Signal, power wiring technologies drive changes in best practices

A creative tension typically exists between accepted practices that have served an industry well and the need for new practices that take full benefit of innovative new technologies. The current state of signal and power wiring is a good example of this truism.

By Brad Woodman, Molex AEP Division August 28, 2008

A creative tension typically exists between accepted practices that have served an industry well and the need for new practices that take full benefit of innovative new technologies. The current state of signal and power wiring is a good example of this truism.

Since separation between power and signal wires is a critical electrical engineering function, companies tend to rely on traditional, field-proven practices. However, engineers are evaluating new Ethernet and fieldbus-based wiring technologies and methods that are, in some cases, changing the traditional approach to power and signal separation by combining power and signal wires in one cable. It’s important for engineers to understand the traditional approaches, the change drivers and how to leverage new approaches into recognized best practices.

Traditional approaches

The traditional practice of placing power and signal wires in separate cables is based on various established industry standards. The two most common standards are IEEE-518, Guide for the Installation of Electrical Equipment to Minimize Electrical Noise , and NFPA-70 National Electrical Code . IEEE-518 establishes four major wiring classes or noise susceptibility levels. Common practices usually take the four major classes and convert them into three key levels.

Level 1: High susceptibility with analog signals of less than 50 V and discrete instrument signals of less than 30 V. Examples of these signals include:

  • Various fieldbus systems such as DeviceNet, Profibus and Foundation Fieldbus

  • 4-20 mA signals

  • Discrete input and output signals such as proximity switches, limit switches and indicating lights.

    • Level 2: Low susceptibility with switching signals greater than 30 V, analog signals greater than 50 V and 120 to 240 Vac lines less than 20 A. Examples of this level include:

    • Discrete input and output ac signals — including pressure switches, limit switches, indicating lights, relays and solenoids

    • 120 to 240 Vac lines of less than 20 A.

      • Level 3: Power ac and dc buses of 0 to 1,000 V with current of 20 to 800 A. Examples of this level include:

      • Power lines for motors

      • Power lines for welders and robots.

        • Spacing distances for the different levels are outlined in Table 1.

          If a signal cable must cross a power line (Level 3), it should do so at right angles.

          The susceptibility level guidelines outlined in Table 1 are very effective and have been standard industry practice for many years.

          New technologies, standards

          However, new technologies are driving a re-examination of these susceptibility levels. These technologies include Power-over-Ethernet (PoE) and various fieldbus technologies such as DeviceNet, Profibus and Foundation Fieldbus. These technologies combine power and signal wires in one cable.

          Moving from point-to-point wiring to consolidated wiring methods that combine power and signal wires, such as trunk and drop topology, can provide cost savings and the ability to perform enhanced troubleshooting with diagnostics. This alone is an important reason to re-examine traditional susceptibility level standards.

          For example, existing standards set the maximum current power for PoE at 15.4 W at 48 V, as specified by IEEE 802.3af . This provides levels up to just 31 mA to one device. The current draft of a new standard, IEEE 802.3at PoE plus, specifies up to 30 W, which would provide a tenfold increase in current levels — up to 350 mA — to one device. This change would ease current limitations in the existing PoE standard that restrict the types of new devices — such as IP Phones, wireless access points, pan-tilt-zoom cameras, RFID readers and building automation — that can use PoE. A new standard would enable creative development of new and useful technologies.

          New standards will also allow for greater system operating efficiency — a key goal for advanced technology networks. For example, one of the advantages of fieldbus technologies such as DeviceNet is that many devices can draw power from the same cable since the communication pair is part of the same cable going to the device module. Currently, DeviceNet cables are capable of carrying 8 A for Class 1 cable or 4 A for Class 2 cable. The current capability is much greater than can be used in PoE under existing standards. A new standard would allow more devices to be connected directly to the network for improved diagnostics and troubleshooting.

          There are other avenues that lead to higher efficiency. For higher-current devices (greater than 1 A), most fieldbus systems use auxiliary power networks to supply current directly to the devices instead of drawing current from fieldbus network cables. These auxiliary power networks can also be used as power sources for networks that normally do not have power, such as Ethernet networks.

          System designers should also take advantage of information available for powering new information networks. ODVA, an international association comprised of members from the world’s leading automation companies, has a planning and installation guide for both DeviceNet and EtherNet/IP. The guide, accessible on ODVA’s Web site (, states the common separation distances for these network cables from other cables installed in an industrial facility.

          Also, several standards bodies are implementing into their Ethernet standards a concept called Mechanical, Ingress, Climatic/Chemical and Electromagnetic (MICE). MICE classifies the type of environment in which cables will be installed. The MICE classification is currently in ISO/IEC 24702, Generic cabling for industrial premises , and TIA-1005, Telecommunications standard for industrial premises . The MICE classification can help designers select proper cabling for individual environments. For example, designers may need to select a hardened connector/cabling solution or use local mitigation or isolation techniques to protect the connector/cabling solution.

          Moving toward best practices

          While the change process is not complete, the industry is moving toward innovative solutions and best practices that use new technology that is more flexible and can also reduce installation and operating costs. For example, changes to NFPA-79 and NFPA-70 in allowing factory-applied connectors molded to TC-ER cable types are leading to more flexible solutions for industrial premises. These changes allow a more flexible solution than the traditional wire and conduit solution.

          Other standards bodies are making substantial changes as well. ODVA and ProfiNet, among others, are moving toward defining a standard auxiliary power network that will simplify and streamline system design. Currently, ODVA is finalizing a common 24 Vdc auxiliary power network for all CIP networks. This network standard can be used for either DeviceNet or EtherNet/IP networks, and covers:

        • Topologies such as daisy-chain and trunk-and-drop

        • Connector types and pole count

        • Cables styles with current vs. network length tables

        • Power supplies.

          • Out of the control cabinet

            Another industry trend is to move both power and signal cabling and equipment out of the control cabinet and onto machines closer to devices that require the power. By making this change, system designers can gain more floor space by using a smaller cabinet footprint. As a result, more devices are readily available for use outside of the cabinet, which have been made environmentally hardened to IP65/67 levels such as I/O devices and Ethernet switches.

            For communication cables such as those used in Ethernet systems, Table 2 states the necessary spacing between high-voltage conductors and signal wires when wiring outside of the cabinet.

            Another trend is to use factory-floor diagnostics in auxiliary power systems for troubleshooting initial start-up operations and downtime situations. These diagnostic systems include a range of devices. For example, they can use simple devices that monitor voltage levels on power conductors, which indicate the voltage level by various colors of LEDs showing under/over voltage, and the normal operating range. Other, more complex devices can be used to monitor auxiliary power network current draw for unusual current spikes or in-rush currents.

            As the trend to higher fieldbus network speeds gains momentum across industrial applications and as more Ethernet networks are installed on factory floors, spacing between power and signal/communication conductors is critical because of the possibility of electrical noise coupling onto the cables. However, it is clear that the industry can, by working diligently to create and test new standards, develop solutions that safeguard system integrity while creating more flexible and cost-effective networks. The industry cannot continue to use standards that, while being highly effective in handling traditional power issues, do not take into account the changing world of signal and power wiring on today’s factory floor.

            Level 1 Level 2 Level 3
            Level 1
            Metallic conduits 75 300
            ON tray & in-trench 150 650
            Tray-conduit 100 450
            Level 2
            Metallic conduits 75 150
            ON tray & in-trench 150 200
            Tray-conduit 100 150
            Level 2
            Metallic conduits 300 150
            ON tray & in trench 650 200
            Tray-conduit 450 150

            Voltage Level Minimum Distance
            0-100 V 8 centimeters (3 inches)
            101-200 V 11 centimeters (4 inches)
            201-300 V 13 centimeters (5 inches)
            301V-400 V 16 centimeters (6 inches)
            Author Information
            Brad Woodman is a senior controls engineer at the Automation and Electrical Products Division at Molex. He is responsible for engineering design and support for industrial networks. Woodman has more than 20 years of experience in the industrial automation industry. He has three patents for industrial connector design and is the chair of the DeviceNet Physical layer SIG, secretary for TR42.9 and editor for TIA-1005 for Industrial Ethernet.