based on a patented real-time extension; in the field of communication, we naturally have own communication stacks and communication specialists for a very wide range of industrial protocols; our hardware specialists design our own ASICS. In I/O technology and Drive Technology we have a command of all ‘small and large currents and voltages’ combined with a great deal of intelligent software. And, of course, we have many automation specialists on board who are proficient down to ‘the last bit’ in their respective areas of automation - such as PLC, Motion, control technology and measurement technology - and we continue to push them forward. This deep understanding of the fundamental automation technologies opens up many further development possibilities to Beckhoff well into the future.

elektro AUTOMATION – at that time still ‘elektro anzeiger’ – published the first front page story on PC Control from Beckhoff as far back as 1992

elektro AUTOMATION – at that time still ‘elektro anzeiger’ – published the first front page story on PC Control from Beckhoff as far back as 1992

 

elektro Automation: The combination of PLC, Motion Control and robotics functions as well as the integration of C/C++ programming and measurement data recording are trends to which Beckhoff has also dedicated itself. How do you judge these today and for the near future?

Beckhoff: We are convinced of the convergence of IT and automation technology for 25 years and we think that the next consistent step will be for measurement technology and
C-programming to also become standards in automation. PC technology is quite capable of providing this ‘Scientific Automation,’ whereas the standard PLC control technology may have greater difficulty. More effective machines will need to be built in the future and that’s why we are sure that the concept of Scientific Automation (in other words, the integration of measurement technology and automation technology) will extend the application possibilities for our PC control technology. The PC Control revolution will continue, driven by hardware and software advances in IT and automation technology (AT). Multi-core technologies allow the provision of virtually unlimited computing power and new software tools are improving engineering. All automation engineers are called on to capitalize on these increases in performance, for example, to control machines, plants and buildings with greater efficiency and with less use of resources.

 

elektro Automation: The EtherCAT real-time Ethernet system has also been successful for eight years. How do you assess its development and future potential?

Beckhoff: Above all, I see three success factors here: first of all the technology itself, which is simply convincing for engineers and most attractive as a result. EtherCAT offers precisely what engineers need and they understand the unique selling points of the technology. Added to that is the consistent disclosure. Absolutely any company that is interested in the implementation of EtherCAT may join the ETG user organization and get support. Even though Beckhoff created EtherCAT, the ETG members are actively contributing to the further development of the technology – this way, profiles are defined for different applications, for example. Thanks to this cooperation, EtherCAT was quickly ‘rounded off’ and ready for use on the one hand, while on the other, the basic technology itself has been stable since 2004. We consider that to be a further important success factor. However, EtherCAT offers further development potential in every respect. The latest milestone here is the integration of the EtherCAT interface into several microprocessor product lines, which Texas Instruments announced at Hanover Fair 2011. EtherCAT is thus becoming a mainstream technology, even beyond the pure automation market. We are naturally very pleased that EtherCAT is the first real-time Ethernet technology to have made the jump into standard products.

 

elektro Automation: Beckhoff has also recently become a servomotor manufacturer through the joint venture company Fertig Motors. How does this fit into the image of a PC-based automation company?

Beckhoff: We are a PC-based automation company, but essentially we are more broadly an automation company. It naturally makes a great deal of sense for an automation company to have a command of the entire range of automation devices, in other words, not just the control technology, the intelligence, of the machine, but also the actuators, and these include the motors and the corresponding drives. We are very proud of the fact that we are not only masters of the motherboard BIOS, the last bit in the I/O sector and 100 µs real-time, but now also of magnetic field design. It will be possible to achieve very good synergies from the combination of all of this knowledge. This applies in particular to the balance of functionality; in other words: what does the PC take care of in a drive controller, what does the Servo Drive do and what is included in the motor itself? Our communication know-how will also flow decisively into our new motor designs. We are very confident of being able to present the first products from Fertig Motors by the end of the year!

Published in: elektro AUTOMATION 6/2011