
Case study
Early warning on failing cardan shaft coupling in runout table roller
Read how SAM4 caught a failing cardan shaft coupling in a runout table roll for this steel manufacturer.
Failure mode
Looseness caused by a fatigued cardan shaft coupling.
How SAM4 helped
SAM4 flagged a sudden spike in the first current interharmonic at all the roller’s operating points. Several events with the same signature had been previously diagnosed on this and other rollers at this customer, so we were fairly certain we were looking at a failing coupling. We notified the customer and put the asset on red alert. One week later the customer performed maintenance, but the interharmonic values remained high. The customer explained that they hadn’t had time to replace the coupling during the planned window. Two days later they did replace it, and the roller’s electrical signature returned to normal.
Outcome
“SAM4 has caught every failure in advance on these specific couplings, and never flagged one that wasn’t there. We view Samotics’ notifications as a trusted source and act accordingly.”
Values over time for the first current interharmonic indicator. Each color represents a different operating point. The graph shows a spike in intensity at all of them (red arrow). The two black arrows indicate previous coupling faults in this same roller, both of which SAM4 detected in advance of failure.
A heat map showing the intensity over time across the current frequency spectrum (blue is low, red is high). Time runs down the vertical axis from oldest to newest. The arrows mark the three coupling detections. All three show a clear rise in the noise floor (interharmonics).
See SAM4 in action
Related case studies

Early warning of failing heated godet roll prevents $90k in production loss
SAM4 picked up motor overloading on the godet roll early enough for maintenance to be planned, before the fault could stop the production line.

How ArcelorMittal prevented 31 hours of downtime by detecting 27 failures ahead of time
SAM4 detected each fault far enough ahead that the repair could be scheduled into planned maintenance, rather than stopping production without warning.

9x common faults detected in steel manufacturing
Nine detections drawn from anonymised SAM4 data across steel plants, each shown with the signal that triggered it so engineers can judge the evidence.
How zinc smelter Nyrstar got 800% ROI in 11 months
Continuous monitoring on the smelter's critical drivetrains caught faults early enough to plan repairs, avoiding more than €1 million in lost production.

Condition monitoring for shot blasting machines: a case study
Read how SAM4 caught a loose blast wheel belt guard before it caused damage to this steel manufacturer's shot blasting machine.
