When I posted last week on WirelessHART reliability, I had a chance to speak to Emerson’s Russ Muller who is a senior PlantWeb specialist. As we discussed the reliability figures, Russ mentioned that the sites that applied the best practices saw reliability figures much higher than 99%. If you’ve seen the PlantWeb University Wireless course, in the Wireless 203- Self Organizing Networks section, it shows this figure:
Even in this extreme example of low reliability links, by designing multiple paths for each device, a self-organizing network can reach that level of performance by constantly choosing the path that offers the best reliability at the time. With self-organizing networks, it’s important to note that site surveys are not required. Russ shared with me some best practices learned from the early wireless installations which I’ll pass along to you.
The first consideration is the size of your facility. If you have a large facility like a refinery or chemical plant, the wireless field network should be scoped to a single process unit. For vertically arranged facilities like power plants or some pharmaceutical sites, the self-organizing network should be scoped to a single floor.
Next, it is extremely helpful to have a scaled drawing of the single process unit or floor where the network will be installed. In an earlier post, I discussed the creative use of Google Earth to zoom in on an outdoor facility where they didn’t have scaled drawings handy. These building drawings are typically available for inside facilities, which is a good thing since the satellite photos can’t see inside a building… yet!
With the scaled drawing, plot the location of wireless devices. Consider the immediate ones you want to install as well as possible future ones. Every wireless device should have multiple neighbors to provide path redundancy for higher overall communications reliability. Based on the experience gathered from hundreds of installations to date, each wireless self-organizing network should be designed with a minimum of five wireless devices to provide this path diversity.
As you look at the devices plotted on your scaled drawing, it’s ideal that each device have three neighbors as potential paths of communication.
Next, consider the placement of the wireless gateway. In small networks, the smart wireless gateway should be located in the center of the network. For larger networks or installations that require the wireless gateway mounted in a control or rack room, you should build the self-organizing network around the location of the wireless gateway, closest ones first, per your plot plan. Also, remember that the gateway needs to connect the network to your host automation or asset management system using common industry communications standards like OPC, MODBUS and MODBUS TCP.
The wireless gateway should have a direct wireless connection (connected without a hop through another device) to 25% of the devices in the self-organizing network. It will still be reliable if less than 25%, but greater than 25% is optimal. You can add wireless devices or repeaters to help achieve this best practice.
During installation, add devices outward from the gateway to reach other areas in the process unit. This installation process helps you see the devices as they are being added and helps verify the robustness of the communications.
I hope sharing these best practices in addition to the PlantWeb University Wireless courses provides you the background to try a wireless field network application in your facility.