Technology is changing along with the times. As the quality of materials and lubricants improves, so does the need to take advantage of such improvements. If we didn’t push those materials and lubricants to their new limits, then the competition would soon push some of us out of business. It’s happening all around us faster than some of us can comprehend. One problem with handling change is trying to discard the established paradigms. We have to look no further than the use and maintenance of bearings to see an example of these changes.

How many times have you heard the phrase “the bearing is going bad because it’s so hot that I can’t keep my hand on it”? Well, in today’s world there are bearings that run hot all of the time, and they are not failing. The following case study illustrates how such an entrenched paradigm created unimaginable problems.

Figure 1

A new split bearing design (Figure 1) was being considered for replacement of a spherical roller bearing in a pressure-lubricated system. There were two advantages for trying this change. The first advantage was that the new bearing could be quickly replaced in the field. This would shorten a typical bearing replacement of the inboard bearing by 16 hours, a phenomenal savings of time. The second advantage was that because the split bearing was rated for higher temperatures, it could tolerate the heat load that it would experience from the intended service. Therefore, the failure-prone and maintenance-intensive circulating oil system was no longer needed to cool the bearing. All of the angles of the proposed bearing upgrade were considered; all except one . . . the most important one . . . operator and supervisor training.

The change was performed on three units. All three units ran flawlessly through the first winter. The trouble began the first summer with the introduction of hot weather. The bearings on all three units were hot to the touch and were considered to be bad. Orders were given to replace the bearings post-haste. The bearings were replaced and immediately the operators requested that they be replaced again, complaining that the maintenance mechanics didn’t know how to install the new bearings.

The lubrication mechanic was stymied when the third set of bearings also ran hot. The mechanic swore that the correct amount of grease was in all of the bearings. Air hoses were placed on the bearings to cool them down. Production and maintenance personnel alike rallied around the bonfire to hang the engineer who thought up the stupid idea of using split bearings. But before a lynching could occur, the hot weather broke and the bearings cooled down enough to appease the operators who had bigger problems to contend with and were distracted momentarily.

The third set of bearings on all three machines made it through the winter and then began to run hot again in the summer. Once again there were accusations that the lubrication mechanic was sleeping on the job and the bearings weren’t getting enough grease.

All of these judgments were made on the basis of the old paradigm that “a hot bearing is a bad bearing.” There was no other basis for the judgment. The vibration signatures looked fine, but to some operators, vibration analysis is just another form of witchcraft. The engineer in charge of assessing the problem recovered every bearing that was removed from service. To his surprise, they were all in pristine condition. With the bearings again in the limelight, failure was not an option to the engineer.

The engineer decided to set the operating temperature limits of all three sets at a higher level. In addition, training meetings for all operators and mechanics, supervisors and foremen were scheduled to review the findings and the recommendations for higher operating temperature limits. It was touch-and-go the whole summer long trying to allay the fears of everyone. To the astonishment of production and maintenance personnel alike, the bearings performed without failure. The bearings were disassembled during a fall outage and inspected. They looked good and were reinstalled. Everyone learned a valuable lesson. An old paradigm was laid to rest, at least for a while.

Why the Heat was Generated
Consider why this bearing had a tendency to run hotter than the bearings previously used in this application. The vast majority of roller and ball bearings contact the inner and outer raceways through rolling contact. In this type of contact, as the rolling element moves through a given point on the raceway, it squeezes the lubricant. The squeezing action creates a fluid pressure wedge that keeps the two surfaces apart. However, let’s take a closer look at the split bearing shown in Figure 2.

Figure 2

The roller bearing contacts the inner race in a rolling motion, just like any other bearing. However, look closely at the sides of the roller bearing. It is not an optical illusion; the rollers nearly contact both locking collars. When the split bearing is used as a thrust bearing, the rollers carry the thrust load against the face of the split-locking collar. The contact is a circular sliding motion. This is a difficult place to establish and maintain a full fluid film. Consequently heat is generated. Why is this? Is the type of grease used in these bearings contributing to the hotter bearing temperature?

Figure 3

The principle of hydrodynamic lubrication requires a balance between surface area, speed and load to create the lifting effect that the lubricant uses to separate the moving surfaces as seen in Figure 3. In this highly localized load zone there is insufficient fluid film to accomplish hydrodynamic lift. Without a complete film between these sliding surfaces, the metal component surfaces collide, causing an elevated temperature condition.

The bearing manufacturer states four requirements for good lubrication:

1. The normal operating temperature range for these bearings is 32ºF to 212ºF.

2. The type of grease used should be an NLGI #2.

3. The amount of grease used should be significantly reduced to a “smear” at high speeds where the dn value > 200,000.

dn factor = n (d + D) / 2
n = speed
d = bore diameter (mm)
D = outside diameter (mm)

4. Extreme pressure (EP) additives should be used for high radial and axial loads. The manufacturer goes on to recommend that oil lubrication is desirable for fixed bearings with high axial loads because of the need to ensure continuous lubrication of the roller ends where the roller interferes with the collar.

All four requirements were met in this application. However, the last recommendation mentioned above by the bearing manufacturer emphasizes the problem that is discussed here. The roller ends are under high load and it is imperative to establish an oil film.

Oil vs. Grease
Why does the OEM suggest the use of oil instead of grease with high axial loading? The edge of the roller is turning and sliding at the same time, and the roller is working the grease. Some greases do not respond well to relentless working because they are not mechanically stable. These inferior greases will break down and release the oil prematurely. At this point, the grease is useless. The bearing manufacturer is keenly aware of this fact.

Should we have used oil instead of grease? We used the existing oil circulating system. The temperature was not any different. There were reliability issues with the system, so it was eliminated.

What about the use of a grease with a heavy base oil? When we tried this approach, the temperature skyrocketed toward 300ºF. At this temperature, the bearing manufacturer states that the radial loading of the bearing is reduced. This is due to a loss in material hardness of the roller. It was decided that going back and using the NLGI #2 grease made good engineering sense, and that running the bearing at 190ºF did not threaten the ultimate reliability of the bearing.

However complex, the bearing action does develop a fluid film, evidenced by the fact that the bearings continue to run. The film may not be as efficient as one developed in the normal fashion, but the design does work.

The technological development of new production systems creates economic opportunities for production facilities. Sometimes the new systems behave differently in operation than the old systems. As the plant changes, operators and maintenance personnel must be trained to view the standard operating procedures differently as well. Failure to retrain can contribute to bad judgments and wasted effort. To prevent this kind of misperception from occurring again, it would be useful to inform operators of changes and then vigorously reinforce the need to establish new paradigms.

Author Richard Adler has more than 25 years experience within the fields of maintenance and maintenance engineering. He has worked for several companies in the petrochemical, oil refining, specialty chemical and pharmaceutical industries. Photos and articles about actual failure analysis events can be viewed on his Web site: www.RESnapshot.com.

Photos © 2001 R.H. Adler