If you ever have dealt with small steam turbines, chances are that issues of bearing selection and lubrication had to be addressed. Among these are issues of rolling element bearings vs. sleeve bearings, types of lubrication and application of lubricants. Even the definition of a turbine's suitability for a certain service is often debated and using such terms as "General Purpose" and "Special Purpose" begs definition.
While a common differentiation of general purpose (GP) and special purpose (SP) steam turbines is indeed made at 1 MW, process industry users frequently have deviated from this rather arbitrary rule. Reliability professionals often have specified general purpose turbines if there were some spare capacity or if there were redundancy in the form of installed spares, or nonessential services, or moderate speed and so forth. A good example is a plant where there were several 2 or 3 MW GP steam turbines driving cooling water pump sets. Here, it would be very uneconomical—even wasteful—to insist on SP turbines.
Examples in the opposite direction exist as well. Take, for instance, a 600 kW steam turbine selected to drive a small, but essential, process gas compressor. In this situation, a special purpose steam turbine would be specified for very good reasons. It will cost more than the GP version, but it will have a lubrication system that makes it less prone to cause unscheduled outages.
Keep hybrids in mind
Hybrids are a composite of GP practice and SP practice. Packaged systems may have to be upgraded by identifying the (potential) weak link(s) and specifying certain elements that are otherwise primarily associated with GP or, conversely, SP equipment.
Competent engineers frequently will specify hybrid support systems, such as lube oil consoles (Fig. 1), electronic governors, oil mist lubrication (Fig. 2) or oil flinger disks, instead of potentially unstable and risk-inducing slinger rings, etc. Even the rather customary carbon seal rings that have been supplied for seven or more decades with GP steam turbines are due for an upgrade. Forwardlooking users are now often replacing carbon rings with high-temperature mechanical seals and, in doing so, reduce both operating and maintenance expenditures.
The best judgment of competent reliability engineers uses lots of experience and sound life-cycle cost assessments. The practices endorsed by these engineers often lead established industry practices and written standards by a decade.
Sound judgment recognizes that the life cycle of a machine is inevitably influenced by operating and maintenance practice. These practices have been the subject of thousands of written pages in books and articles. Moreover, they vary from location to location and are difficult to generalize across the board. While the practitioners may be disinclined to share this information, we nevertheless can observe what true reliability professionals do: They look at the specifics of each case and carefully weigh the alternatives. They will then submit their findings to responsible management in writing.
Lubrication frequency and oil analysis
Questions on re-lubrication frequencies and how these are to be determined recently have been explored by some readers, as have issues relating to fitness for service or extending the life of existing oils. Of course, the major lube suppliers provide products and services—either directly or through contract distribution channels. They also offer lube oil analysis as part of the supply contract.
Some plant engineers seem to have engaged the services of major and minor suppliers without regrets, and they have pointed out reductions in both lube amounts and the number of different oil types being inventoried. One of our readers noted that his company is now using a single or a couple of oil types and maybe just one grease. "Yet," he went on to write, "I'm not so sure this consolidation is the best way, as you could end up being held to ransom, and maybe not get the best properties of oil for expensive equipment. I would be interested to hear your opinion on this subject."
The same reader had looked at oil sampling valves and wondered which was best; he also noted copper greases were not to be used at his olefins site. He found two products that enjoyed name recognition and attempted to narrow it down to the best one. In terms of oil analysis sampling points, he opined that primary and secondary locations might be justified in some cases. Likening it to his analogous experience in vibration monitoring and analysis, he determined there was no need for "overkill of data" or complexity of the sampling system. "Data overload and complexity," he stated, "might tend to become problematic and maintenance-intensive."
Our reader pointed out, just as is the case with vibration analysis, that there is not the same need for detailed knowledge in all foreseeable instances. There are times when identifying the defective bearing component may be less important than simply knowing that the bearing should be replaced and determining the optimum time to effect the change-out. He indicated that component identification will obviously aid in finding certain root causes, and reliability engineers should strive to have a balanced approach.
Because the reader looked forward to our supplementary comments and thought his letter (and our answer) would be of interest to others, we are pleased to share this information.
Certainly, the bulk of this reader's questions have been answered in many currently available books and articles. Without taking the time to go into more detail, the use of a single type of grease for all applications in a modern olefins plant will lead to two alternatives. To maintain reliability and on-stream time would defi- nitely require more frequent preventive maintenance. If this diligent maintenance effort is not expended, the plant will give up a measure of reliability. If, then, an equipment user in such a facility is willing to sacrifice equipment reliability and equipment run length, he might opt to use a single grease type. However, the one advocating a single type of grease should be asked about ultimate life-cycle cost of the machines so lubricated. If the advocate were to take the time to study life-cycle cost, he might find some surprise answers. If he does not take the time to study the matter, he's just guessing and is thereby putting his plant at risk.
A similar statement could be made if an olefins plant had been persuaded to limit its oil types to just one or two. The probability of such a facility gaining Best-of-Class status is so close to zero that it is certainly not worth debating. Nevertheless, some consolidation of lubricants is feasible, as long as one looks at the lube requirements item-by-item and machine-by-machine. Sweeping generalizations are just that, i.e. sweeping generalizations are rarely adding value in a reliability-focused process plant environment. We are not familiar with the preferred brand of oil sampling valves. We are, however, reasonably familiar with some providers of oil analysis. Some folks advocate this predictive approach in situations where it makes absolutely no economic sense. The times and places and frequencies where sampling and analysis make sense are again discussed in many books and merit closer study. Having a preferred lubricant provider often is feasible and sometimes even beneficial. Unfortunately, there are also instances where this provider tries to sell the user too much stuff, or the provider's representative is inexperienced or neglects a user's account because he prefers to expend most of his energy finding new customers. There also have been instances where the lube provider does not offer the most suitable oil and then makes the user-buyer his test bed for trial-and-error solutions.
In summary, there will never be a substitute for the user educating himself/ herself on these issues. Whenever some users think training is expensive, we often ask them to calculate the expense with no training! Whenever they tell us they have no budget to purchase books, we leave them to do what they've become accustomed to accept as their normal routine: repair, repeat the repair, and repeat the repair again…