A Good Case For Improved Oil Filtration

Learn how a well-designed filtration-improvement program is helping specialty-metals producer ATI Wah Chang realize significant savings through enhanced equipment reliability.

By Ray Thibault, CLS, OMA I, OMA II, MLT, MLT II, MLA II, MLA III, Contributing Editor

ATI Wah Chang, a division of Allegheny Steel, produces reactive and refractory specialty metals used in a variety of unique high-performance applications for engineered products and material solutions. The operation was the first to process zirconium and produce a number of other specialty metals such as hafnium, titanium, niobium, tantalum and vanadium. It is a highly innovative company in a highly specialized field.

In 2011, faced with significant vacuum pump failures attributed mainly to contaminated oil, ATI Wah Chang’s maintenance department embarked on a proactive lubrication program to enhance fluid cleanliness through improved filtration. This program was spearheaded by Dale Jones, who has extensive lubrication experience and currently holds six lubrication certifications. Jones worked with his lubricant marketer Moreland Oil and, through Moreland, with Hy-Pro Filtration, in an effort to develop a filtration improvement program. The result was a dramatic improvement in fluid cleanliness that significantly enhanced bottom-line savings for the company.

This case history focuses mainly on improvements made to vacuum-pump reliability and the operation’s Z-Mill rolling oil. They have led to enhanced product quality, dramatic equipment life-cycle extensions and increased productivity.  

A quick review of basic filtration principles

It’s been shown that approximately 70% of premature equipment failures are caused by particulate contamination, and over two-thirds of equipment wear is from abrasive particles. Since 70% of premature failures are caused by contamination, it stands to reason that controlling particulate contamination through better exclusion and filtration would provide a high return on investment.

Although the basic principles of fluid cleanliness have been discussed in previous articles in this and other series in LMT, let’s review some of them here:

Fluid cleanliness is measured primarily with the use of laser particle counters. Particles are measured in many size ranges but are expressed in a simplified system ISO 4406. This system expresses cleanliness in three size ranges: > 4µm(c), > 6µm(c) and > 14µm(c). Table I is used to arrive at the ISO 4406 Cleanliness Code.

Notice that for each increase in range number, the number of particles doubles. This method is a convenient shorthand way of assessing fluid cleanliness without worrying about the actual number of particles. As an example, assume you measure 1042 particles/ml > 4µ. 412 particles/ml > 6µ and 152 particles/ml > 14µ. Referring to Table 1, the range number for particles > 4µ is expressed as 17. The number of particles > 6µ is 462/ml, which translates to a range number of 16. Finally, the number of particles > 14 µ is 152/ml, which is expressed as a range number of 14. Putting this three-number code together results in expressing the above fluid cleanliness as 17/16/14. Remember, the first number is always greater than the second, which is greater than the third.

Filter performance should be based on the absolute filter rating, which is determined in a laboratory with the Multi-Pass Filter Performance Test. Dirt in milligrams per liter is introduced in a test fluid at a constant rate. The fluid is circulated at a constant or variable rate through a test filter. The test concludes when the terminal pressure drop of the filter is reached. This is the point when the filter manufacturer designates the filter as no longer operable and needs to be changed. The filtration efficiency of the test filter is expressed as the filtration ratio here:

For example, a filter with a Beta Ratio of β₆ = 200 indicates that for every 200 particles greater than six microns in size entering the test filter, one will pass though. Therefore, 199 will be captured. The efficiency of the filter is calculated as β – 1/β x 100. The efficiency of a filter with a Beta Ratio of 200 is 99.5%. The standard today for calling a filter absolute is a minimum Beta Ratio of 200, and some filter companies express their filters with a Beta Ratio of 1000, which eventually will be the standard for absolute filters.

Vacuum pump improvements
At ATI Wah Chang, a large number of Stokes 412 vacuum pumps (similar to the one shown in Fig. 1) are utilized to create a non-oxygen atmosphere by producing a high-vacuum environment in a vast network of furnaces during a chemical reactive process. The pumps operate in a highly hostile environment requiring frequent oil changes. The volume of oil consumed is so great that an oil reclamation unit has been established to recondition the oil. Bath lubrication of the pumps with an R&O ISO VG 150 lubricant is used in this process. Major pump failures occur due to wear of the piston slide plate; as the tolerances between the piston and slide plate increase, the integrity of the piston’s sealing surface is jeopardized, resulting in loss of vacuum-producing efficiency.

It was established that the major cause of wear was from abrasive particles that accumulated in the oil during the process. The cost to rebuild each pump is $22,500 (calculated on the basis of 225 man hours at $100/hr). Before the change in the company’s filtration program, an average of 24 pump rebuilds—costing a total of $540,000—had to be done each year. The reclaimed oil process is illustrated in Fig. 2.

The used oil is collected and sent to the oil reclamation unit for reconditioning. The initial step is to use a settling tank to remove the free water and heavy solids. The next step is to use course filtration through a β₂₅ = 200 filter. The oil is then heated to 160 F and passed through a vacuum dehydrator to remove much of the remaining emulsified water. After the oil is vacuum-dehydrated, it’s sent through clay filtration to remove the acidic contaminants generated during the process.

The final step—and one of the most crucial—is fine filtration. The major change in the process was to upgrade filtration in that final stage: The original filtration was a β₃ = 75 and was upgraded to a β_2.5 = 1000 filtration system from Hy-Pro. Figure 3 illustrates the results from the improved filtration.

Fig. 3. Examples of vacuum pump oil before and after implementation of filtration upgrades in the final phase of the site’s oil-reclamation process

The total savings from the upgrade is $382,500/year, with a minimal increase in filter cost. The most recent information from the plant notes a further drop in the need for pump rebuilds—down to five or six per year.

Z-Mill rolling process improvements
ATI Wah Chang’s rolling mill produces very thin metal sheets—to under 0.001 inches thick. The rolling mechanism is enclosed and flooded with 60 cSt oil during the process. Due to the extreme amount of force applied to the work piece during the rolling process, the oil produces an elastohydrodynamic lubricating film between the work rolls and the product. Because an elastohydrodynamic lubricating film is very thin (estimated at less than 1 micron), virtually all of the dirt that is suspended in the oil is larger than the lubricating film separating the work rolls and the product. The dirt in the oil makes contact with the work piece and work rolls, having a significant effect on the surface quality of the product and the longevity of the rolling mechanism.  The original filtration system was evaluated by draining, cleaning and filling the reservoir with new oil and installing new original filters. After several months of normal operation, a baseline fluid cleanliness was established. A study was then conducted with the following steps:

• Research new filter options
• Test and evaluate options
• Determine if an improvement opportunity exists
• Initiate the improvement and monitor the results

The original filtration system utilized a β₃ = 75 filter, producing an average ISO cleanliness code of 23/20/12. The filtration system was modified using a custom-built Hy-Pro filter, with a specification of  β_2.5 = 1000. The ISO cleanliness of the fluid improved to an average of 13/10/4. After two months of operation with the improved system, the department engineer identified a 15% product yield increase due to surface-quality improvements. The resulting yield increase was valued at $100,000/year.

Additional savings were realized in roller-bearing cost. Historically, bearings were replaced every six months at an average cost of $70,000. Referring to Table II on bearing-life extension, it can be concluded that over a fivefold bearing life extension may be realized with the increased cleanliness. The following cost-savings estimate relates to bearing life extension and a conservative threefold life-cycle increase was used for the sake of generating credible estimates. The estimated savings was $93,333/year. After nearly three years of operation since the improvement, only one bearing has failed. The actual results are a realized sixfold increase in bearing life.

Actual savings from this activity are in excess of $200,000 annually. It is interesting to note that the increase in filter cost is just $1700/year. Put another way: an annual investment of $1700 for the purchase of high-quality filtration has a return of over $200,000—without the plant having to turn out any extra product. 

Coupled with the oil filtration improvements associated with its vacuum pumps, ATI Wah Chang’s fluid-cleanliness improvement in its Z-Mill operations means the company has captured approximately $600,000/yr in savings—from just two assets.

Other filtration improvement projects

Cold Roll Mill gearboxes. . .
The large and small gearboxes in the plant’s Cold Roll Mill (Fig. 4) were also investigated. Both gearboxes showed normal to low wear on the oil analysis reports, with emission spectroscopy results of 34 and 16 ppm of iron, respectively. As shown in Fig. 5, evaluation through an in-house 0.8µm micro patch revealed a significant number of large metallic particles in the fluid of the relatively new large gearbox. Particles >5 microns cannot be seen with emission spectroscopy, rendering them undetectable by the lab analysis.

The large gearbox is lubricated with ISO EP 220 oil. After the contamination was discovered, Dale Jones brought in a 4-gal.-per-minute filter cart fitted with a 36” tall Hy-Pro filter element (β₅= 1000) to handle the high-viscosity fluid. The unfiltered fluid had a cleanliness rating of 23/23/18. After 24 hours of filtration, it had a cleanliness rating of 17/14/11. The condition was corrected before any problems occurred and this gearbox has been operating flawlessly for three years. This is a clear example of the value of proactive maintenance.

Hydraulic fluids. . .
Because of the vast number of hydraulic systems in the ATI Wah Chang plant, a proactive approach to hydraulic fluid cleanliness has also been implemented. To accomplish this, Dale Jones established a program for pre-cleaning new bulk hydraulic fluids. (He had tested new AW 46 hydraulic fluid and discovered a typical shipment had a fluid cleanliness of 18/16/13—which he interpreted as not clean enough.) As shown in Fig. 6, the new pre-cleaning process involved a tote fitted with an off-line filtration system to polish new hydraulic fluid before it was used in the plant’s equipment. Using a β₅ = 1000 filter helped achieve a cleanliness code of 15/13/10. 

Conclusion
ATI Wah Chang’s well-designed, proactive lubrication pro-gram and approach to improved filtration has significantly reduced particulate contamination in the plant’s fluids: Savings of over $574,000/year have been documented. This is a result of properly structured efforts by Dale Jones, coupled with a strong cooperation between ATI Wah Chang, Moreland Oil and Hy-Pro Filtration.

The plant is well on the way to establishing a world-class lubrication program by identifying the importance of fluid-conditioning and implementing successful strategies and techniques to achieve world-class status, including:

• Use of a portable oil diagnostic system (PODS) with both online and bottle-sampling capabilities for on-site condition-monitoring and rapid analysis.
• Filtration of new hydraulic fluids to meet a cleanliness level of 15/13/11.
• Testing, evaluation and improvement of existing systems.
• Failure prevention  through the use of various condition-monitoring tools.
• Continuous improvement efforts aimed at enhancing equipment reliability.

Based on my experience, I believe this facility will continue to identify opportunities for lubrication-program improvements, and thus be able to generate numerous additional cost savings. LMT

Acknowledgement
This article would not have been possible without the information provided by Dale Jones of ATI Wah Chang: He is a true lubrication professional. He and his peers are doing a remarkable job of improving equipment reliability and providing their company with noteworthy savings that go directly to the bottom line.

I also want to thank ATI Wah Chang for allowing me to share the results of a well-designed program and the benefits they have realized from utilizing clean oil in their processes. There are still companies across industry that don’t believe oil cleanliness has an effect on equipment reliability. I trust this article will help change their perception.

Long-time Contributing Editor Ray Thibault is based in Cypress (Houston), TX. An STLE-Certified Lubrication Specialist and Oil Monitoring Analyst, he conducts extensive training for operations around the world. Telephone: (281) 250-0279. Email: rlthibault@msn.com.