What Is PROFINET DCP?
Before a PROFINET IO device ever exchanges cyclic data it has to be found, named, and given an IP address, and DCP is the protocol that does all three. Any time an engineering tool scans a network and lists the reachable devices, or assigns a station name, it is speaking DCP. This page explains what DCP is, how the identify-and-set exchange works, and why it lives below IP.
PROFINET DCP in one line: PROFINET DCP (Discovery and Configuration Protocol) is a layer-2 Ethernet protocol that discovers PROFINET devices and assigns their station name and IP settings before any cyclic connection begins. It rides directly on Ethernet frames, not on IP, so it can reach a device that has no valid address yet. Engineering tools use DCP Identify to browse the network and DCP Set to write the station name and IP.
Why DCP Sits Below IP
A brand-new PROFINET device often ships with no IP address and no station name, so a normal IP-based scan cannot reach it. DCP solves this by running as a raw layer-2 protocol carried in Ethernet frames on the local segment, addressed by MAC rather than IP. That is why an engineering tool can see and name a device that has never been configured, and why DCP traffic does not cross a router - it is confined to the same broadcast domain the tool sits in.
This layering also explains a common commissioning surprise: a device is perfectly visible in the vendor tool's device list yet unreachable by ping. The tool found it with DCP over layer 2, but no IP has been assigned yet, so IP-level tools cannot talk to it. The fix is to complete the DCP set step. Because DCP is confined to the local segment, a device on the far side of a router will not appear, which shapes how you plan a commissioning connection and any OT VLAN segmentation.
The Identify and Set Exchange
DCP has two families of messages that cover everyday work. A DCP Identify request is a multicast that asks every PROFINET device on the segment to answer with its identity - MAC, station name if any, IP, device type, and role. This is what populates the device list when you browse. A tool can also target the identify at a specific station name to confirm exactly one device answers to it, which is how it detects a duplicate name.
DCP Set writes configuration to a specific device: most importantly the station name, and the IP address, subnet mask, and gateway. A DCP Set of the station name is exactly what assigns the name a controller then checks at connection time, and it is the manual half of resolving a device name mismatch after replacement. DCP also offers a flash or signal service that blinks an LED on the target device so you can physically confirm which box in the cabinet you are about to name.
Where DCP Ends and the IO Connection Begins
DCP's job finishes once a device has a name and an address. From there the IO controller establishes the cyclic application relation using the station name to identify the device, and the real-time data exchange takes over. DCP is not used during normal cyclic operation; it reappears only when you commission, rename, re-address, or diagnose. Some devices support storing the name permanently while others expect the controller to reassign it after a swap, which is the basis of automatic device replacement.
For anyone building or auditing a monitoring layer, DCP is worth understanding because it is where PROFINET identity actually lives. The station name, not the IP, is the durable identifier a controller trusts. A platform that trends controller tags never issues DCP itself, but knowing that names come from DCP explains why a mislabeled or duplicated station name shows up as a control-side fault rather than a network ping failure.
Frequently Asked Questions
Does DCP use IP addresses?
No. DCP runs directly on layer-2 Ethernet frames, addressed by MAC, which is what lets it reach a device that has no IP yet. It is used precisely to assign the IP address in the first place. Because it is not IP-based, DCP traffic stays on the local segment and does not cross a router.
Why can I see a PROFINET device in the tool but not ping it?
The engineering tool found the device with a DCP Identify request over layer 2, which works without any IP. If the device has no IP address assigned yet, IP-level tools like ping cannot reach it. Complete the DCP Set step to assign an IP, and the device becomes pingable and available for cyclic IO.
How does DCP help find the right physical device?
DCP includes a flash or signal service that blinks a status LED on the targeted device. When several identical devices sit in one cabinet, you select one in the tool and trigger the flash to confirm which physical unit it is before you assign its station name. This avoids naming the wrong node.
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.