Device networking 101: Your First Installation
Originally appeared in
Control Engineering Magazine, May 2001, (C) 2001 Cahners.
Somewhere in the world each day, an engineer decides to use a device level network instead of point-to-point wiring. A survey taken at a DeviceNet workshop in Chicago showed the #1 reason for investigating networks was: "They’re sexy." True, we all like to blaze new trails, but in the pragmatic world of manufacturing, there must be a sound justification.
Q Why are you doing this?
- Modularity–aids disassembly and re-assembly of the system; and allows easy replacement of failed components;
- Wiring simplification–reduces system assembly time and mistakes;
- Open architecture–maximizes choices and keeps prices competitive;
- Preventative maintenance–so devicescan warn users before something goes wrong;
- Configurability–allows parameters to be set "on the fly," which adds functionality and saves time;
- More information–networked device may provide additional process data, which may also be important for sending data to the enterprise; and finally
- Cost savings–on huge installations, you will save money. However, most benefits are realized over time, not with the first installation. Subsequent advantages will prove much more valuable than shaving pennies up front.
Q Topology requirements?
A The dimensions and physical layout of your system make a big difference in which network you choose. Primary topologies include:
- Trunk with drops–DeviceNet, AS-i, CANopen, ControlNet, Interbus, FOUNDATION Fieldbus H1;
- Linear "daisy chain"-Profibus; and
- Star–Ethernet 10Base-T, 100Base-T
Is your system a long conveyor, a compact machine, or does it cover an entire shop floor? Physically draw out your system and see how nodes will be clustered together.
Q Speed, data requirements?
A Does your I/O need to be scanned every 2 msec? Or is 500 msec okay? Do your devices send a few bits of data, or hundreds of bytes? Are you mixing simple and complex devices on the same network? To answer these questions, determine the "hard limits" and worst-case requirements of your design.
Which networks are available on specific devices you’ve chosen? This issue may force your hand, or at least limit your choices, if you are required to use a specific brand of PLC or other component.
Q How will you distribute power?
A Some networks deliver 24 V or loop current to devices, while others do not. This can be a convenience and/or a complication that you should consider.
Q Which network to use?
A After you’ve answered all of the previous questions, you’re ready to choose a network. Secure some reference information.
Q Using multiple networks?
A Some OEMs must ship machines with any one of two to four different networks, especially if they have international customers. Careful planning can make this much easier. Some more questions to answer include:
- Can you find devices that use the same mounting holes across multiple networks?
- Can you use the same programs and software, instead of rewriting it each time you switch networks?
- Can you use common configuration tools?
- Can you use the same or similar connectors and wiring trays?
Q What about connectors?
A Most networks support a variety of connector schemes, from terminal blocks to waterproof connectors. Sometimes the expense of connectors is offset by convenience and reduced wiring problems.
If this is your first network-based system, start with a small project and grow from there. If it’s a large system, engage a systems integrator who has done this before. Control Engineering May 2001