Oil analysis has been an integral part of the preventive maintenance (PM) and predictive maintenance (PdM) programs at American Axle & Manufacturing’s (AAM) Detroit Gear & Axle facility for several years. The program began when the company’s lubrication supplier, ChevronTexaco Global Lubricants, offered its services to complement AAM’s PM/PdM programs. The use of oil analysis as a reliability maintenance tool rapidly increased throughout the facility, with proactive results.
Savings resulting from the on-site oil analysis program have convinced AAM that an effective lubrication-monitoring program should be an integral part of any PM/PdM program. The data provided through oil analysis was key to improving the overall effectiveness of the PM/PdM programs.
Documentation from AAM’s oil analysis program clearly shows that the Detroit Gear & Axle (DG&A) facility saved $75,000 to $100,000 in three years on lubrication-related expenses in 102 systems, containing approximately 7,300 gallons of hydraulic fluid.
Because these 102 systems represent only 38 percent of the hydraulic systems in one of five axle-manufacturing facilities, it is easy to see that on-site oil analysis can mean big savings for AAM.
Working with ChevronTexaco to effectively utilize the information, AAM was able to:
Before on-site oil analysis was implemented, AAM’s off-site oil analysis process was relatively simple. The samples were pulled and provided to the lubrication representative. Then they were mailed to the laboratory, analyzed and mailed back to the lubrication representative. The results were interpreted by AAM’s PM personnel and any necessary action was taken.
The only issue with the program arose when a nonroutine (emergency) sample needed to be analyzed. The issue was time. The typical process for routine samples had a turnaround time of seven to 10 business days from the time a sample was pulled until the results were returned.
For nonroutine samples, the turnaround time was two and a half to three days; again, from the time the sample was pulled until the results where returned. Typically, actions were required prior to the return of the sample results if production levels were to be maintained.
AAM’s DG&A facility began to look for process improvements in the oil analysis program. The key problem was not the program’s quality, rather it was the program’s inefficiency, particularly for nonroutine sample analysis. Approximately 66 percent of the process time on nonroutine sample analysis involved shipping and handling.
Therefore, AAM decided to try on-site oil analysis. The overall cost of the equipment, which was purchased from CSI, was approximately $31,000. This initial investment included a computer, a dual port, a basic oil analyzer, a viscometer and a particle counter.
The greatest advantage of the new system was time savings. Nonroutine samples could be analyzed more efficiently. With the on-site lab, the basic analysis required approximately 10 minutes to process.
The total turnaround time for pulling the sample, analyzing and returning the results was reduced to approximately 30 minutes - a 99 percent reduction in turnaround time when compared to the off-site analysis. Furthermore, the oil analysis could be used as an integral part of determining the root cause of a problem at hand.
There are several reasons for nonroutine sampling at AAM’s (DG&A). Much like any facility that utilizes lubricants, the integrity of the lubricant is questioned from time to time. Therefore, nonroutine samples are taken for various reasons, including:
It should be noted, however, that only about one percent of the annual sample volume represents nonroutine samples. Most are routine samples taken to monitor lubricant condition in major systems.
Since the first full year of on-site sampling (1997), the number of samples processed on-site has increased each year. In 1997, AAM processed 519 samples - approximately 43 samples per month. In 2002, AAM processed approximately 114 samples per month, a 265 percent increase in sample volume since the initiation of on-site analysis.
The full details of these samples are stored in an oil analysis database. For simplicity, AAM maintains a log that documents the sample information (plant, equipment identification, system identification and lubricant) and plant reaction. The recommended plant reactions are: filter, drain and refill, or clean. Table 1 tabulates the number of samples processed and the plant reaction based on a percentage of the samples processed.
Table 1. Annual Synopsis of Lubrication Analysis at AAM DG&A
The data provided in Table 1 shows a positive trend in using oil analysis as part of the PM/PdM program. The trends are easily seen in Figure 1, which shows that the percentage of clean results continues to increase, while the percentage of drain and refill results continues to decline. In addition, the percentage of systems that require filtering has decreased.
Figure 1. Trends in AAM DG&A Oil Analysis Program
So what does this mean to AAM? It means that routine lubrication condition monitoring of equipment has allowed AAM to:
Oil Analysis Audit Results In June 1999, AAM implemented Maximo as its computerized maintenance management software (CMMS) program. Data from this system documents the effectiveness of the on-site oil analysis program. This data was analyzed in two methods due to the acquisition of data into the Maximo CMMS system after the equipment was in operation.
First, the data was used as is. That is, the number of system corrections reported in the Maximo CMMS was compared to the life of the fluid in the system. Second, the data was extrapolated to cover the entire period the equipment was operating. This calculation is a gross overstatement of the true number of system changes that occurred over the life of the system.
The decay or breakdown of a lubricant occurs in an exponential fashion, not a linear fashion as the extrapolation shows. Nevertheless, portraying the data in this manner provides a nonjudgmental perspective with maximum and minimum boundaries.
This audit consisted of hydraulic units used in the manufacturing environment. These systems were selected to illustrate the benefits of monitoring the lubrication as well as selecting the proper lubricant for the operating conditions.
For the most part, the lubricants in these systems were in place since the equipment’s initial fill. The assumption that the OEMs’ recommended change out cycle is 2,000 operating hours or six months, whichever comes first, was used in the calculations.
AAM analyzed 102 primary systems with a total lubricant capacity near 7,300 gallons. These systems contained one of two products, re-refined antiwear hydraulic fluid or Chevron Group II antiwear hydraulic fluid. Both of these products are approved under the General Motors LS2 specifications. They were selected for the individual systems based on the operating conditions.
The larger reservoirs that operate under ambient temperatures, have efficient residence time and do not have critical tolerance components, contain the re-refined antiwear 46 hydraulic fluid. The smaller systems that typically operate at temperatures greater than 120°F, have critical tolerance components and operate under more severe pressure, use the Chevron Group II antiwear hydraulic fluid.
The audited systems were in operation for an average of 29.6 months, or approximately 14,800 hours. According to OEM specifications, to maintain warranties, this term of operation would warrant approximately seven system changes.
Therefore, approximately 51,100 gallons of lubricant would have been used had an effective lubrication-monitoring program not been in place. The Maximo system recorded a total of 13 system change outs since the database was developed.
The total reservoir capacity involved with these change outs was less than 600 gallons. When the data is extrapolated over the equipment’s operation life, there would be a maximum of 42 system change outs with 1,775 gallons total reservoir capacity.
This information shows that AAM has avoided changing lubricant in approximately 700 systems thanks to lubrication analysis. When the range is calculated by selecting the OEM-specified change outs minus the extrapolated data and real data, the cost avoidance obtained through lubrication analysis, although soft, is very substantial. AAM has avoided the cost of:
For simplicity of these calculations, the product cost and labor cost to change the lubricants were investigated. Labor costs were calculated assuming that the average worker makes $18 per hour and that the change outs are typically scheduled during periods when time and a half is paid.
Furthermore, it was assumed that an average 50-gallon hydraulic reservoir change out takes two men approximately two hours to complete. Therefore, the estimated labor cost to change out a system is $2.16 per gallon. Table 2 details the cost to maintain the systems according to OEM specifications, and both actual and extrapolated data from the Maximo CMMS.
Table 2. Comparison of Costs to Maintain Systems
The volume of lubricant needed to complete change outs was based on specific system change outs.
The cost avoidance has been calculated under the assumption that the lubricant cost for the OEM-specified change out frequency was the lowest cost in Table 3, $1.00 per gallon.
Table 3. Matrix of Possible Lubricant Costs for OEM-specified Change Outs
The calculations in Table 2 clearly show how oil analysis can lower costs. However, the question of how much it costs to analyze the samples must also be addressed. This cost is broken down into two factors. First, the cost of analyzing the lubricant samples and second, the associated costs, such as the cost of generating the request, handling the samples and recording the results. A comparison of on-site and off-site analysis costs was generated and is illustrated in Table 4.
*The above calculations include the total cost of the analysis apparatus consumed in three years.
Table 4. On-site and Off-site Analysis Cost Comparison
AAM is capable of running two samples at once with a total cycle time of 10 minutes. Particle counts and viscosity analysis are completed during normal analysis cycling. Therefore, it was assumed that only one sample is analyzed at a time and that the time to complete a particle count analysis is insignificant because it can be performed simultaneously with the other analyses.
In addition, it was assumed that 50 minutes of a skilled tradesman’s time is consumed in generating the request and handling and recording the results. Therefore, the total process time per on-site sample is one hour. Additional assumptions used to define these costs were: sample processing occurs during normal working hours (straight time), the average skilled tradesman’s (chemist or laboratory technician) wage is $23 per hour, the cost of outside analysis is $9 per sample and shipping cost is $1 per sample.
According to Table 4, the cost of on-site and off-site oil analysis is relatively even. However, the values in Table 4 do not account for emergency or nonroutine samples. AAM has used the analysis equipment on several occasions to determine if a system in question needs to be drained and refilled, if it can hold out without significant damage to the system until the weekend, or if it is perfectly fine for continued use.
Therefore, the costs associated with stopping production, draining and refilling a system, new product costs, emergency deliveries, or significant damage to a system are not captured. If these costs where captured, the use of on-site oil analysis equipment would far outweigh that of off-site analysis.
These calculations clearly show the benefits of maintaining an effective lubrication- monitoring program as an integral part of a PM/PdM program. Even when considering the equipment purchased, manpower and the selected systems, AAM’s cost to operate the program over the past three years has been less than the cost of an off-site lubrication program.
Without the cost of the equipment, the minimum cost avoidance obtained through the oil analysis had nearly a 75 percent return on investment. Of course, this return on investment is captured in soft savings, but the documentation from the oil analysis program clearly shows an avoidance of $75,000 to $100,000 (Table 2) based on 102 systems containing approximately 7,300 gallons of hydraulic fluid.
Through this program, AAM has effectively increased change out frequencies on its critical systems and reduced the amount of waste oil generated in its facility. Furthermore, personnel have been able to address contamination issues head on, schedule maintenance for planned shut down periods and reduce the maintenance overhead associated with insufficient lubrication.