With new API categories for both diesel and gasoline-fueled engines on the horizon, its a busy time for the industry. ILSAC GF-6, the new gasoline engine oil specification, is adding six new engine tests, including two that are measuring completely new parameters. The other four GF-6 tests, meant to be simple replacement tests for the older Sequence tests, dont appear to be the easy substitute anticipated.

Meanwhile, PC-11, the heavy-duty diesel oil specification, is adding two new engine tests focusing much more attention on challenging lubricant performance requirements.

It doesnt end there. Both of these new categories have two sub-categories. PC-11A and GF-6A are the traditional SAE viscosity grades, while PC-11B and GF-6B are new lower-viscosity grades that may not be backwards compatible with earlier products.

Additionally, ACEA (the association of European auto industry) is working on category upgrades in Europe and General Motors has issued its own Dexos 1 Next Generation oil specification, supported by four new engine tests.

The scope of these challenges is unprecedented. The industry has never introduced so many engine tests at once. Adding urgency, some of the existing engine tests will run out of test parts as early as 2016. As a result, the engineers on the test development panels and the lubricant formulators who have to balance their formulations to meet the performance challenges of all the engine tests are working very hard.

There is another group that is very busy. The American Petroleum Institute (API) Base Oil Interchange & Viscosity Grade Read Across (BOI/VGRA) task force is meeting frequently. Additive company program managers and oil marketers all know that if they have to run every viscosity grade and base oil combination in each test it could take a long time to satisfy the market with all the new oils customers need.

So why does all this testing create lots of activity on the BOI/VGRA task force? The reason is simple. Lets look to the words:

VGRA refers to a guideline to test a formulation in one viscosity grade, then read a passing result to a different grade without testing. From a formulating or deployment outlook, this is extremely helpful. If you can run and pass a test as an SAE 5W-30 and then use that pass to support a data package for your 10W-30, 15W-40 and maybe even a straight-weight SAE 50 or 60, thats a lot of testing time and money saved. Both money and time are in short supply at the beginning of a new category.

Similarly, BOI is about interchanging one base oil for another. So, if a company has access to an API Group II base stock that was not run in a test for a specific approved formulation, BOI guidelines may allow the blender to substitute the new base stock for the tested stock without testing the final formulation.

Clearly, BOI/VGRA guidelines are powerful tools for the formulators and the deployers who adapt core additive packages to meet specific customer needs. BOI/VGRA is important in getting new products to the market rapidly to satisfy customer and original equipment manufacturer needs. It does, however, come with a risk – a risk that must be tempered with data to be sure that the reads that are approved protect the consumer, the manufacturer, the marketer and the formulator.

Its important to talk about risk with regards to lubricating oils. Your typical automobile in North America is going to see somewhere around 20 to 25 oil changes in its driving life. A heavy-duty, Class 8 diesel tractor should last a million miles and, depending on drain recommendations and drive cycles, can see between 17 to 40 oil changes. Its uncommon for one oil change to cause a vehicle to fall off the road, as some think of the risk of a bad oil reaching the market, but there have been some spectacular cases where engine oil sludge has crippled large numbers of vehicles during cold winter months. The more prevalent risk from an individual oil change is the cascading damage it can unleash during the thousands of miles and operating hours it spends in the crankcase.

Sludge from one bad oil can block passages and cripple hydraulic systems, or reduce functionality and accelerate early engine failure. Wear builds up on cams over years of operation. That bad oil can start a spiraling chain of trouble as the lift is lost on the valves, or the timing chain elongates and emissions start deteriorating. Carbon deposits build behind the rings or on the top-land over time. These deposits cause wear and increase blow-by gasses in the crankcase, reducing power and accelerating oil deterioration.

The failure of an oil to perform in the field is not usually about catastrophic failure. Its usually about reducing performance and shortening engine life in ways that are hard for the consumer to see but are very real. When the consumer sees that trademarked API starburst or doughnut logo on a product claiming licensed quality, they deserve to know there is data supporting that claim. That is where BOI/VGRA supplements the whole licensing process.

To do VGRA right, a couple things need to happen. First, reads need to make sense. In a wear test, the base oil viscosity has a huge impact upon film thickness. Most engines run near 100 degrees C, so the parameter to look at is the base oil kinematic viscosity at 100 C (or KV100).

If the KV100 base oil viscosity is reduced, the oil film will be thinner and the parts will have more opportunity to contact each other. Viscosity modifiers can help formulated oils achieve the targeted KV100, but viscosity modifiers shear down, both temporarily and permanently. This leads to the principle that wear tests should be run in the lowest KV100 base oil viscosity to be approved, and then reads could be considered for higher viscosity oils.

That makes sense. This is what is referred to as a technical principle in API 1509, the document which guides engine oil program testing. There needs to be some logic, or as the Japanese OEMs put it, mechanism, behind a read. Most of todays reads, in the GF-5 and API CJ-4 era, have sound logic behind them. (There are a few that are hard to understand.)

The second thing necessary is data to support a read. Every principle should be confirmed with data, and extrapolation should be avoided. All of the engine tests for GF-6 and GF-5 were designed to operate using oils with a viscosity grade of at least SAE 5W-20. Most were designed for a 5W-30. Experience with the Sequence VID fuel economy test showed some advantage for a 0W-20 oil in this test.

Now it generally makes sense that lower viscosity oils will reduce hydrodynamic drag. With lighter oils, however, some parts of the engine may also move from hydrodynamic to boundary lubrication, increasing drag and wear. GM makes the engines used in the Sequence VID and VIE tests. They have lots of engineers who optimize pumps, bearings, cams and other components to maximize fuel economy for the companys engines. They optimize for 5W-30 viscosity grade in these engines because thats what GM recommends. Testing of oil viscosity grades the engine never was designed to operate with introduces real opportunities for unexpected results. Clearly, this is an example of why principles must be confirmed with data.

Looking at Base Oil Interchange, its a similar story, but with additional problems. Most of the BOI guidelines were drafted in the days when API Group I and Group II oils were dominant and Group III was not a factor in actual formulating. It was easy to characterize Group I base stocks by measuring three things: their sulfur content, greatly differing saturates, and widely differing viscosity indexes (80 to 120).

Its a different story today, almost 20 years later. Group I oils are out of the modern passenger car engine oil market and shrinking in the HDDO market. As Group III base oils were introduced and Group II began to dominate formulations, the traditional differentiation factors didnt discriminate well between these two types. All of these oils have essentially no sulfur. All Group II and III oils have greater than 90 percent saturates. Worse, for highly saturated oils, the designated test procedure (ASTM D2007, measuring saturates, aromatics and polar compounds) is widely known to be imprecise. Some Group II oils are starting to approach 120 V.I., which is where Group III starts.

Meanwhile, Group III oils range in viscosity index from 120 to almost 160. How do you write rules to interchange these oils? BOI is very difficult for highly refined base stocks because readily available physical and chemical tests to measure the characteristics that drive performance in the engine have not been identified.

This was predicted, of course, by almost every additive company and major oil marketer. Papers heard at various ICIS conferences abound with statements about the difficulty of Group III-to-III interchange. Further, viscosity index is really just a ratio of viscosities across a narrow range.

New measures of base oil quality will be needed in the future to do BOI right and protect all parties. How will we resolve this? It will take time, cooperation, money, testing and data.

We need quality data for both BOI and VGRA to support GF-6 and PC-11. It would be great to have the $2.4 million integrated precision and BOI/VGRA matrix of the Sequence VID, which was the only engine test added in GF-5. We all know that is unlikely to repeat itself in the near term. Still, running structured VGRA and/or BOI matrixes where we can identify principles in the priority tests is the best substitution available.

Engine tests should be run at least in duplicate with multiple base oil slates for BOI and at least three additive packages from two different suppliers to ensure the principles hold across technologies. This approach helps give confidence to the task force that they are protecting the engines as they recommend reads.

Sometimes reads are requested based on user-supplied data pairs. For example, three different companies may bring forward a few data points. Unfortunately, with these small data sets, statistics cant play as big a role as they do when a matrix is planned. Confidence levels of 95% are so wide that differences between oils are almost never statistically significant. To protect the integrity of the read, the task force has to look at trends and be willing to demand more data if necessary. Also, the user offering the data should be willing to state that they are unaware of any other data in that test that contradicts the read they are pursuing.

With budgets under pressure there is a lot of desire for BOI/VGRA reads to simplify oil certification. Frankly, the industry needs the reads. But they must be supported by principles and data that align over multiple additive technologies. This way, we are confident that we arent creating a read that leads to that first cascading crack that shortens the life and efficiency of engines in the field.

The BOI/VGRA task force, as well as the excellent people running similar services for Dexos 1 Next Generation and the new ACEA and OEM specifications, want to ensure that everyone – the consumer, marketer, base oil manufacturer and additive company – is protected. The BOI/VGRA task force searches out the principles and data to support new test reads.

Chris Castanien is technical services manager, Americas, for Neste, a global supplier of high quality base oils. Now based in New Jersey, he has 26 years of lubricants industry experience, including with Lubrizol, and has long been active in engine oil standards setting. For information about this article, email chris.castanien@neste.com or phone (440) 290-9766.