Hydraulic System Contamination Control - Case Study

 Hydraulic System Contamination Control - Case Study

 

 

by Jerry Mayo, Alumax Mt Holly 1983 (my coach and mentor for 3 years, great maintenance technical advisor, he taught me a lot)

It goes without saying that maintenance is necessary to keep Equipment in running condition, but in today's world this is not good enough. Maintenance must also ensure safe, reliable operation, in the most cost effective manner.

 

Most everyone will agree that contamination, other than component caused, is site specific, and usually shows up in the form of a liquid, solid, or gas, in the fluid. 

Contamination caused hydraulic system failures are usually not precise events. Failures are most often processes that take place over time. System reliability, component life, and the life of a system's fluid all depend on a fluid maintenance program, and the attitude of those that carry it out.

A proactive root cause problem solving approach is the best way to stop the chain of events, that lead to high cost hydraulic system operation. In any fluid power program there exists both strengths and opportunities for improvement. How we manage the opportunities for improvement will have the most effect on future overall system operating cost.

This paper looks at some tried and proven ways to enhance hydraulic system reliability, and at the same time how to greatly reduce system operating cost.

This analysis was done on two systems using a Mobil Oil DTE 25 petroleum base hydraulic fluid. However, history shows us that the results are very similar, if not the same, using a fire resistant synthetic polyol ester (Quaker Quintolubric 822) fluid. Likewise, we have had very similar cost reduction results using a synthetic Phosphate Ester (Mobil Pyrogard 53). However, we are not using this fluid, at present, at Alumax of South Carolina.

This paper does not address water base fire resistant fluids used in industrial hydraulic systems. It is not our intent, or the intent of this case analysis, to compare one type of fluid to another, or one type of testing or sampling to another. It is our intent, to show you where we came from, how we got to where we are now, and the results of our efforts.

Common Causes Of Excessive Hydraulic System Cost And System Failure

Hydraulic System Contamination

All working hydraulic systems have at least one thing in common, and that is the potential for contamination. Contamination enters the system from many sources which include the machine's environment, around or through seals, heat exchangers, make-up fluid, initial construction, maintenance, procedures, and internal generation by system component wear.

The probable number one cause of excessive system contamination is the failure to use a proper ISO cleanliness code as a part of the purchasing requirements when the bid package is released. We are usually very specific about all of the production requirements the machine must meet, but not filtration.

The components that make up a hydraulic system are like a "chain", they are no stronger than their weakest link. The contamination level of a system must satisfy the needs of the components that is the most sensitive to contamination.

In many cases we compound the problem of system contamination, after machine start-up, by not having a method for solid particulate testing, fluid trend analysis, and a work-force that is knowledgeable in contamination control.

Think about it. It takes a mind-set change to become knowledgeable in contamination control. Maintenance people have been trained all of their lives to check cleanliness by looking with the unaided eye. Unfortunately. the smallest particle that can be seen with the unaided eye is about 40 microns in size, which makes the particles we are most interested in, for fluid analysis, "invisible'. Also, we cannot assume that just because there is a filter on our system, that there is little possibility for a contamination problem.

The Effects of Contamination at Initial Start-Up

Having been involved in two primary aluminum smelter start-ups, I have had the opportunity to see many new systems commissioned to service. Most of the machines require some sort of minor changes prior to being ready for production, but these are generally routine in nature, and usually do not have any long lasting effects on the system.

However, I have witnessed three machine failures at the initial vendor commissioning, due to contamination, that delayed the start-up for several weeks, and permanently damaged system components.

I have also seen "kidney loop" (off-line) filters plug within hours after the initial commissioning of a machine, which is a sign of severe system "built-in" contamination.

There is no way to evaluate the component damage done by "built-in" contamination from machine construction, before it is filtered out of the system.

When contamination is not severe enough to cause immediate system component failure, it can still cause accelerated component wear. This will eventually lead to a rash of intermittent system problems and component there is a particulate analysis program in place to help break this chain of events. Left unchecked, contamination results in high system maintenance cost, unreliable machine operation, and reduced system fluid life.

Component Contamination Tolerance

Contamination control requirements will vary from system to system. The minimum fluid cleanliness requirements for standard solenoid controlled directional control valves, proportional directional control valves, servo valves, and piston type hydraulic pumps may all be different, and even vary from one manufacturer to another. To illustrate, I will list a few components:

The ISO (International Standards Organization) fluid cleanliness chart is a method to simplify the complexities of particle count and distribution and to provide a system with appropriate tolerance to variance as a function of the cleanliness target selected (see table two). Table Two also equates typical ISO codes to other commonly used fluid cleanliness standards. 

The ISO chart is simple to use. While not a ratio, the numerator refers to the nominal range number for the particle count of particles greater than five microns per milliliter of fluid. The denominator is the range number for the particle count of particles greater than 15 microns per milliliter of fluid. The two point code gives sufficient indication of the distribution slope and is centered around the 10 micron level, the average size of separation of moving surfaces provided by the lubricant.

Establishing Cleanliness Codes

Most component manufactures give a maximum contamination level that their component will tolerate. It is recommended that when you set up initial cleanliness code (s), for each of your system(s), you go at least one and preferable two ISO codes less than the vendor's stated ISO code. We have found that the cleaner fluid (ISO 16/13) results in better equipment reliability, lower maintenance cost, and improved fluid life, in our plant, on conventional hydraulics.

Meeting the Needs or Our System Components

As stated earlier, excessive system contamination is basically the result of insufficient attention paid to filtration as part of the system design. 

As an example, at start-up of Alumax of South Carolina, not one hydraulic system would meet an ISO 16/13 cleanliness code. Several came close to an ISO 19/16, and three or four were too contaminated to try to measure, as 15 milliliters of fluid would completely plug a 10 micron, 25 millimeter patch. 

To meet the needs of your conventional, proportional, or servo hydraulic systems, the total contamination ingression rate must be controlled by a filter (s) to maintain the established ISO cleanliness code. This means that system design and airborne contamination represent variables that prohibit accurate, or off the top of your head answers. At best, you must rely on published contamination ingression charts and make an estimate. 

How many hydraulic filters will your system need? What micron ratings? These are good questions and the following offers answers. 

We have standardized on 3 micron absolute filters, with a good beta rating (beta = 200). We like to oversize, and standardize. As an example, with a 100 gpm flow rate, have a filter rated for 150 gpm, with a very low pressure drop. Likewise, the same size elements can be used for many different flow rates by ordering filters that allow for "stacking" of the elements. 

Most, (not all) of the hydraulic systems, in the Alumax of South Carolina plant, maintain an ISO 16/13, or better, with a single return line 3 micron filter, like the one stated above, and a small kidney loop system. A minimum of two filters are recommended on both return and pressure lines for proportional and servo systems, using one hydraulic pump. 

However, the object is to lower the system contamination, and maintain it at, or below, the established ISO cleanliness code. If that requires both return and pressure filtration, on a conventional system, it should be installed. The need for an ongoing in-house, solid particulate testing program combined with a trend analysis program that will tell you the condition of the system fluid, on a regular basis, cannot be over stressed.

How We Started Our Fluid Conditioning Program At Alumax Mt Holly

"A proactive root cause, problem solving approach" toward contamination control is a must if you are to achieve the cost saving potential available by reducing machine down-time and system maintenance cost. Your work-force involvement is imperative to the success of your oil analysis program.

The Mind Set Change

Alumax of South Carolina has a multi-craft mechanical maintenance force. This group basically came from plants within a 75 mile radius (of our plant), and none had primary aluminum plant experience. 

Some of our mechanics were familiar with preventative and predictive maintenance programs, but not solid particulate testing and trend analysis for hydraulic fluid and lube oils. In order to encourage the mechanics to "buy in" to the oil analysis program, we started training our mechanics to use a "microscopic solid particulate patch test kit". These are available from Pall, Millipore lab products, and others. 

Mechanics broke the seals and removed samples of hydraulic fluid from new drums, operating hydraulic systems, and samples that had been run through our portable filter buggies, and tested them. 

This enabled them to "see" that new fluid is not clean, that simply changing fluid did not clean up the system, and the fact that filtration works. I believe that this training has done more for our contamination control program than anything we have, or could have done. The opportunity for mechanics to see the results of their efforts becomes a driving force for changing the traditional "mind-set".

System Sampling

There is more than one recommended place and method of removing a hydraulic fluid sample, from the system, for testing. It depends on the fluid analysis program, and what result is expected from the specific sample. Regardless of the analysis program used, the procedure used should give repeatable results for the same type of sample taken. "Built in' sampling ports, and a written procedure are recommended. 

Our sample ports were designed and fabricated in-house. They are installed at mid-oil level, in the reservoir, on the clean oil side. From these ports we draw trend analysis (bottle) samples, and also, with the use of an auxiliary pump, take direct ISO particulate readings, without creating a waste stream, with a direct read out analyzer (Diagnetics). 

The ISO particulate readings, taken from each system, can be down loaded directly into a computer, under each unit's equipment number. 

This provides a proven, repeatable test procedure, without opening the system, and it can safely be done while the system is in operation, at operating temperature 

We still use microscopic fluid patch testing, but we are slowly moving toward our direct read out analyzer. Switching involves a learning curve, and another 'Mind-Set' change. 

Types of Samples Taken at Alumax

Solid Particulate: These samples, as described above, are done in-house, on a monthly basis. 

 

Trend Analysis: These samples are used to verify the fluids chemical condition for continued use. 

Our interest here is not in "wear metals", but in the quality of the fluid, as we sample from the reservoir, after the fluid goes through the return line filter. We sample quarterly.

Fluid Use Reduction

Changing the hydraulic fluid and lube oil when the chemical condition dictates it, through trend analysis, is reducing our plant waste stream, as well as our new product purchases.

Adding Fluid to Systems

Each of our area maintenance shops has a portable filter buggy. This unit serves two purposes. It is sometimes used to supplement a systems filters, in case of a problem, and it is used to add all fluid to the hydraulic system. They can be purchased from Pall, Schroeder, and others. 

Case Study Analysis

This case study is based on two identical systems that were severely contaminated well beyond the component manufacturer's recommended cleanliness code. These two systems set side by side, have the same running time, do exactly the same job, have the same duty cycles, and operate at the same system pressure. These systems were in service for approximately 3 1/2 years prior to the filter revision and extended yearly cost survey. There were no component changes made to the systems when the filter revisions were installed. 

This case study assumes that pump cavitation, or false cavitation is not present, in the system. Depending on the type of system, and the severity of pump cavitation, failure can occur in seconds, days, weeks, or over a longer period of time. During this failure period, the pump would be contaminating your system. In a properly designed and maintained system, pump failure, from cavitation, is rare, but it can occur. 

The results indicate that by keeping the particulate level at, or below an ISO 16/13, in the hydraulic reservoir, on a conventional hydraulic system, we have broken the chain of events that lead to system failure, caused by particulate induced excessive wear. This is the basis of my case analysis.

The Original System Make-Up

The component list as stated here is for one machine only, As previously stated each machine is exactly alike. The system was designed for "synthetic Phosphate Ester base fluid". However, the fluid used in this system, at start-up, and through this case study was Mobil DTE 25. 

All Pump motors are 1200 rpm,, and the gpm stated for each pump is for this motor speed. 

The system came with in OEM (Original Equipment Manufacturer) kidney loop filter system, that would not meet the required system cleanliness code, for the system's components. The OEM kidney loop system had an 11 gpm gear pump (Parker), and a 10 micron nominal filter (Parker 3 1 P10CEEC50PPI). 

The reservoir capacity is 180 gallons, and the two reservoir vents were standard wire mesh screens, rated at 15 microns nominal. 

The system has 15 double acting differential area cylinders, which cycle once every 2 minutes. The atmosphere around the machine contains carbon dust. None of the cylinders were supplied with "Boots". 

The system fluid is supplied by 2 hydraulic pumps. 

Pump No. 1. is a Pressure compensated Axial Piston (Vickers PVB45FLSF20C10), fitted with a 2 pressure remote compensator. The low setting is 1325 psi, and the high setting is 2200 psi. The flow rate is 24 gpm.

Pump no. 2, is a fixed displacement vane pump (Vickers V201P11PD10-L). The solenoid controlled relief valve is set at 1740 psi. The flow rate is 10.5 gpm. 

The system solenoid directional control valves were Parker. There were four 3/4 inch, and six 3/8 inch sub-plate mounted valves. Some valves controlled more than one cylinder. 

There were seven pilot operated check valves, a pressure reducing valve, and a regenerative valve, on the system, as well as flow controls, relief valves, pressure switches, gauges, etc.

Contamination Problem Awareness

During commissioning, of the system, the 10 micron nominal, kidney loop filter plugged, within a few hours. Replacement elements were purchased, at the same micron rating. 

When the system was put into production, system contamination escalated quickly. We were unable to control it. 

The first change made was to the existing filter, and reservoir vents. We changed from a 10 micron to a 3 micron absolute, and hooked our portable filter buggy to the system (the filter buggy was only 3 gpm). We installed 3/1 0 micron filters on the reservoir, instead of the screens. We were still unable to clean the system. 

Visually, new Mobil DTE 25 is a 'Honey" color. Ours was "jet black', and too 'dirty" to test. We needed to increase our filtered fluid flow, by at least the volume of the two system pumps. 

This would require system down time, and a complete "re-piping" of the return lines. The plant was in start-up and this job was a back burner item, at this time.

Root Cause Problem Solution

We finally got the job done. We added a return line filter, with a 3 micron absolute element. 

We pumped the oil out of the reservoir, through the filter buggy, into drums. Cleaned the reservoir and filtered the fluid back into the system. The system was returned to service. It took about two filter changes to clean the system. 

The system is now able to maintain an ISO 16/13 or cleaner. Visually, now, new fluid looks like system fluid. The carbon is gone. 

Savings From Contamination Control

As is indicated by Figures One and Two, system operation cost dropped steadily from 1983 to 1987. Present component costs are, on average, 90% lower than at the time the project was begun. Monitoring component replacement costs for two more years (1988 & 1989) shows that the system operational cost will probably remain at about the 1988-99 average.  

 

The component cost, for the base year (1983), came from our closed out work orders, in our computer files. For the purposes, of this cost study Both rebuilt and new components were considered "new'. No "burden" has been added to these graphs, for labor, or machine down time, as these cost will vary, from machine to machine, and plant to plant, But, these "soft" costs are significant and should be considered as you estimate the cost of contaminated fluids in your plant or application. 

Enhancements After Contamination Control

A good waste reduction program will reduce the volume of fluid you are purchasing, the volume you are sending out for disposal, and the shipping and handling charges, for receiving and disbursement, of both new and used fluid.

There are reasons why we might decide to scrap, otherwise good, lube oil and hydraulic fluid. Free water, which could come from a leaky heat exchanger, or added with new fluid, and excessive solid particulate 5 microns or less in size, called silt. 

While it is usually not economical to filter out free water, or, to maintain your system in a nearly silt free condition at all times, this can be done, economically, when, and if, the need arises, with a portable oil purifier 

I suggest that you consider a portable oil purifier, as an enhancement, to your "use reduction, by contamination control program."

Conclusions

Clearly, controlling contamination is a very easy and cost effective way to improve machine reliability and minimize machine failure related costs. An aggressive contamination control program should include: establishing fluid target cleanliness levels sufficient to meet component life extension and machine reliability objectives, aggressive exclusion and removal of contamination, and a feedback and monitoring system to assure cleanliness targets are achieved and maintained. 

* The author wishes to acknowledge the assistance of Drew D. Troyer, of Diagnetics, Inc., Tulsa, Oklahoma, in the development of this paper. 

The Author: Ricky Smith CMRP, CMRT, CRL

About:

To all my friends, The Maintenance Community on Slack is an incredible free space where over 1,500 maintenance and reliability professionals like myself share real life experiences with each other.   

To join us, sign up here: https://upkeep.typeform.com/to/icC8EKPT