How WirelessHART Mesh Routing Works
A WirelessHART field device rarely talks straight to the gateway - its message usually hops through several neighbors first. How the network decides those hops, and why it keeps more than one path ready, is what separates a mesh that survives a blocked line of sight from one that drops readings. This page explains graph and source routing, why redundant paths exist, and what routing does when a node disappears. It is for engineers who want a mental model of how data actually travels across the mesh.
WirelessHART Mesh Routing in one line: WirelessHART routing moves each message hop by hop across the mesh to the gateway using paths the network manager precomputes. Graph routing gives a device several alternative next hops toward the gateway for redundancy, while source routing pins one specific path for diagnostics. Because the manager provisions more than one route, a message can still reach the gateway when a single node or link fails.
Graph Routing and Source Routing
In a WirelessHART mesh, most field devices cannot reach the gateway directly, so their messages are relayed through intermediate devices that also act as routers. The routes are not discovered on the fly by each device; they are computed centrally by the network manager, which knows the whole connectivity graph and installs routing information into the devices. Two routing methods coexist, and knowing the difference explains most of what you see in a network diagnostic view.
Graph routing is the normal, resilient method. The manager defines a graph - a named set of paths toward the gateway - and gives each device more than one possible next hop that belongs to that graph. When a device has a message to send, it can forward it to any of its listed next hops, so if one neighbor is momentarily unreachable, the device simply uses another. This built-in choice is the redundancy that lets the mesh route around a blocked or weak link without waiting for a full reconfiguration.
Source routing pins a single, explicit path from device to gateway, listing every hop in order. It is used mainly for diagnostics and commissioning, because it lets an engineer force a message down a known path to test connectivity along it, rather than letting the graph choose. Source routing carries no redundancy - if any hop on the fixed path fails, the message fails - so it is a testing and troubleshooting tool, not the mode you rely on for routine reporting. Routine data uses graph routing precisely because it can adapt.
Redundancy and Surviving a Lost Node
The reason a WirelessHART mesh keeps delivering data when a node drops is that graph routing gave the surrounding devices alternatives before the failure happened. Suppose a device normally forwards through a neighbor that suddenly loses power. Because the device's graph lists other acceptable next hops, it forwards through one of them on its next attempt, and the reading still reaches the gateway. No central decision is needed for this immediate reroute; the alternatives were already provisioned.
Over a longer horizon the network manager reacts too. As devices report degraded links or a node stops responding entirely, the manager updates the graphs and the communication schedule so the mesh settles into a new stable configuration that no longer depends on the failed element. This is the deeper layer of the self-healing behavior: the graph provides instant local resilience, and the manager provides the slower, network-wide reoptimization. The page on the WirelessHART network manager covers that central reconfiguration in detail.
For design, this routing model translates into a simple guideline: give the mesh enough neighbors. A device with only one viable path to the gateway has no graph redundancy, so a single obstruction takes it offline until the manager can find another route, if one exists at all. Ensuring each device can hear several others - through placement, density, or adding a routing-capable node - is what gives graph routing the choices it needs. The broader picture of adding points without wire is covered in the WirelessHART overview, and the mesh-topology page explains why density matters for any OT mesh.
Frequently Asked Questions
What is the difference between graph routing and source routing in WirelessHART?
Graph routing gives a device several acceptable next hops toward the gateway, so it can route around a failed link automatically - this is the resilient mode used for routine data. Source routing pins one explicit path with every hop listed and carries no redundancy, so it is used mainly for diagnostics and commissioning to test a known path rather than for normal reporting.
How does WirelessHART keep working when a node fails?
Graph routing already provided each device with alternative next hops, so when a neighbor drops, the device forwards through another one and the message still reaches the gateway with no central decision needed. Over a longer horizon the network manager updates the graphs and schedule so the mesh settles into a new configuration that no longer depends on the failed node.
Why does WirelessHART need multiple paths per device?
Because a single path has no redundancy: one obstruction, a parked vehicle, a dead node, and the device goes offline until a new route is found. Provisioning several next hops through graph routing gives the mesh choices, so it can reroute instantly. In design terms, ensuring each device can hear several neighbors is what gives graph routing the alternatives it relies on.
Automation services
Need help turning this into a working system?
Merobix integrates SCADA, programs Allen-Bradley and Siemens PLCs, and designs and fabricates industrial control panels.
Meeting requests are reviewed before confirmation.