Over 30 process-automation network protocols exist in use today. Why is this important, and how do you decide which ones are best for your heat-treatment application?

Peter Sherwin from Invensys Eurotherm discusses this topic with Chris Mooney and Mikael LeGuern (process experts in control and power) to understand applications and reasons for the different (language) protocols.

Overview of Network Protocols

[Peter] Chris, why do we have network protocols, and how are they used in a heat-treat process?
[Chris] Simply put, network protocols allow us to transmit data from point A to point B (or point C, D, E, etc.). This is relevant in heat treating because we can have a master programmer device sending out instructions to many so-called slave devices as well as communicating with a central SCADA unit.

Most modern process equipment – from the industrial controller handling your furnace temperature to the PLC or PAC controlling your mechanical motion – can be outfitted with some manner of fieldbus communications.  

[Peter] What is the current status of these protocols?
[Chris] The older, costlier standard of retransmitting data with multiple 4-20mA signals is being replaced by Modern Industrial Communications. These newer networks allow a wide variety of information to be communicated from sources all over the plant floor, sometimes over a single network cable (significantly reducing the cost and complexity of wiring). Whether you are reading the power-draw levels from a network of SCRs to calculate your power factor or the flow from a flow meter to monitor your gas usage, the data is available, accurate and does not suffer from the retransmit errors associated with nondigital communications.

Industrial communications networks allow an engineer to connect dozens of field devices and share data between devices for process-control purposes with corporate leadership who analyze this information to determine performance against a set criteria of KPIs (Key Performance Indicators).

These communications networks all require some language, or protocol, to transfer data. Increasingly, Ethernet-based communications are becoming more prevalent amongst the numerous protocols available to suit a wide variety of products and applications.

[Peter] What are the main protocols in use today?
[Chris] Among these various protocols there are two types that are most widely used: CIP and Modbus. CIP, or Common Industrial Protocol, was developed by Allen Bradley (of Rockwell Automation) and includes a number of network hardware types. CIP contains older serial hardware-based network platforms, with Controlnet and Devicenet being the most popular. But the most recent addition to the CIP family is EtherNet/IP, an Ethernet-based protocol.

EtherNet/IP
EtherNet/IP, or Ethernet Industrial Protocol, is an open network that makes use of most networking hardware available off the shelf, communication lines and physical media. What this means to a plant engineer is that a simple CAT5(6) cable is all that is needed to connect the devices on the network. The creation of Ethernet-based communications stems from the need for interconnectivity between wide varieties of networking platforms – both industrial and business-related. EtherNet/IP does not require its own independent network. It can operate under an established plant network. Functionally speaking, EtherNet/IP communication offers the ability to move large amounts of data at very high speeds. This allows the passage of real-time information to all levels of an organization, from plant floor to top management.

Like all of Allen Bradley’s protocols, EtherNet/IP is considered an “open protocol.” This means that it conforms to the Ethernet IEEE 802.3 and to the CIP standards. The CIP standards are governed by an organization called ODVA, which maintains all compliance standards. If a device wishes to be ODVA compliant, it must be tested and approved at an ODVA-approved lab. The benefits of open protocols provide users with the surety that connecting with other ODVA-compliant devices should be trouble free. But there are other protocols that offer similar advantages as alternatives to the CIP platform, such as the Modbus protocol.  

Modbus
Modbus was developed in the 1970s by the PLC manufacturer Modicon (now owned by Schneider Electric). Like CIP, Modbus is an open platform but is instead governed by the Modbus Organization. Like Allen Bradley’s CIP, Modbus has both a serial and Ethernet base. Modbus RTU is a serial platform that makes use of the widely available single-drop RS232 or multi-drop RS485 network. Presently, most Modbus networks operate on an RS485 network rather than 232 as it is faster and offers the ability to address multiple units on a network. While originally on serial links, Modbus is also available on an Ethernet-based network called Modbus TCP. Again, because of the ubiquitous nature of business networks, Modbus TCP can coexist on a plant network alongside normal network traffic again allowing data flow from plant bottom to top management.

Even though these two Ethernet-based protocols are as prolific as they are, not all field devices are available with both choices. As communications become more complex, many of the newer types of networked devices can operate multiple protocols from one Ethernet link.

Data Capture
A primary use for these communications protocols is data transfer and collection. Additionally, field devices can send process variables to controllers and monitor devices such as HMI panels and SCADA software packages. This data is ultimately collected, stored and either analyzed to make the process more robust and efficient or saved for regulatory and customer-demand purposes. Depending on the level of security and regulation a company needs to meet, this data can be found in a few formats.

Most data collected over these communication networks can be viewed in an open CSV format. Once in this format, the data can be brought into SQL databases or imported into various analytical software packages. However, data security is a concern with this open format. From the moment the data is captured, there is no security. The data is free, open and can be tampered with without anyone’s knowledge. Many heat-treating organizations follow strict regulations for data surety (e.g., AMS 2750D requirements), and storage in a CSV-like format is unacceptable.

The good news is that there are a number of companies that now offer data-acquisition platforms (both hardware and software) for processes that need to meet strict tamperproof regulations. Included in this list is the nanodac Controller/Recorder (Fig. 1) and 6000-series Chart Recorders from Invensys Eurotherm. For the majority of products in this category, the data files are checked for attempted manipulation against some proprietary algorithm. This is critical to ensure that the data can be trusted implicitly. These historical records must also be kept for extended periods of time – even years – so it is important to question the memory capacity of these units.

[Peter] Chris, what protocol options do typical controllers have?  
[Chris] Serial communications (e.g., Modbus RTU) are typically found in the less expensive, smaller, single-loop controllers, but as you move up in complexity of control and have more information/data that can be passed from device to device, there tends to be a variety of protocols offered.

Applications of Network Protocols with Power Controllers

[Peter] Mikael, your expertise in power controllers and knowledge of network protocols has promoted the use of a variety of network protocols within the latest Power Controllers/SCR Technologies from Invensys Eurotherm. What practical advantage do these protocols offer a heat treater?  

[Mikael] Over the past few years, we have included network protocols in the latest EPower Power Controllers, especially Ethernet protocols like EtherNet/IP, Modbus TCP or Profinet. While at the core these are still SCR units, the inclusion of protocol technology has allowed these devices to start communicating to each other and the outside world. To explain the benefits of this technology, we can think about splitting this into two areas: (1) communicating process, diagnostic and energy information to HMI or SCADA systems and (2) communication between devices to maximize energy-efficient usage.

Communicating Energy Information to HMI or SCADA Systems
The electrical power controller is determining the use of energy in your process in KW/hour. In an energy-intensive industry such as heat treatment, control, data capture and analysis of energy use is more important today than it has ever been in our history. The base of the power controller is the SCR technology acting as a fast electrical switch to allow electrical energy to flow to the heating system of your furnace (see sidebar for Power Controllers/SCRs vs. Contactors).

Today’s power controller not only acts as this fast switch but is configurable over a network. It also has the ability to monitor the electric usage and, using network protocols, can provide this data to a DataLogger, HMI or SCADA system. This data can be translated into information about energy use per process (particulary on a batch-based system), idling energy use and peak-demand information. This information can then be used to accurately predict future energy use running a specific process (helping pricing calculations) and also to give peak-demand information that can aid decisions in avoiding blackouts due to running over capacity. However, this information is after the fact. What can we do with current technology to act on data/information in real time?

Communicating Between Power Controllers
Utilizing network communications, it is now possible to link power controllers together. Why is this important? If you have a multi-zone furnace or multiple furnaces with a number of power controllers, you can now have automated fast communication between these devices. This means you can schedule the firing of these units to avoid firing at the same time, thus minimizing the peak-load demand and also balancing the power across multiple controllers. Everything is done automatically at high speed with minimal configuration from the operator.

Without rapid communication between the devices, the time lag would delay firing and impact heat-up times or, worse, impact the ability of the furnace to maintain process temperatures. Not only can you address issues with peak demand, if you have multiple processes using power controllers, you can use a shedding technique to keep below a user-defined set limit on power/current usage to prevent overloading the grid or to reduce cost of energy by limiting demand charge.

Latest Network Technology and Freedom of Choice

SIDEBAR: Power Controllers/SCRs vs. Contactors

Older-style mechanical contactors have been used for many years in controlling the heat input to furnace heating systems. For high duty cycles, however, life of these contactors can be measured in months due to mechanical wear. Other drawbacks are their slow speed in switching cycles – typically not less than 30 seconds. The advent of power controllers and SCR technology has improved the life of the switching mechanism (these are fast electrical switches and do not suffer the mechanical wear), and their switching speed is measured in the low milliseconds rather than seconds. This precise control of the load results in holding tighter temperatures and typically provides the user with a more consistent and reproducible process cycle.     

Modern power controllers have the ability to operate with different firing modes (even within a single cycle) gaining an optimum balance of energy costs through load balancing and shedding, speed of operation, power factor and harmonic disturbances. To read more about power controllers, please refer to the white paper “Energy Cost Reduction through Load Balancing & Load Shedding” by Mikael LeGuern at http://www.eurotherm.com/products/power-control/epower/.