SIG200 PN loses IP address after a power cycle

Why is the SIG200 Profinet not available via the IP address after it is disconnected from the PLC. 

The "problem" here is the DCP protocol of PROFINET. 

What is DCP?

PROFINET DCP is part of the protocol suite and stands for “Discovery and basic Configuration Protocol”.  It is used by the engineering tool and controller to discover devices, identify device information, and configure device settings such as PROFINET device name and IP address on a PROFINET network. PROFINET DCP is an Ethernet link layer protocol and offers multiple services.  It is normally used in PROFINET systems to handle the network address and name management as each PROFINET device is assigned a unique device name (ex: “sig200”) based on Domain Name System (DNS) naming conventions and an IP address (ex: 192.168.1.2).

What happened so that the SIG200 lost the IP address?

First the Controller /PLC sends an Identifier protocol to find the projected devices. 

DCP Identify
The ‘Identify’ service is used when a device needs to be found using a particular/known device name.  This is typically used by the PROFINET controller at startup to identify each device and check its IP address settings, name and if it has the expected device before parameterization.  Another use might be in the engineering tool; for example, let’s assume that we already have setup a device with the meaningful name “sig200”.  You can then check if the name has been set on the device by using a check name service in the engineering tool (or by using the previous function: ‘Identify All’).  In this case, the tool is using a DCP ‘Identify’ service basically to say “Device sig200 are you out there and do you have the IP address 192.168.1.2?”  If the device is there and has the name and IP address, the check is successful as the device responds.  If not, the controller/PLC set the name and or the IP address with the PROFINET DCP ‘Set’ service.

DCP Set
The ‘Set’ service is used to set the name or IP of the device.  It also has some other special functions we’ll mention next such as resetting a device to factory settings and flash of device LEDs.

In the engineering tool the typical way of setting up a PROFINET device initially is to configure it using the manufacturer supplied GSD file, then setting the parameters and device name offline.  Once this is done the device name must be written to the device using a naming function in the engineering tool.  When the tool writes the name it uses a DCP ‘Set’ command.

A DCP ‘Set’ can be permanent or temporary.  A permanent set (retentive, the default) means that the name is stored on the device permanently in memory (even across a power cycle); temporary means the name is used until a power cycle and then returned to the default value (IP: 0.0.0.0, no IP address set).

Normally when a controller starts up a device, it attempts to find the device by its configured device name using a DCP ‘Identify’, then the controller checks the configured IP address which the engineer setup in the PROFINET projects. If the IP address is not set or wrong, the controller will write the IP address to the device using a DCP ‘Set’ command.  If the controller finds that a different device / or different node already has the IP address it cannot set a duplicate.  In that case the user would either need to change the device IP in the engineering configuration or on the conflicting node.  The IP can also be set permanent or temporary.  In the case of temporary after a power cycle the IP is typically returned to zero settings (0.0.0.0), if retained the IP address is kept.

That means that the PROFINET name was identical to the hardware config but not the IP address, so the controller/PLC does not change the PROFINET name but sets the IP address which is projected in the hardware config temporarily and after a power cycle the IP address is 0.0.0.0