Simple Connect Network Hardware and Switches

General Overview:

The Simple Connect “PwrNet” cable provides 4-wire Ethernet connectivity and 24VDC power on a single custom designed cable. The cable can distribute 10 amps of 24VDC throughout the robotic system.

In addition to distributing power, the cable also carries 4-wire Ethernet that is used to form an Ethernet/IP network to various devices on the network. This “PwrNet” cable thereby allow daisy-chaining of devices which results in:

• A minimized number of needed ethernet switch ports
• Minimized cable connections to the robot controller/tophat
• Less overall cabling and costs

Bypassing an off-line device:

There may be instances where you wish to physically disconnect a disabled/damaged device. This can be done by removing the cable connectors from the Simple Connect device and mating the cables to one another as shown below. This removes the device from the physical network but additional steps are needed to remove the device from the software depending on device type. The cable connections below allow the network itself to continue to pass signals and power up/down stream in open or closed ring configuration. See the separate post entitled “Removing a Simple Connect device” to disable the device from the software and scan list.

Star vs Ring Networks:

Simple Connect supports either of two common network architectures depending on the type of Ethernet switch that resides within the robot cabinet. Let’s first explain the differences in cabling between the two types of networks:

Star Network:
A “Star Network” connects each device to the ethernet switch directly and does not allow daisy-chaining devices. Many devices only support “Star Networks” because they only have one Ethernet port (LAN) so they must use the switch to communicate device-to-device.

Ring Network:
A “Ring Network” can be formed by connecting from one device to another without using a switch in between the two devices. This however dictates that the device itself must have two LAN ports as it acts as a switch itself and the device relays the data not intended for it to the next device in the chain. This minimizes number of ports needed on the switch itself but requires that all devices on the chain be powered so that the messages can get passed along the “Ring”.  When combined with a managed switch and connected in a “Closed Ring” network, redundancy can be achieved so that if one cable is broken or disconnected traffic can re-route to avoid passage across the broken segment.

PwrNet Connector Module Fusing:

The PwrNet Connector module provides access to the Ethernet signals & 24VDC power at an individual device. This module allows a transition between a rugged exterior cable & connector to within an enclosure where more commonly accessible cables and induvial wires can be used to distribute the signals to individual devices.

Ethernet Connections:
The Ethernet signals are made available via two RJ45 connectors on the backside of the connector module.  These two RJ45 connectors are then connected to the Ethernet module being used and that Ethernet module (Turck module) then serves as a switch taking messages in one port and passing messages not intended for that particular module out the other port where it message is passed downstream in a ring network.

24VDC Power Connections:
Access to the 24VDC power that is available on the PwrNet cable is made available for individual devices on the backside of the connector module. There are two possible pick-off points for 24VDC that are separately fused. Typical fusing is 2.5A at various devices (reamers, op-station, beacons) or 6.7A at the robot cabinet where the 24VDC is injected onto the PwrNet cable. The 6-pin black connectors next to the fuses are the pick-off points for the 24VDC (or the injection point if connected to the robot cabinet). Most times only one of the two connectors is populated with a cable so if a fuse blows, the unused fuse on the other connector can be used to replace the blown fuse.

EtherNet Switch Types and Rings supported?

Back within the robot controller, there is either a managed or unmanaged ethernet switch depending on the type of equipment purchased.

Managed Switch Settings required for Ring Operation:

Yaskawa provides a managed ethernet switch in the case of ArcWorld cells for the following reasons:

• A managed switch allows better network performance as more devices are networked onto a single network.
• A closed-ring network is achievable and provided on all ArcWorld cells. This provides redundant paths for signaling in the event of a broken cable.
• As PLCs and HMIs are added to ArcWorld cells they can disrupt networks if they are not carefully implemented by broadcasting large amounts of data. The managed switch will break-up those large communication so the critical Ethernet/IP devices can be serviced at regular intervals.
• Yaskawa sets-up the provided managed switch with “port mirroring” enabled so that the “service” port on AW cells can see all the communications occurring anywhere on the network. This is helpful if a “WireShark” capture is requested for diagnosis of network related issues.

Yaskawa sets up some basic functionality within the 708TX switch prior to shipment. The shipped configuration is basic in nature adding things like:

• Yaskawa administrator log-in credentials
• Setup-up “fault” indication operation
• Increasing broadcast rate limits (from a default value of 3% to 8% on all ports)
• Setting an “Admin Edge” on VLAN1 for operation with some Lincoln Electric equipment
• Details of loading the above configuration and individual settings can be located in Yaskawa’s in-house SSGW-610 instruction.

If a replacement 708TX switch is sourced by a customer, the settings contained in the above configuration provided by Yaskawa ARE NOT critical. Operation of a factory default 708TX switch should operate without issue.

Default port assignments for typical Yaskawa components:

Known issues when a Non-managed switch is configured in “closed ring” network:

If a user unintentionally connects a non-managed switch into a “closed ring” system, the performance of the system can diminish quickly. Essentially, you’ve placed an jumper cable between two ports of the unmanaged switch which then causes network collisions that will cripple the network. During this time, devices will sporadically go off-line, alarms will start to stack-up, and in some situations the programming pendant might become slow to respond and possibly go off-line itself. If the pendant is operational, you can verify the likelihood of this by looking at “SYSTEM” à “NETWORK SERVICES” in management mode.