Emergency trip avoids consequential damage

Crosshead vibration initiated immediate shutdown of the machine preventing consequential damage

This case study illustrates the advantage of utilizing segmented analysis (8 segment peak to peak) rod position as part of a multi-layer protection strategy together with crosshead acceleration for reciprocating machinery.

Sequence of Events

  • Compressor commissioned in 2011
  • On Aug. 15,2016 at 5:47:04 am the system was tripped on high 1st stage crosshead vibration after 3 month of running on steady state at 100% load
  • Analysis of crosshead vibration data shown no increase
  • Analysis of peak-peak rod position data shows a slight increase in Seg. 2 several minutes prior to failure, but protection limit of 36 mils was not violated
  • Machine was restarted by operations risking further damage

Several Trend Views for Analysis

See how the signal developed during the month, days and hours before the event and machine shutdown occurs.

3D Crosshead Trend (4 month)
3D Trend Peak-Peak Rod Position (4 month)
2D Trend Peak-Peak Rod Position (all segments 4 month)
Trend Rod Position for Segment 2 (4 days) – 1st Upper Threshold = 18 mils 2nd Upper Threshold = 24 mils Protection Limit = 36 mils
Peak to Peak Rod Position Trend (4 days)
Revolution Prior to Break
Next Rev. – Rod Breaks in Tension

Key Learnings

  • The development of this type of rod fracture did not show significant signs of impending failure using x-head vibration
  • Crosshead vibration initiated immediate shutdown of the machine preventing consequential damage
  • Segmented peak to peak piston rod position analysis did show an increase in rod movement leading up to the event, however similar levels had been seen in past trend data
  • Data review confirmed that for the type of failure experienced, the machine protection parameters were properly configured requiring no adjustment

1 Comment

  1. Good afternoon, thanks for the case study. After looking at this it is unclear why the protection setting is not below the point in which the rod breaks. Surely in this case would rod displacement in future above 25 mils not now indicate a fracture has occurred. Subsequently a trip at this point would have the ability to stop the machine before the disconnected rod crashes into a free floating broken rod & piston assembly?

Leave a Reply

Your email address will not be published. Required fields are marked *

1 Comment
scroll to top