Depending on the equipment analyzed, a thorough OA program can include tests for oxidation, acid number, contamination, moisture, particle counts and more. All are important, but according to TLT readers, viscosity remains the single most important data element in an OA report. Readers noted that, in addition to viscosity being a lubricant’s most important attribute, viscosity testing is fast and reveals information on more than one lubricant property. Wear metals data was the No. 2 most-cited survey answer, although readers stressed the importance of understanding the lubricant’s work environment for best results. Those involved with metalworking fluids were understandably concerned about additives issues, which they ranked second in importance only to viscosity. Readers agreed that the biggest problem with OA testing programs also might be the oldest—improper sampling technique.
All lab-reported data. Need to see the wear metals for equipment. Need to see the additive contents to ensure no cross contamination. Physical properties to see the operating conditions.
I first look at the attention code and the highlighted source of less-than-normal code. Then the iron and copper for abnormal wear. Then I look for high silicon for intake problems. I then look for high sodium and/or potassium for coolant contamination. Last but not least is the oxidation and nitration levels, followed by base number or acid number for the life condition of the oil. Silicone and coolant are probably the most common problems detected, followed by wear metals. I sampled the oil on a freshly re-manned Cat engine. When I checked the intake system we found a hole the size of a thumbnail in a rubber elbow rubbing on a metal component. But it was too late. The engine had to be re-manned again. Cost of two re-mans: $16,000 plus down time.
ISO count, viscosity, wear metals and degradation readings (oxidation, sulfation, nitration and base number) are all used together to get the complete picture. Sometimes there is a question with some of the elemental and ISO if it’s suspected that the tech got dirt in the sample during collection, so those could be considered the least trusted on some samples.
Neutralization number and elemental analysis.
Viscosity. Incorrect oil viscosity can cause severe machine damage or failure.
No single test can determine the total oil condition, but viscosity is probably the most important since it indicates more than one property, followed by wear metals.
Water content and wear metals. Always ask—what happened and why?
Viscosity is always the key.
I would have to say that wear metals are the most used information in our lab due to the fluctuation of how much sample our customers send in. It doesn’t take much sample to test for wear metals, so we can almost always get those results. It’s also some of the most useful information that we gather, since the elemental analysis shows us the additive package of the oil, wear metals and also contamination.
Depending on the system, typically particle counts and water are the most important information to track as they are the first indicator of a change within the system. If it’s for a turbine, acid number also is on the top of this list. If there are varnish-related problems, then MPC is on top. Once a change has been detected, this is when we look at wear particles, etc.
We look at all the basic info, including viscosity, water content, wear metals, additives and particle counts. We just want to ensure the right product is being used and is not contaminated.
Viscosity and wear metals identify the cause of most failures.
For metalworking fluids, viscosity and additives concentration are the key parameters.
ISO count, acid number and moisture content.
We use the entire range of results as they are all interdependent.
Viscosity is required as it tells us if the lubrication film is thick enough. Total contamination is an easy test and very important as the contamination of a lubricant is closely related to wear and lubricant-related equipment problems.
Wear metals, acid, elements, viscosity.
Wear metals is the information that predicts failures.
Viscosity to determine any comingling or severe service. Acid number to determine the amount of oxidation degradation. Wear debris to help determine if components are failing. Particle count to determine cleanliness level. Additive levels to determine if the additives are being depleted or comingling with another product. All of these values are important—which is more important depends on the kind of equipment.
Viscosity because it’s a very fast test.
To me it’s always a combination of information and the individual type of equipment analyzed.
Viscosity is probably the most trusted. ICP/RDE has limitations that can be misleading. ISO particle counting also has room for error. Trending all of our oil analysis results is the most powerful tool we have.
pH, acid number and viscosity are most relevant for PAGs.
Viscosity is the most frequently used piece of information for our facility. It helps us head off equipment issues due to our two most-common problems: incorrect oil added to a piece of machinery and excess water in a system.
Viscosity: lubricant health. ISO count: lubricant health, machine condition. Flashpoint: lubricant dilution, safety concerns.
Viscosity at 40 C and 100 C, as this is the most important property of a lubricant. ASTM D445 test is a well-established and reliable procedure.
Oxidation. This is how I judge if the oil needs to be changed.
Viscosity is most important in the initial use of fluid. Other things become more important later.
Viscosity. Because your lube could degrade quickly.
Moisture percentage. If it’s wet, nothing else matters.
We do not perform an oil analysis program such as is used for in-use lubricants.
Visual, viscosity, X-ray and FTIR show contamination and performance additive levels.
Viscosity and particle analysis of any sort. Those two pieces of information can help predict/prevent catastrophic failure.
Both the viscosity and wear metals profile results are key. In addition to providing insight into the lubricant’s condition, additional information such as airborne contaminants, dilution and introduced contaminants can be detected and are indicators of the system’s overall health.
Wear metals, especially in a dusty environment.
Viscosity, ISO count, wear metals, RULER, varnish potential MPC (dE) are the most used and trusted information in regards to our oil analysis program to help predict and trend potential bad actors before they become catastrophic and take down the equipment.
Elemental analysis via ICP. It gives you a good indication of whether your oil is indeed your oil and if it is contaminated or shows wear metals, etc.
Viscosity. It is a quick and easy way to check the integrity of the fluid, whether it has oxidized or is contaminated with another fluid.
Wear metals and fuel dilution.
It is very hard to determine which test is the most important, as they all are valuable in their own way.
Viscosity/ISO count and wear metals. Dirt levels are important to valves, and wear metals help determine if you’re having motor problems.
Wear metals. Done properly, it gives a good picture of the overall health of the system.
Elemental spectroscopy (contaminants, wear metals and additives) because it measures contaminants at their most damaging size, wear metals in their earliest form and helps confirm (along with viscosity) that the correct lubricant has been applied.
Viscosity. It is the most repeatable measurement from our lab.
What percent of your condition monitoring program is based on oil analysis?
Less than 25%
10%
26%-50%
38%
51%-75%
22%
76%-100%
30%
Based on responses sent to 15,000 TLT readers.
Q2. What is the No. 1 fault/factor that invalidates or calls into question your oil analysis results?
No. 1 problem: incorrect info on analysis sheet with the sample. I ask for copies of the sheets so I can check them.
Silicon, aluminum and ISO have been thrown off by poor sampling practices. To combat we discuss it with the techs when we notice a trend and work on keeping training up to a high standard. Continuous improvement.
Oxidation.
Poor sampling technique/procedure. Must obtain a quality sample to produce quality analysis results. I collect the critical oil samples myself.
Validating alarm limits. The lab has no idea of conditions of environment that the asset is ingesting.
Sampling method, including interval.
Particle counting; faulty sampling procedure. Supplying clean and suitable sample bottles, plus training.
There are so many factors that can invalidate our oil reports, from customers not indicating which product they’re using to not having a baseline for FTIR comparison. We encourage all of our customers to fill out all the information on their sample forms and also to send in a clean oil to use as a baseline.
Improper sampling methods. We do not support the bag method as bottles typically have 10 ppm 10 micron and larger, and this will skew results in clean systems.
Improper sampling. Repeated training classes.
Often our lab will report the heavier gear oils, 320 and 460, as being light, but when they are resampled they test normal. The have told us they do so many lighter samples that cross contamination can occur.
Time and distance of the damage site from which the oil samples were taken.
Contamination/filtration.
External contamination with different fluids/lubricants.
Calibration or verification for analyzer.
Delay between getting the sample to the lab and receiving the results. Samples couriered to the lab and results are cloud based and easily accessible.
To have a better understanding of the issue.
Responsible people are not always familiar with the advantages of oil condition monitoring. I have organized many training sessions dedicated to maintenance people.
Response time to review reports and after-sales service and technical support.
Contamination of the sample. Solution is proper training and procedures when obtaining the sample.
Water in the sample. Make sure the oil is up to operating temperature before taking a used-oil sample.
Customer not including enough information on oil type, equipment type, service time on oil, etc.
Oil sampling practices. Dedicated oil sampling ports fitted to all critical plant, train personnel on the importance of correct sampling procedures and educate on the principles of rubbish in, rubbish out.
Variation in acid number from the lab.
Sample quality is the biggest issue we have with the accuracy of our oil analysis. We have had a huge turnover in people, and it is difficult to get someone trained and keep those same people on that sampling routine.
High number of largest particle sizes suggests poor sampling technique. We then resample.
Additives and wear metals concentration. Some elements are harder than others to detect, a situation also affected by contamination. We often retest to confirm results.
Improper identification of the equipment sampled. We have gone to entering all equipment into a database and using QR codes to tie the equipment and sample together.
Sampling. Error taking and/or properly labeling samples.
Additive metals don’t match baseline.
Need to trust that the sample was taken properly. If you don’t know what you are testing then it’s impossible to draw any conclusions.
Calibration of the X-ray.
Inconsistent results endanger confidence in the oil analysis program/company. Mitigation efforts include training in oil sampling and submission processes, re-submission alerts to the analysis company and, in rare instances, changing to another analysis company.
The use of automatic viscometers and duplicate runs.
Particle wear—it’s key to make sure sample oil is collected at the right spot.
The variance of a Varnish Potential MPC (dE) and Karl Fisher oil sample result. We usually re-test and compare the results.
Elemental analysis inaccuracies. Modify the method used accordingly.
When there is an obvious error and it is apparent the person has not looked at the sample before being released.
Base number measurement. Sent out samples to outside labs. Ordered a new titrator.
High particle count and/or silicon with no associated wear metals often indicates a poor sampling procedure. I usually re-sample to confirm but recommend that a better sample valve and/or procedure be implemented in the future.
Has oil analysis saved your company (or your customer’s company) substantial money through allowing predictive maintenance?
Yes
87%
No
13%
Based on responses sent to 15,000 TLT readers.