Developing a Reliable Oil Analysis Programme

An effective lubricant analysis programme increases the reliability and availability of equipment while reducing the costs associated with labor, repairs and downtime.  Carefully selecting an oil analysis programme that best fits the needs of your company will help you to fulfill your specific goals.

The most important focus of any oil analysis programme should be quality assurance. Most people tend to think of quality assurance or quality control (QA or QC) programmes as being for laboratories only and so rarely originate the quality assurance and control process in their own plant.  But in oil analysis the QA/QC process should not begin and end in the laboratory.  Some of the most important aspects of control must initiate from within – let us consider why.

Take control

Without the proper controls in place prior to analysis, testing may be performed on non-representative, mislabeled or out-dated samples, which in turn will lead to the wrong corrective actions and to added costs on the oil analysis programme.  If these issues of control are not considered prior to the sample arriving at the laboratory, any test results obtained will be of small value.

There are also control issues to consider once the analysis has been performed and the results have been obtained.  Matters to think about include:

• Storage of new lubricants

• Testing of new lubricants  

• Sample labeling

• Sampling techniques

• Selecting applicable test methods

• Timely sampling, analysis, and corrective action.

Supply and storage

Quality assurance and control begins as soon as the supplier delivers the lubricant. Ensuring every time that the correct lubricant has been obtained from the oil vendor and that the lubricant container has been handled properly will prevent problems from being introduced into the reservoir.

New lubrication containers must allow for easy identification of the lubricant type and grade.  Proper storage of the lubricants is also critical.  Although it may sound basic, storing the containers in a clean, dry place will aid in ensuring that the lubricant is not contaminated with debris or water. The storage containers should also be kept in a controlled environment, protected from excessive heat or cold.

Remember that it is possible to receive mislabeled or contaminated lubricant from the oil vendor.  At least periodically, new lubricant samples should be submitted to the laboratory for analysis as part of the quality assurance programme. You may be surprised to find out how “dirty” your “new” oil really is.

The process of transferring lubricants from a storage container into the reservoir can easily be one where contamination is introduced.  It is important to set out and implement clear procedures to reduce the risk of contamination.  If the product is not being dispensed directly from its original container, then a clean, dry receptacle must be used for the purpose of moving the lubricant to its point of use.  Each container or device used for dispensing the lubricant should be clearly marked and used exclusively for one product.

Good sampling practice

Special precautions should be made when obtaining a sample from the machine reservoir for testing.  It is vital to use a technique that will provide a representative sample.  The sampling procedure must ensure that the technique used is consistent each time a sample is drawn to send to the laboratory.  This will guarantee that tracking and trending values received from the laboratory are consistent and representative over time.  In other words, the results will be meaningful.

How exactly can this be achieved?  With any oil analysis programme, it is important that the sample is drawn at the same location every time, because variations in the oil can sometimes be seen at different locations in the reservoir. Whenever possible, the sample should be drawn while the unit is operating.  If for safety or other reasons this is not an option, the sample should be drawn as soon after shutdown as possible.  This ensures that separation of any particles or water in the reservoir does not occur and lead to a falsely ‘clean’ or ‘dirty’ reading.

One answer may be to install a valve specifically for sampling purposes.  The valve should be installed upstream of the filter so that problems such as wear in the unit can be identified.  Periodically taking a sample after the oil passes through the filter can also be useful.  This will give a good indication of how well the filter is working. When sampling a unit with a valve, make sure the valve has been cleaned and thoroughly flushed before collecting the sample. 

When sampling a unit that does not have a valve, use a siphoning pump and tubing.  It is critical that all sampling equipment be clean and dry.  When using tubing, cut the length so that the sample will be drawn from the mid-point of the reservoir.  Each piece of tubing should be used only once and then discarded to minimize the risk of contamination.  Extreme caution must be exercised when using tubing to obtain oil samples from units that are operating.  The tubing can be caught by moving parts and pulled into the unit, resulting in damage.  In these cases, it is better to shut down the equipment before obtaining a sample.

The most undesirable method for obtaining a sample - but surprisingly still wide-spread in use - is through a drain plug.  If this procedure really must be used, an adequate amount of oil should be drained first so that particles, debris, and water from the bottom of the reservoir are removed before collecting a sample.  Be cautious of the results obtained from a sample that has been drawn through a drain plug.  It is more likely that they will not be representative of the reservoir.

What happens next

Once obtained, all samples should be placed in clean, dry bottles that have never been used.  Depending on the type of analysis, bottles can be purchased that have been certified to meet specific cleanliness levels.  Labels should be placed on sample bottles immediately after sampling, to prevent confusion. Label information should include - but need not be limited to -reservoir identification, date of sample, lubricant hours (if applicable), and run time hours (if applicable).  For tracking and trending purposes, the same reservoir identification should be supplied to the laboratory each time a sample is drawn from that reservoir.

Once a sample has been drawn, it is important to have it delivered to the laboratory as quickly as possible.  It is equally important that the laboratory give prompt results. When selecting a laboratory for oil analysis, look for attributes such as quality assurance, turn around time, testing capabilities, and good customer service.  A quality laboratory will have an in-house quality assurance programme, for example the 10-CFR-50 Appendix B quality assurance and control programme, or an ISO certification.  A laboratory with good quality assurance should provide accurate, dependable results on every report it provides.

You and your lab

Building a good relationship with the laboratory can only be advantageous to your oil analysis programme.  Whenever possible, tour the laboratory, and perhaps bring along your maintenance technician or whoever will be taking the samples.  This will allow the opportunity to see first-hand the way the samples are handled and how the tests are performed. You can also ask to receive copies of audits - or you could actually audit the laboratory’s quality assurance programme yourself

Provide the laboratory with as much information about the equipment, lubrication types, environmental conditions, and maintenance practices as possible.  In return, the laboratory should be able to provide an in-depth interpretation and recommend action in the oil analysis report. .

Once you have received an oil analysis report from the laboratory, it is imperative to act quickly on any findings or recommendations.  It may be useful to carry out a second sample to verify the results, just in case the original sample was contaminated or not representative. If you have any questions concerning the data findings or the recommendations given, be sure to contact the laboratory directly.  A lack of open lines of communication will only compound any problems.

It is crucial that you give consideration to the type of analyses you are going to request for your oil samples. You will need to select a sampling frequency and set of tests for each of the different types of machinery in your plant.  For example, engine samples should be tested for coolant and fuel contamination, but these tests would not be applicable to lubricants taken from hydraulic equipment.  And to make best use of testing, ensure that there will be sufficient data to give you information not only about the state of the lubricant but about the equipment from which it came.  Basic tests will only give an idea of the oil quality without telling you  anything about the condition of your plant. It is by incorporating condition monitoring into your oil analysis programme that you will genuinely be able to save pounds or dollars on your maintenance bills. To this end, critical equipment should be sampled more frequently, and analysed more extensively than other less essential plant. If you are setting up an oil analysis programme for the first time in your factory, or have introduced a new piece of equipment to the scheme, it is good practice to analyze the oil once monthly for the first three months to establish a baseline or ‘normal’ reading for the unit.

Conclusion

Keeping a close eye on your oil analysis programme can really ensure the maximum value for the company’s investment in its equipment.  The process of setting up your programme should include auditing the laboratory’s practices, and should establish clear and well-understood routines for lubrication storage, sampling procedures, sample handling, testing methods application, and responsiveness to the analysis results.  Ongoing monitoring of the whole package will help you control, and ensure a more reliable oil programme.

as seen in:
Machine Plant & Systems Monitor
Jan/Feb 2000