Don't stop with condition monitoring

So much of our emphasis in reliability organizations is placed on the application of technologies and the savings associated with finding a problem and preventing an unplanned outage or catastrophic failure. Many times, we repeat this same procedure, over and over throughout the plant, because we stop at identifying the physical problem and our resulting actions do not address the latent cause. We must become more proactive and take the next step when identifying problems with condition-monitoring technologies and determine the system or latent cause and apply the subsequent solution and/or learnings throughout the plant. The following case study involves a 450-horsepower, 1,200 rpm, 4,160-volt motor (Photo 1).

Photo 1. The huge motor in question at
the Eastman Chemical plant.

In this study, a vibration analyst noted a significant increase in vibration levels on the subject motor (Graph 1).

Photo 1. The huge motor in question at the Eastman Chemical plant.

The vibration levels had risen from less than 0.1 inches/second to 0.25 inches/second. No other changes were noted on the associated machine train. Analysis of the motor outboard bearing spectrum revealed a high amplitude peak around 7,200 cycles per minute and another significant peak around 71 times motor turning speed (Graph 2).

Graph 2. Analysis of the motor outbound bearing spectrum.

The analyst’s first suspicion was an electrically related motor problem. Therefore, he requested that the motor analysis group evaluate the suspect motor. The motor analysis group carried out a current analysis (Graph 3) and power analysis on the subject motor, and no electrical problems were identified.

Graph 3. Current analysis readings for the subject motor.

The vibration analyst then decided to proceed with more in-depth analysis. A lower-frequency, high-resolution spectrum was acquired that revealed that actual peak around two times line frequency was actually at 7,239 cpm. Further review of the motor components determined that this frequency was equivalent to the ball pass frequency, outer race (BPFO), of the inboard motor bearing. Based on these findings and the fact that we had previous problems with this application, a decision was made to replace the motor during an upcoming scheduled preventive maintenance assignment for the machine train.
Don’t Stop Here
Many times, our reliability groups want to stop at this point and claim the savings for preventing an unplanned outage or catastrophic failure. But to obtain a greater benefit from our condition-monitoring technologies, we must take the next step.
Our motor analysis team followed the subject motor to our local motor repair shop to verify the bearing problem and try to determine the causes of this problem. Upon removing the grease fill and discharge tubes, the team noted that the grease in the fill tube was not our specified grease for motors. The fill tube contained Interlube Red Hi-Lo grease, where our specified motor grease was Exxon Polyrex EM (Photo 2).

Photo 2. Analysis of fill and discharge
tubes reveals problems.

When the inboard bearing was disassembled, the team and repair shop also noted that the grease had hardened in the bearing. Further analysis of the grease contained in the discharge tube found that the bearing had been lubed with Chevron Black Pearl grease when it was previously rebuilt. The parties determined that the two greases were incompatible and resulted in the grease hardening. Further analysis of the bearings also confirmed that the outer race of the bearing was damaged.
It was also noted that we were using a spherical roller bearing in a belt drive application. A decision was made to change the bearing to a cylindrical roller bearing to increase the radial load capacity.
Don’t Stop Here Either
Many times our reliability groups are tempted to stop at this point, and we pat ourselves on the back and claim our problem solved by correcting the bearing and grease specification on the subject motor. But to obtain the greatest benefit from our condition-monitoring technologies, we must take the next step. We must identify the system or latent cause of the failure and address these causes to gain the most benefit for our company. How did the wrong greases get into this bearing? Are there other motors in this area or in the plant that are receiving non-specified greases? Why is the motor repair shop using something other than the grease we specified for our motors within the plant?
Some of the other actions that were taken as a result of the findings and subsequent root cause analysis were:

1. Findings were communicated to our Lubrication Services Group to insure that the subject grease was not being used in other applications. Although we no longer specified this grease on any plant equipment, it was determined that many of the lubricators still had an inventory of this grease in their areas. The Interlube Red Hi-Lo grease was removed from all lubrication storage areas.
   
2. The Lubrication Services Group had changed the grease specification within the plant, but this had not been communicated to the affected service providers. A meeting was held with our motor repair shop to communicate our findings, the result of mixing incompatible greases and our expectations for the future.
   
3. A new repair specification was developed as a result of this and several other significant motor failures. The grease specification was included in the repair specification. The previous repair specification did not include a grease specification.
   
4. As a result of these findings and findings on several other motors, a decision was made to develop a motor repair evaluation process and team.
   

Conclusion
This case illustrates the extra benefits of taking your condition-monitoring program to the next level. By becoming more proactive and taking the next step when identifying problems with condition- monitoring technologies, you can determine the system or latent cause and apply the subsequent solution and/or learnings throughout the plant.