Many drivers never ask these questions; they just accept on faith that the oil put in their engines will work. Very few know about the long history (50 years) behind the engine test protocols that are at the core of the American Petroleum Institutes oil categories.

Im writing about this as the result of an e-mail exchange I had recently with a reader who was trying to understand the concept of reference oils. How were these developed, he asked, and what part do they play in engine testing and oil approvals? In addition, what is the difference between a test reference oil and a category reference oil?

To explore in more detail the role played by reference oils, I contacted some experts who have a history in the engine oil approval business. My first call was to Dick Kabel, formerly with General Motors Research and one of those present at the beginning. Kabel shared some of his recollections of what GM did to develop the original Sequence II and III engine tests, which were introduced in 1960. The Sequence II measures rust protection, while the Sequence III has grown to cover a number of properties, including oxidation, wear and deposits.

From Field to Lab

Before these fired-engine tests were created, GM had been running field tests (as it does to this day) on various engine configurations and driving cycles, such as high-temperature trials and so-called Aunt Minnie low-temperature short-duration tests. Police fleets (where idling is common) and high-speed tire testing were also useful, Kabel recalled. Originally the automaker had been running these tests using commercially available oils. However, the researchers noticed that the oils they were testing could vary in composition and were not a good basis for evaluating the engines performance.

To counter this problem, GM began blending oils for itself using components from the additive industry and commercially available base stocks. It formulated engine oils that gave satisfactory performance, as well as oils that were substandard. Most important, these test reference oils did not vary from test to test.

Using these oils, GM was able to start the development of engine sequences that wouldnt require field testing but could be run on laboratory test stands, where field conditions could be mimicked and the various operational parameters controlled. Since the reference oil itself was consistent, GM engineers could tweak other test parameters – speed, load, temperature, duration, etc. – until they got lab results that replicated field results. In order to keep the field and engine tests under control, GM kept these test reference oil formulations confidential.

In 1976, GM turned over the composition and blending of its test reference oils to ASTMs newly created Test Monitoring Center (TMC). Since then, TMC has been responsible for blending and distributing engine test reference oils to any laboratory running a standardized engine test. The other original equipment manufacturers involved in sequence engine test development followed a similar pattern.

Category References

That brings us to the related topic of category reference oils. These are needed whenever a new engine oil category is being developed, such as the ILSAC GF-5 engine oils that will debut in October.

When a new oil category is under development, the OEMs typically identify some set of performance parameters (needs) that they want to improve, or that they feel the existing oil category does not satisfy. Examples might be improved wear protection and greater fuel economy. Always, the proposed new category will include test limits that reference oils for earlier categories now fail. At that point, new category reference oils must be designed to match the higher performance parameters the OEMs desire. However, all of this development proceeds with the backing of ongoing field tests.

For a closer look, at category reference oils, I got in touch with Frank Fernandez, who was with Chevron Oronite before his retirement and was based in San Antonio, where the two largest independent test laboratories are located. Fernandez was also the longtime chairman of the ASTM Passenger Car Engine Oil Classification Panel.

He explained that it is easier to make an oil that clearly passes one very difficult test than it is to make an oil that passes all the tests in a category. If you want to demonstrate the sensitivity of one particular test, a set of test reference oils can be developed. Such oils can be useful in detecting subtle severity shifts in the test.

However, the additive and oil companies will argue, such a set of oils, while demonstrating the ability of a test to discriminate and facilitating the detection of severity shifts, really is not relevant to the real world. In the real world, oil formulations have to meet performance targets in many difficult tests. These performance targets can tend to be mutually contradictory in terms of oil formulation. The additive components used to control piston deposits, for example, may make it harder to meet fuel economy targets.

Because of this, a category reference oil may be less sensitive to severity shifts than a test reference oil is. But it does demonstrate that the entire category can be met with a single oil formulation.

Reference Oils in Action

Most testing for API engine oil approvals is carried out at the so-called outside labs, which are independent of the private in-house labs owned by oil and additive companies. Two of these currently are in operation in North America, both in San Antonio. The first, Southwest Research Institute, was started after World War II. The second was Automotive Research Associates, which has undergone several ownership changes over the years and is now called Intertek. These two labs run sequence tests for all of the additive suppliers, which generally are responsible for managing the engine test programs for their oil marketer customers.

All engine tests use test reference oils to define limits but also to check for severity shifts in engine test results. This is a very serious issue for test laboratories as well as for oil formulators and additive companies. The reasons for severity shift are many, and can include changes in test parameters, operating conditions or engine test parts. Two examples of this are really instructive.

Back a number of years ago, the Caterpillar 1-H test was run to measure an oils ability to protect against piston deposits. The 1-H was a single-cylinder, diesel-fueled test engine which could be, shall we say, erratic in test performance. Reference oils could pass or fail without any clear reason. One private test lab observed that more tests passed during certain seasons than in others. Further examination of the reference test data and environment showed that inlet air humidity was a significant factor in how the test ran.

Since there was no procedural restriction on air humidity at the time, the lab began to run the inlet air through a humidifier to get as much moisture into it as possible. Low and behold, Cat 1-H reference tests started passing like crazy! With this new understanding, the lab experienced much better pass/test reliability for its test program oils year-round. The operational change would never have been made without reference oil data to point to the seasonal problem.

Heres another example. The Sequence VIII (formerly known as the L-38 test) has had problems with bearing materials over the years. This is a single-cylinder gasoline engine test designed to measure bearing corrosion protection. For a while, it seemed that every time a new set of bearings was produced for the test, the test results would shift. By examining the results of tests using reference oils, the variation in bearing materials was identified and adjustments made to bring the test back into line.

The Final Analysis

Every test has standardized fuel, parts and procedures. But with something as complex as an engine – with upwards of 400 parts – and the necessary instrumentation to run it, it is not uncommon for tests to shift severity either to the severe side or to the mild side. Only reference oil test results can keep the tests centered on the tested values and provide reliable data.

Of course, since the reference oils are supplied by ASTMs TMC, all labs run the same reference oils, so inter-laboratory correlations are maintained. This provides assurance that all labs are testing to the same degree of severity.

So the answer to how do you know you are getting a good quality engine oil? is really based on whether or not you trust the oil source. If the supplier follows the protocol found in the API Standard 1509 and tests the oil according to the protocols developed by the American Chemistry Council, you should be in the clear. I know that Im comfortable with my oil suppliers, and I hope you will be as well.