Engines are unpredictable, living, breathing machines and our industry is responsible for the blood that pumps through their veins. But how can we know we have the right blood type? Engine testing determines whether a lubricant is truly capable of protecting against a multitude of health issues that might otherwise result in early failure.

Developing tests that can be used to compare lubricants demands sophisticated life support systems, experienced engineers and an appropriate test procedure. It is critical to define a robust method capable of differentiating products with the level of reliability, repeatability and reproducibility that oil and additive developers require to refine one of the engines most critical components.

An engine oil must keep pistons free of carbon build-up; protect bearings, valve trains, rings and liners against wear; minimize oxidation, acidity and viscosity increase; inhibit sludge or ring sticking; and safeguard against low-speed pre-ignition. Testing performance in each of these areas requires particular hardware configurations, special fuels and specific operating conditions. There are many levers that a test developer can pull to increase or reduce the severity of a particular parameter, but every change starts a chain reaction of effects that must be pre-empted, analyzed and evaluated. The development process is a minefield that must be trodden carefully.

Getting the Test Ball Rolling

Who is responsible for what within lubricant engine testing? In the European market, the Coordinating European Council develops engine tests on behalf of additive, oil, hardware and fuel manufacturers (the Additive Technical Committee, Technical Association of the European Lubricants Industry, European Automobile Manufacturers Association and Concawe, respectively).

If an issue is reported to an original equipment manufacturer – low-speed pre-ignition is a good recent example – and that OEM determines that the engine oil is at fault or could have better protected its engine, it may report this to other ACEA members. In this case, the CEC may initiate an industry-wide test development to uncover the issue, which starts with drafting instructions to describe what the goals and boundary conditions of a test must be, known as the terms of reference. A test development group is formed and is responsible for the creation of a method that fulfills these terms.

Alternatively, an OEM could develop an in-house test. While the development process is similar, the politics are far simpler. There are advantages and disadvantages to both, but the focus here is on how things currently work in a CEC environment.

The test development group comprises extremely experienced professionals from all areas of the lubricant industry. They come together to determine how a method should evolve to best assess and differentiate lubricant products. Their primary concerns include protecting hardware, passing profitable products, encouraging innovative new formulations and ensuring robust testing practices. Not all of these concerns align, so we end up with a battleground of political motivations, technical opinions and testing experiences.

Phase One: Test Development

The first crucial decision is the hardware, or test engine. In fact, in most CEC groups the hardware is fixed but it must be fit for purpose. Valve train wear is something that an oil must protect against, but not every engine suffers from large amounts of wear. A sliding cam and tappet combination operates in the boundary friction regime separated by a monolayer of lubricant molecules, and we can expect to see wear levels in the 5 to 100 micrometer range – a small but measurable level of wear. A roller-follower valve design, by contrast, may only produce 1 micrometers or less of wear over 1,000 hours of operation. This is not significant enough to differentiate oils and, moreover, is not an issue for the hardware. It makes no sense to test for valve train wear in this design of engine.

Assuming the hardware to be used is susceptible to the issue under investigation, the next step is to define a suitable test fuel. Generally speaking, it makes sense to use a fuel that is representative of a worst case but is also readily available on the market. This will likely be a B7 biodiesel or an E10 gasoline but the choice is dependent upon the testing goals. Another would be a sludge-test fuel, a very poor-quality fuel containing gum components considered to be pre-cursors of sludge formation. The final consideration for fuel is whether or not changing the fuel batch is critical to the test outcome, since slight variations could cause significantly different test results. If so, the batches must be tightly controlled and perhaps even individually approved.

Even with the most appropriate hardware, every test requires a defined cycle comprising the various elements that exaggerate the issue of interest. A field issue that manifests after thousands of kilometers of operation must be reproduced in a reasonable length of time, typically no more than 300 hours. This is where the test development group participants years of experience makes a real difference.

It is only with a combination of understanding the required precursors, the effects of operating conditions and the hardware response that test conditions are selected. There is room for improvement, however, and better use should be made of advanced measurement technologies.

Through an iterative process, in which the test cycle is lengthened or shortened, fuel specification optimized and operating conditions adapted to modify severity, the test procedure is refined. Ultimately, the test must reliably (without engine failure) and with acceptable repeatability discriminate between high and low reference oils – those that provide the performance being tested and those that intentionally do not. The number of iterations varies significantly and can be very unpredictable. This can introduce budgetary concerns and could require additional funding from each participant.

Phase Two: Round Robin

Signing off on the final procedure requires data that shows that the test is not only repeatable with the same setup by the same observers but is also reproducible independently by others. The groups statistician determines how many tests are required from each participating laboratory and in which order the reference oils should be tested. If the results satisfy predefined statistical limits then the first official issue of the test procedure can be published and test limits for the various ACEA or OEM approvals can be specified.

As the complexity of engine hardware has increased it has become more and more difficult to maintain stable, repeatable conditions from one test to the next. Removing the influence of the hardware on the result can be very difficult. It can also be incredibly difficult to anticipate the hardwares response to a particular change in operating conditions and hence the sensitivity of the test to a change in operation. The result is an unacceptably long test development period.

What can be done differently? There are so many innovative measuring technologies at our fingertips, technologies that allow measurement in real time of component wear, soot content, fuel content and even oil emission.

While developing a sludge test, it is critical to know how much fuel the oil contains. This could range from zero up to 35 percent, depending on the engine design and test conditions. In one single investigation, it is possible to determine the speed, load and coolant and oil temperatures required to stress the oil and introduce a repeatable fuel dilution.

Wear can be very accurately measured by comparing pre- and post-test measurements, but when asked which operating point was the most severe you would not be able to provide an answer. The radionuclide technique – in which thin layers of components surfaces are radioactively activated – has been around for more than 25 years. It has been refined and improved over that time to such an extent that pairs of components can be activated to analyze how much wear one is causing to the other, on a nanometer scale, in real time. There is no better tool with which to develop a wear test.

A particularly impressive technique is the analysis of molecules present in exhaust gas, which can be used to estimate how much unburned oil is being emitted by the engine via the combustion chamber. This can be achieved for individual cylinders at a crank-angle-degree resolution, such that the opening and closing of the exhaust valves can be seen. With technology like this, an engines oil emission can be mapped in every operating condition in a matter of hours, a task that would have previously demanded days of tedious and potentially inaccurate measurements. And the result could help to avoid extremely high oil-consuming conditions so that test length can be extended without having to top-up the oil or to analyze the effect of temperature, viscosity or fuel characteristics on the oil emission.

Combined Effort

These measurement technologies – combined with the knowledge and expertise of test engineers, hardware manufacturers and oil formulators – have the potential to reinvigorate and advance the test development process.

Utilizing these technologies of course demands a fundamental change in approach. It requires a large portion of a test development budget to be apportioned to a small number of highly scientific tests. This, in turn, demands intricate planning, the involvement of measuring experts and component experts to further inform the group and interpret the data acquired. The aim however is to generate a fundamental understanding that can not only reduce the iterative phase of the test development (hence reducing overall development costs) but also inform severity investigations later on in the tests lifetime.

Timothy Hadaway is the lubricant testing team manager for German testing company APL Automobil-Pr ftechnik Landau in Mannheim. He has an academic background in automotive engineering and previously held a position in the testing department of Lubrizol.