Half the arguments around a fault disappear when you show a recording. The trace function built into S7-1500 CPUs and SINAMICS drives is a multi-channel recorder sampling at cycle time — and most plants never open it.
What to record and when
For intermittent faults, configure the trace with a trigger on the fault condition and use pre-trigger recording — the trace shows what led up to the event, which is where the cause lives. Typical captures that solve real problems:
- Command, feedback and fault bits of a drive around a random trip.
- Analog level of a sensor together with valve command during a quality deviation.
- Actual cycle time and OB execution during sporadic overruns.
On SINAMICS, drive-internal traces sample far faster than the PLC sees the world — for torque ripple, encoder issues or current-loop questions, record in the drive itself via Startdrive.
Practical settings
Sampling every cycle fills memory fast; for slow processes, sample every n-th cycle and extend duration. Save recordings as measurements in the project — a library of “healthy machine” traces is priceless six months later when someone asks whether the axis always oscillated like that.
FAQ
Does tracing load the CPU? Marginally; the recording runs in firmware. On tightly-cycled machines, check the cycle-time reserve before adding channels.
Can operators trigger a trace? Yes — traces can be armed from the program, so a “record next occurrence” button on the HMI is perfectly achievable and very effective for phantom faults.
Zone Otomasyon uses trace-first diagnostics before replacing any hardware. Being asked to swap parts on suspicion? Get a recording first.