TIA V17, CPU 1515F-2 PN, six ET 200SP stations with F-DI and F-DQ modules, all PROFIsafe. We shipped the machine last year on a line topology.
The customer had their switches changed to Scalance XC208 and asked me to configure MRP, which I did on a site visit. Since then the far station on the ring passivates two or three times a day. Their buffer says "F-monitoring time exceeded" on the F-DI, then it reintegrates after an acknowledge.
I tried swapping the IM 155-6 PN HF on that station with a spare, same thing. The customer wants MRP taken out again, "it worked before".
The F-monitoring time is whatever TIA proposed, 100 ms I think, and I never changed it. I'll own that. Fourteen devices in the ring.
Is it the ring, or the monitoring time itself, any thoughts?
Ring reconfiguration time. Standard MRP is the 200 ms class for up to 50 devices, fewer devices is faster in practice but 200 is the figure you design to. A 100 ms F-monitoring time is under that. Is a link in the ring actually going down, though? MRP only reconfigures when something breaks.
Separate the layers before touching the F-parameters. Switch logs first: is a port flapping? Then the ring: does the manager report reconfigurations at the times in the buffer? Then the safety side. If nothing on the ring changes state, the monitoring time is not your problem, the cable is.
Checked with the customer's IT. Second XC208 shows a port going down and up a few times a day, old patch lead to the far station. Ring manager logs a reconfiguration each time. Times match the passivations. So the ring does break, MRP reconfigures, and the far station misses PROFIsafe frames while it does.
Then it's both, and only one of them is the safety fix. Replace the patch lead, yes. But the F-monitoring time has to survive a ring reconfiguration or MRP buys you nothing on the safety side, the next flap does it again.
Device view, the F-module, Properties, F-parameters, F-monitoring time. Raise it above the MRP reconfiguration time with margin, 200 ms is what the 50 device ring is specified at, check TIA's own proposed minimum for the station. Do it per F-module on that station, and look at the other five, they were lucky, or just closer to the manager than the break.
Then recalculate the safety response time, the monitoring time goes straight into it, and write both numbers into the validation file. The customer's safety engineer signs, not you.
F-monitoring time raised on every F-module across all six stations, above the ring reconfiguration figure with margin, the customer's safety person set the exact value. Response time recalculated with the Siemens spreadsheet, still inside what the risk assessment needs. New collective F-signature, documented. Patch lead replaced too.
So the F-monitoring time was sitting at the TIA proposal, under the MRP reconfiguration time, and every ring flap passivated the far station. Raised, response time redone, signed.
the other five never passivated on the same ring. dieter says position in the ring, I'm not sure that's the whole story. Thanks dieter_k and hkraus.
F-monitoring time goes in the response time sheet. Every time it changes.