Systematic Oil Analysis Interpretation

Tags: oil analysis

Oil analysis is a powerful condition monitoring tool and an important contributor to plant reliability. This technology can be applied in both predictive maintenance and failure root cause investigation and is a keystone of proactive maintenance.

Unfortunately, even with established oil analysis programs, companies may not always realize the full benefits available to them. Oil analysis programs must overcome certain challenges such as incorrect sample port locations, incorrect sampling procedures, dirty sample bottles, incorrect sample labeling, delays in sending samples to the lab, mishandling of samples, contaminated test solvents/reactants, noncalibrated instruments and incorrect/misleading interpretation of oil analysis results - which lead to a waste of time, energy, administration and resources.

This article focuses on interpreting oil analysis results. Even if every task is performed correctly, the success of the oil analysis program hinges on accurate interpretation of test data. The objective is to provide a method of translating data into useful information through a disciplined, systematic approach, such as the SACODE Method.

This methodology allows you to get the most from your program while preventing you from jumping to conclusions too early in the process, thereby reducing errors in interpretation.

Background on the SACODE Method

The SACODE method is a systematic method of oil analysis interpretation, where:

  1. “sa” stands for individual oil properties

  2. “co” stands for contaminant materials in the oil, and

  3. “de” stands for wear metals

The proposed methodology follows this same order when interpreting oil analysis results.

The SACODE Method, Step by Step

When reviewing an oil analysis report, people often focus on the wear metal data. While wear-related information is important, focusing on this alone is not recommended because more attention is generally paid to the effects of the problem rather than the causes. This is similar to viewing an iceberg (Figure 1), in which a person will likely reach the wrong conclusions when considering only what can be seen on the surface.

It is better to use oil analysis as a proactive maintenance tool to investigate the root causes (the hidden part of the iceberg), which are oil health and contamination.

Table 1. Setting Oil
Analysis Limits and Targets
(click here to enlarge)

The following 12 steps should be followed when using the SACODE method:

1. Read carefully - Consider all the information in the oil analysis report concerning the equipment, operation conditions, sample date, recent maintenance performed between samples, etc.

2. Take into consideration all sample details -

Not all information may be available; nevertheless, it is important to consider all available data/information to make the best recommendations.

3. General observations – Include type of machinery, type of industry, equipment work environment, etc.

4. Normalization - Normalize the data if necessary. See “Terminology” at the end of this article for details.

5. Identification of oil properties – Identify and label each property measured by the SACODE categories: S for health, C for contamination and D for wear. An example of this is shown in Table 2.

6. Baseline and last sample data analysis – Compare oil analysis results with baseline information. Refer to previous oil analysis results and review data of historical samples, then identify trends.

7. Setting limits - Based upon baseline data, determine caution and critical limits for each property. These will vary based on factors such as machine type and criticality. Some guidelines are provided in Table 1. Write the calculated caution and critical limits next to each property.

8. Review data - Review the oil analysis report, beginning with health properties (properties designated with the letter S). Next, review the contamination-related data (properties designated with the letter C). Finally, review the wear data values (properties designated with the letter D).

9. Data qualification – Use the following terms to qualify the data.

Mark the first abnormal value as Pivot No. 1. Review historical data to better understand what happened to that particular property/characteristic. Continue this process to review all health properties (those noted with an S), contamination data (values noted as C), and finally, the wear data (values noted as D). Depending upon sample condition, it is possible to designate several pivots in each SACODE category.

10. Partial conclusions - Note partial conclusions for each property.

11. Pivot analysis, conclusions and recommendations - Upon completion, assemble all the notes together. If pivots (or trend data) indicate a relationship, analyze and summarize the findings. Avoid making early conclusions, and conduct a field investigation if necessary.

12. Proactive and environmental actions - Define a plan of action for handling the oil and equipment. Use a color coding system to indicate conditions such as:

Example of Applying the SACODE Method

The following example in Table 2 illustrates how the SACODE method is used to systematically interpret oil analysis results.

Table 2
(click here to enlarge)

Steps 1 – 3: Known sample and equipment conditions are provided in Table 2.

Step 4: Normalization of the data is not required because all of the samples were taken approximately every 1,000 hours.

Step 5: The following measured properties can be categorized relating to oil health (S), contamination (C) and wear (D).

Step 6: Baseline data is provided in Table 2.

Step 7: Example limits are set for each property in Table 2.

Step 8: Data review is shown in the final column of Table 2.

Steps 9 – 10:

Step 11: In this example, the first pivot is the change in viscosity and the second pivot is the change in flash point. Concentrating on these items, review the common causes for viscosity increase, which include:

  1. Polymerization
  2. Oxidation
  3. Evaporation losses
  4. Insoluble materials increase (such as soot and some oxidation products)
  5. Emulsions due to water contamination
  6. Incorrect oil (such as higher viscosity)

Oxidation signals may be an increase of color and a higher acid number than the baseline reference. If the sample is clear and bright and AN value is under the limit, oxidation is not the cause of viscosity increase. Contamination with a higher viscosity oil is a possible cause for viscosity increase. Operational temperature is not high, despite an increase of 9°F (5°C).Therefore, evaporation of lighter base oils may not be the reason for the viscosity increase. There are no insoluble materials, no sediment observed and water content is under limits.

Also, current AN value is similar to baseline AN and antiwear additives (zinc).

Flash point is higher than the baseline reference. The temperature has increased to 9°F (5°C), which may happen when the viscosity of fluids in hydraulic systems is increased due to fluid friction.

In the plant, it is recommended to verify if a higher viscosity fluid (such as hydraulic oil brand X ISO 68) has been added to the reservoir.

Step 12: The following actions are recommended based on the available information:

  1. Investigate the root cause of the viscosity increase, systematically verifying the causes of viscosity increase listed above. Pay attention to the possibility of contamination with a similar hydraulic oil with higher viscosity.

  2. Once the root cause has been identified, eliminate the possibility of recurrence.

  3. Flush out the reservoir when oil viscosity is out of critical limits (refer to Note at the end of the article).

  4. There is no evidence of high wear. Add oil within specifications and continue to monitor for wear metals to make sure the equipment is not adversely affected.

  5. Color-code qualification is shown in Table 2.

NOTE: As an exemption action, and considering that oil is within additive content and acidity limits, it is not oxidized and is not contaminated with water or dirt particles. An environmental and economical option is to add a lower viscosity oil (ISO 46, there are viscosity tables for oil blending) in calculated doses to lower the contaminated oil to original viscosity specifications. It is necessary to verify with the oil supplier if both oils (ISO 46 and presumably ISO 68 or higher ISO grade) share the same additive formulation. If not, this option is not recommended unless a compatibility assessment is made.


Baseline – This represents the original characteristics and properties of the new oil to be applied in the equipment (viscosity, AN, BN, additive content, oxidation stability, RPVOT for turbine oils). It is important to measure the baseline data from the beginning when implementing an oil analysis program. Note that the data in product data sheets (PDS) is not useful for defining baselines because the lubricant manufacturer includes only general data. Changes in formulation (such as additive changes) are not always printed in PDS, which may lead to confusion and errors when interpreting results. All new oil should be analyzed.

Caution Limits - Exceeding caution limits results in abnormal conditions and requires corrective actions.

Critical Limits - Exceeding critical limits indicates a critical situation and requires immediate action.

Goal-based Limits - These limits are set as predetermined values of properties (such as ISO code cleanliness, maximum water content, etc.).

Aging Limits - These limits are a result of the oil’s normal aging process. For example: the highest permissible limit of acidity, oxidation or nitration; the lowest additive concentration, etc.

Statistical Limits - These limits are statistically determined. Data average and standard deviation are obtained. Caution limit is set at an average of +/-1 standard deviation, and critical limit is set at an average of +/-2 standard deviations. Statistical limits may be applied to wear metals.

Normalization - When collecting samples with different time intervals in reference to set frequency, it is easy to make mistakes and come to the wrong conclusions. Consider the following example: If the objective is to monitor iron wear every 500 hours and the data is 40 ppm (400 hours), 55 ppm (580 hours), 30 ppm (450 hours) and 68 ppm (500 hours), then the analyst observes which samples were taken at different time intervals Therefore, the data should be normalized in the following manner: For the first data, if during 400 hours the iron wear was 40 ppm, then what would the iron wear be in 500 hours? Answer: (40 ppm iron) (500 hours) / 400 hours = 50 ppm iron. This is the normalized iron wear data. For the other examples, values would be (in respective order): 47.4 ppm, 33.3 ppm, 68 ppm. NOTE: When taking samples, if time intervals vary less than +/-10 percent versus set frequency, normalization may not be necessary.


  1. Oil Analysis I and II Seminar. Instructor Gerardo Trujillo, Noria Corp.

  2. Fitch, Jim. Sourcebook for Used Oil Elements. Noria Corp.

  3. Fitch, Jim and Troyer, Drew. Handbook of Used Oil Analysis. Noria Corp.