MRO Today - Lubricant sampling design

Lubricant sampling design

by Kevan Slater

The promise of enhanced reliability and increased foresight in the way we conduct maintenance practices on capital equipment has always been a selling feature of oil analysis and condition monitoring. Unfortunately the steps taken to reach the goals mandated by our lube oil program often overlook a key element vital to its success. Lubricant sampling and extraction is perhaps the most important and highly variable step taken prior to the analysis of a sample and it is also the easiest to make consistent.

Before we decide the appropriate way to extract the sample from our equipment for analysis, we first need to determine the desired end result. At this, we can work backwards to ensure the outcome can actually be achieved. We determine the end result first by examining the mandate of our lube oil program. We then can make decisions on the location of where we should be extracting our sample, what tools to use and what procedure to follow.

A common mistake made by many companies beginning a lube oil program is that lubricant sampling is not considered an important part of the program. The methods and procedures we use for sampling will determine the amount of useful data we can acquire from our sample. The goals of our lube oil program will determine the amount of useful data we need to acquire in order to perform the proper analysis to make significant judgments on the maintenance of our lubricants and equipment. At the outset of our program, we need to determine whether our program will aid us in decision-making regarding the condition of our equipment by way of the lubricant, (condition based maintenance) or simply the condition of the lubricant alone. These are two very different goals of a lube oil program and each requires a specific approach to be successful.

Historically, many programs work toward condition based maintenance, but often they do not start there. So it is important that we begin our program with an ultimate goal in place. A common misconception in the evolution of a lube oil program is that the transition between lubricant monitoring and machine condition based maintenance is the difference between the types of testing that are done on the sample. This is not necessarily true. Of course the type of test run on a sample will provide specific information either on the condition of the lubricant or on the condition of the equipment it came out of. The difference is the location the lubricant was extracted from the equipment, the method that was used to obtain the sample and the tools that were used.

For example, samples taken from hydraulic reservoirs will offer little significant data on the condition of the equipment. Reservoir samples, however, can provide excellent data on the homogenous properties of the lubricant such as acid number, remaining useful life, viscosity and additive properties. Samples taken at the reservoir or sump are typically referred to as primary samples.

Primary samples
If the intent of your lube oil program is only to monitor the health of the lubricant, a primary sample is all you will need. It is very easy and inexpensive to draw a sample for these types of tests. Often in lubricated pieces of equipment, a primary sample is the only type that can be installed. For example a bearing housing or small gearbox will only have one primary sample port.

Secondary samples
Secondary samples are taken at specific strategic locations throughout the system. On a hydraulic system, you may have a primary sample port on a main return line and secondary sample ports downstream of major components such as pumps, motors, coolers, valves and filters (Figure 1).

Figure 1: Hydraulic system often have primary sample ports on main return lines and secondary sample ports downstream of major components such as pumps, motors, coolers, valves and filters.

Secondary samples support what may be found in a primary sample and help to determine the cause. For example, if a particle count is done on a primary sample and the result exceeds a target level, it is difficult to determine the source of the contamination. It may have been ingressed through the breather, or lack thereof, introduced through contaminated oil during an oil top-up or generated internally by excessive wear. By testing secondary samples after the pump, motor and filter we can compare the ISO particle counts to that of the primary sample port to determine the cause.

If all the particle counts are similar, we can conclude that there is no excessive wear being generated by the pump or motor. We can conclude that if the particle count after the filter is the same as all the other particle counts, the filter is no longer capable of removing contaminants from our lubricant. Changing or upgrading our filter should reduce our contamination level throughout the system. Without the use of secondary sample ports, we never would have determined that we needed to change the filter.

Methods
There are several methods used for drawing oil samples from equipment. Of course, some are more effective than others; the idea is to make the method in which you sample oil consistent among those sampling. We want to make sure that each time a sample is drawn, the end result is the same regardless of the technician drawing the sample. Written procedures and specific training are vital to the success of your program.

Drop-tube sampling
Drop-tube sampling is an effective, low cost way to draw a sample with a vacuum pump for the analysis of chemical and physical properties of the oil. When using this method of sampling, there are many points to consider. For instance, in order to draw a sample, the machine must be opened and therefore the oil is exposed to the environment. Opening a machine potentially allows significant amounts of airborne contamination to enter the oil and cause damage.

The key to an effective oil analysis program is the ability to draw an oil sample from a specific location while the machine is in operation and under normal load. However, using the drop-tube method on a gearbox while it’s running poses several concerns. For one, the plastic tubing may be pulled into the gearbox. This presents specific safety concerns for the person taking the sample.

Other problems associated with drop-tube sampling include large required flushing volume, difficulties in getting a consistent sample from the same location, and problems with sampling high-viscosity fluids. In summary, this method of oil sampling should be avoided when possible.

Drain port sampling
The ideal location for drawing an oil sample from a sump or reservoir is to get it as close to the return line, gear set or bearing as possible. Another rule of thumb is to sample at 50 percent percent of the oil level. Sumps and reservoirs were designed to hold a large volume of oil, to dissipate heat and to allow air to rise and contaminants to settle. Therefore, the most concentrated contamination is on the bottom of the sump or reservoir and the cleanest oil towards the top.

Avoid using the drain plug for sampling if it sits on the bottom of the sump, even if you flush high volumes of oil before drawing a sample. If the drain port is the only way to obtain a sample from the gearbox, there are commercially available sample tubes that can be installed on the bottom or the side of the sump (Figure 2).

Figure 2: If the drain port is the only way to obtain a sample from the gearbox, commercially available sample tubes can be installed on the bottom or the side of the sump.

These inward pilot tubes can be manipulated to ensure that the sample is drawn from the most appropriate location of the sump or reservoir, and that the sample is taken from the exact same location inside your system each time. This method is a more consistent and representative way of sampling oil than drop-tube sampling.

Minimess sample valves
There are several commercially available sample valves offering different features from which to choose. Perhaps the most effective choice, typically used on larger systems, is the minimess sample valve. These special sample ports are similar to a check valve, i.e. the valve is normally closed until the sample port adapter is threaded (or pushed) on. High quality sample ports have a dust cap with an o-ring for second stage leak protection. The adapter has a hose barb on one side that accepts standard 1/4" O. D. plastic tubing.

As the adapter is threaded onto the sample port it unseats the check ball in the valve and allows fluid to flow. These valves can be used on systems from zero psi (assuming the line is flooded) to 5000 psi.

On pressurized systems ranging over 2000 psi, consider safety. For example, handheld pressure reducing valves can be used with sample ports and adapters to reduce pressures of 5000 psi to less than 50 psi. They also come in several styles that are easy to install.

Another benefit to these types of sampling valves is that they retain a very small volume of static oil (dead volume). This results in less oil needed for flushing prior to taking a sample.

Figure 3: Trap pipes are excellent for pulling a sample from a vertical, non-flooded line of pipe coming from a bearing housing or gearbox.

Trap pipe adapters
Trap pipes are excellent for pulling a sample from a vertical, non-flooded line of pipe coming from a bearing housing or gearbox (Figure 3).

Typically, vertical, non-flooded lines of pipe cause the oil to spiral down the inside wall of the pipe, depending on the velocity of the oil flow. The trap pipe essentially temporarily traps a portion of the oil flow, regardless of velocity, and allows the user to draw a data rich oil sample in a representative location.

Oil sample bottles
Most oil analysis labs will supply sample bottles. It is important to be aware of the cleanliness of the sample bottle supplied by the lab or oil supplier. Choose a low cost bottle with a consistent cleanliness.

Most bottles available fall into the “clean” category. “Clean” bottles have less than 100 particles greater than 10 microns/ml.

“Super clean” bottles have less than 10 particles greater than 10 microns/ml.

“Ultra clean” bottles are also available (Refer to ISO 3722 for bottle cleanliness guidelines). “Ultra clean” bottles are usually glass bottles that have been thoroughly washed and dried in a “clean room” environment. These bottles might cost more than the oil analysis and once the bottle is opened in a typical work environment, it is no longer “ultra clean.”

Ask suppliers about the cleanliness level of their bottles as well as how often, and under what circumstances, the cleanliness of their bottles is verified. Beware of the “sanitized” or “sterilized” lab bottles. Sterilized means that there is no bacteria living in the bottle, however, there may still be high concentrations of particle contamination inside the bottle.

Though bottle cleanliness is very important, the cleanliness of a bottle will only affect results at higher oil cleanliness levels. For example, if the cleanliness code is an ISO 19/16, there are between 2,500 and 5,000 particles greater than 5 microns/ml in the sample. At this level of contamination, typically the bottle cleanliness will not interfere with the particle count or ISO Code.

On the other hand, if the cleanliness code is an ISO 12/9 that means there are between 20 and 40 particles greater than 5 microns/ml in the sample. A bottle that falls under the “clean” category will have a disturbing effect on results because the contamination in the bottle can add up to 100 particles greater than 10 microns/ml.

There are several bottle types available at reasonable rates. The most popular bottle is a clear 4 oz. PET (polyethylene terephthalate). This clear bottle allows the analyst to use sensory evaluation on the sample. This bottle is compatible with most industrial lubricants and is readily available.

Opaque four-ounce HDPE (high-density polyethylene) bottles are also fairly inexpensive and offer excellent compatibility with a variety of liquids. The downside to this type of bottle is that they are opaque (not clear) and not conducive to a visual evaluation of the sample.

Glass bottles are excellent with respect to cleanliness, visual inspection and fluid compatibility. The negative side to glass is the high cost and the lack of durability in the plant environment. (Be skeptical of inexpensive glass bottles and question their actual cleanliness.)

Sample port identification
A savvy, successful oil analysis program incorporates the practice of labeling sampling ports with corrosion resistant tags. These tags should display the information needed by the technician to obtain a proper sample.

You may want to include items such as: sample port I.D.; machine I.D.; lubricant I.D.; and target cleanliness level.

Bar coding identification tags is another good way to label the port.

Summary
The measures outlined in this article will ensure that the life of your program does not depend on one person alone.

Try to be consistent with your program. Procedures and guidelines for the program should be set up and monitored. These should include the frequency of the sampling on a given piece of equipment, the amount of oil to flush prior to pulling the sample, the method by which the oil sample is attained, the tools used to get the sample, how the bottle is to be labeled and what information it will contain.

Additional elements include which tests are done for a specific machine on a regular basis and which tests are done on exception, and anything else that will add to the integrity of the sample being analyzed.

World-class programs do not happen overnight. The success of a program is indicative of the people and companies who strive towards specific goals. It takes resources, conviction and knowledge to implement an oil analysis program of world-class caliber. Approaching oil sampling with knowledge and creativity will allow a program to reap significant benefits.

Kevan Slater, CLS, has spent the last decade as a senior technical consultant improving the reliability of industrial equipment for numerous companies throughout North America, including Ontario Power’s Pickering Nuclear Generating plant. His extensive field experience includes spearheading equipment maintenance audits, surveys and the development of in-house oil analysis programs.

Kevan can be reached at 416-439-9425; e-mail: kslater@tricocorp.com; or on the Web at www.tricocorp.com.