Machinery Lubrication magazine published by Noria Corporation
Issue link: http://www.e-digitaleditions.com/i/544388
crackle test can identify water at around 500 ppm and higher, although a good lab technician can sometimes catch it at a lower ppm in certain oils with low additive levels (e.g., turbine oil). The crackle test is approx- imate and not considered quantitative. If you need to know the exact amount of water, have a Karl Fischer titration performed, but under- stand that there can be interferences with this as well. Also, be sure to specify the evaporator or drying method, especially with engine oils (also called co-distillation). Particle counts are usually obtained with a laser particle counter. While this is an accurate method, darker oils, water and air entrain- ment can sometimes cause problems. The pore blockage particle count can be performed in place of the laser particle count when the oil condition is extremely dark or contaminated with water. This method is not as accurate as a laser particle count, but it is a viable option in assessing fl uid cleanliness to avoid the previously mentioned interferences. Another less common way to perform a particle count is with a Millipore patch. This can be time-consuming and expensive, but in specifi c situations, the Millipore patch can be used to help with wear debris analysis in place of analytical ferrography. Special Tests Even with an extensive oil analysis test slate, special tests are often needed to provide additional information when extending oil drains. For example, a rotating pressure vessel oxidation test (RPVOT) can be conducted to determine the oxidation stability of the in-service lubricant compared to a new lubri- cant. The RULER test offers another way of comparing new oil with in-service oil to estimate the remaining useful life. Other tests, such as the varnish potential rating, membrane patch colorimetry (MPC) and quantitative spectro analysis, can help identify the amount of oil degradation byproducts and depleted additives. These soft contaminants can create serious problems if left unattended. A demulsibility test is used to indicate a lubricant's ability to shed water. Air release or foam tests are also important, as air does not provide a proper oil fi lm to keep machine surfaces separated. Rust and copper strip corrosion tests can help identify a lubricant's remaining anti-corrosive additives, while acid and base number tests measure the rate of change between a new and used lubri- cant. See Figure 1 for a list of the ASTM test methods. These are just some of the special tests that can be performed on in-service lubricants to help make decisions about extending drain intervals. Once again, having good communi- cation with your lab will be invaluable when it comes to making sure you are getting all the information needed to achieve cost avoidance and prevent catastrophic failures. Cost Avoidance With lubricants, there is the actual cost (what you pay the distrib- utor per gallon, pound, pail or drum) and the real cost. The real cost of a lubricant is the total cradle-to-grave cost once it reaches your facility. This includes receiving, storing, dispensing, installing and disposal. Several years ago this cost was estimated to be an average of four to seven times per gallon. For contaminated special case oil (radioactive), the cost could be as much as 40 times per gallon. The average industry cost per gallon is $9 to $14 for mineral oil and $20 to $30 for synthetic lubricant, with the exception of specialty synthetics being $60 and more. For the purpose of this discussion, let's estimate mineral oils at $10 per gallon and synthetics at $25 per gallon. At the Seminole Electric facility, which is a two-unit, 1,300-mega- watt combined coal-fi red power plant, the equipment monitored for condition-based oil changes contains 6,043 gallons of oil (see Figure 2). Prior to condition-based oil changes, some of this equipment received oil changes every six months and some every 12 months. With these oil changes, the total increased to 7,948 gallons of lubricant per year. If you were to use the average of $10 per gallon and the real cost of seven times per gallon, the total would be a real cost of $70 per gallon. In a perfect world, if you could go one year without an oil change on all of the equipment, you would have a cost avoidance of $556,360. As can be seen in Figure 2, the average meantime between oil changes at the Seminole Electric facility is now closer to 1.5 years, with some equipment reaching fi ve years. On the fi ve-year oil change equip- ment, the total gallons of oil is 18,145, based on the six- and 12-month change frequencies over fi ve years. Four to seven oil changes are avoided, since the equipment is usually inspected with an oil change at the fi fth year of operation or around 40,000 runtime hours. This equals a cost avoidance of $1.27 million for those six groups of equipment over the fi ve-year period. Please note that turbines, turbine control oil and boiler feed pumps are not included in these numbers. Due to their size, these reservoirs, which are more than 25,000 gallons combined, and fi ltration systems generally receive extended drain intervals. Return on Investment Now let's consider the return on investment (ROI) for condi- tion-based oil changes. The primar y cost is the oil analysis. In Figure 2, there are 792 oil samplings per year for the equipment listed, at an average cost of $20 to $40 per sample analysis. A range of $15,840 at $20 per sample to $31,680 at $40 per sample provides an idea of the cost for the oil analysis. While $40 might be slightly high, that should more than make up for the few special tests needed to obtain additional information throughout the year. Even if you include the salar y of the person managing the oil analysis program at the facility, you would still be well below the projected cost avoidance of $556,360 per year. 16 July-August 2015 | www.machinerylubrication.com