Pharmaceutical processing is all about controlling flow, mixtures, temperatures, pressures, and a number of other variables that influence the quality and yield of the final product. Automating the control and verification of these parameters not only increases quality and yield, but also has the potential to accelerate manufacturing processes. This is where a modern controls strategy comes in. Automating valve operation makes continuous processing possible where previously batch handling was preferred.
STRUCTURED CONTROLS ARCHITECTURE
Imagine a system with a large number of valves and mixers. From the point of view of a distributed control system (DCS) or programmable logic controller (PLC), these devices are controlled by setting outputs. Furthermore, consider the system to have a large number of flow meters, pressure gauges and temperature probes. Generally speaking, these are the inputs to the control system. Historically, input and output devices were directly connected to suitable I/O cards on DCS or PLC backplanes. Because pharmaceutical plants tend to be large, the necessary cable runs are significant. Running large amounts of cable makes installation time-consuming and costly. Besides the cost of the cable, cable bridges, conduit, field-mounted junction boxes, along with the necessary labeling, are other cost drivers that make such installations complex, expensive, difficult to maintain and challenging to expand or update.
A well-structured controls architecture based on a modern industrial network has the ability to deal with those negatives and offer the pharmaceutical industry flexible solutions. These solutions not only reduce the initial implementation expenses and cost of ownership, but also add the ability to quickly modify the process by simplifying the addition of supplementary process signals.
CONTROL VIA THE AS-INTERFACE
For nearly 20 years, AS-Interface has been the preferred control network for low-level digital and analog data. Worldwide, over 24 million/1/ field devices have been installed. These field devices control everything from roller-coasters to automotive assembly lines to chemical plants. Pharmaceutical applications specifically can benefit from this technology, as it addresses many issues that plant operators are frequently faced with. How can processors benefit? The following should offer some enlightenment:
AS-Interface uses a 2-conductor power and communication cable. This alone simplifies installation and startup. In contrast to all other networking solutions availabe, AS-Interface does not require the network to have a predefined topology. Instead, branching is allowed without limitation. As seen in Figure 1, it is even possible to add child branches to parent branches. Since AS-Interface was originally designed with mainly discrete factory automation applications in mind, connecting I/O modules to the network had to be fast, easy and reliable. This goal was accomplished by using an insulation displacement technology where piercing needles penetrate a flat, 2-conductor, unshielded cable to create the necessary electrical connection. While process automation applications can take advantage of this installation method, it is common to see conventional round cable being used in most applications. Since the AS-Interface was designed for use with unshielded cable, it’s recommended system designers use lower cost, flexible unshielded round cable.
Creating a new network branch does not require special hardware; two simple wiring terminals will do just fine. Still, in cases where multiple I/O devices are to be placed in an area, AS-Interface suppliers are now offering junction blocks with M12 quick disconnects (Figure 2).
No installation is 100% correct on day one; expansions and modifications are facts of life. AS-Interface supports change by allowing modules to be connected or removed from the network while under power. Imagine that a new valve must be installed. The process could not be simpler. Technicians simply install the new valve and assign its desired network address using a handheld addressing tool; essentially a calculator-shaped unit with a small display and a few push buttons. The valve can then be connected to network. From the point of view of the DCS or PLC, the new valve is unexpected, resulting in some kind of error indication. But the important fact is that control of any of the previously present I/Os is not inhibited. The control system simply reports the fact that the hardware configuration is not what it used to be. With the valve connected, it is still necessary to accept the valve in the AS-Interface configuration and make logic changes.
Programmers will like the fact that in virtually all instances, it is not necessary to manually modify hardware configuration files. Most typically, the AS-Interface field data is transferred to the DCS/PLC through a gateway. In the past, PROFIBUS, DeviceNet, and Modbus were the preferred solutions. For several years now, Ethernet approaches like PROFINET, EtherNet/IP and Modbus/TCP have gained popularity. Because these gateways are large I/O, single module devices to the DCS/PLC, changes on the AS-Interface side are automatically available. For instance, the I/O occupied by the valve we just added had already been mapped in the control system. The data bits simply did not change. With the valve installed and accepted on the AS-Interface network (typically accomplished by pressing buttons on the gateway), the valve data is immediately available in the DCS/PLC I/O table.