Test Procedures Prove Race Oils Have Differences - Engine Builder Magazine

Test Procedures Prove Race Oils Have Differences

One of the premier racing oil developers finally took the time (18 months) and spent the money (megabucks) to quantify the performance differences between various oil products.

Those of you who know me are aware that I’m a scientist, not totally unlike the guy from the Big Bang Theory. I’m a physicist by training (M. S.), but I’ve always worked as a mechanical engineer, since I’m such a motorhead.

With a background like mine, I’m not much on sales presentations. I’m much more comfortable with real data, not BS; and qualified people, not marketers, to evaluate when attempting to draw conclusions.

This sometimes gets me in hot water in the oil industry (and not to mention with certain magazine editors)!

Figure 1 – ASTM Sequence IIIG is a 100 hour, steady state, engine wear test that is part of the API SN/CJ4 and ILSAC GF-5 motor oil standards. This test measures cam wear and oil oxidation. The test is typically conducted on a 3.8L Buick with a flat tappet cam. Other specs are as follows: Valve Lift: .325 inch; Rocker Ratio: 1.5; Base Circle: 1; Spring Rate: 205 Lbs. (Open); Test Length: 100 hours; and Wear Rate: 6 microns per hour. Photo courtesy of Lake Speed, Driven Racing Oil.

One of the presentations made at this year’s SEMA Motorsports Parts Manufacturers Council (MPMC) Media Trade Conference really caught my attention. For years we’ve had technically developed racing oils competing in the marketplace with armchair formulated racing oils (usually a fully formulated oil plus ZDDP), ZDDP supplements, diesel engine oils and European oil formulations. One of the premier racing oil developers finally took the time (18 months) and spent the money (megabucks) to quantify the performance differences between these various products. This is some great, useful science!

First, let me give you a little lube oil testing background. All oils that wish to obtain API SN, CJ-4 or ILSSAC GF-5 credentials must pass the ASTM Sequence IIIG test (Figure 1, above). Notice that this test utilizes a 3.8L Buick V6 engine with a flat-tappet cam and only 0.325˝ valve lift. Valve spring pressures are only 205 psi open. Not many of these engines can be seen on the racetrack.

Compare these operating conditions to your race engine. NHRA Pro Stock engines utilize about 0.600˝ cam lobe lift (1.1 to 1.2˝ at the valve) and valve spring pressures approaching 1,500 psi open. Realizing that most racers use cams with about 0.500˝ lobe lift and lesser spring pressures, these laboratory engine tests were conducted to reflect the more typical cam lift/ valve spring pressure conditions.

A carefully built small block Chevy engine was repeatedly tested with the conditions as shown in Figure 2 ( at left). The engine utilized small base circle cams with 0.492˝ lift and 300 psi open valve spring pressures.

The tests included both a break-in cycle (lasting 30 minutes) and a 2-hour durability cycle with a full power pull every 5 minutes. All test cams were prepared identically with no coatings and measured prior to break-in, after break-in, and upon completion of the durability test using the Adcole 911 precision camshaft measurement apparatus (an incredibly accurate measuring device). Break-in oils were used if the race oil supplier provided them.

Figure 2 – This test was conducted on a small block Chevrolet with a flat tappet cam. The valve lift was .492; rocker ratio: 1.5; duration at .050: 236 degrees; spring rate: 310 lbs. (Open) – API SN = 205 Lbs. (Open); Break-In: 30 minutes @ 3,000 RPM with 15% applied load. The durability test lasted 2 hours @ 3,000 RPM with 15% applied load and a full power pull every 5 minutes. The cam was measured with an Adcole check (all 16 lobes) after break-in and after the durability test. Photo courtesy of Lake Speed, Driven Racing Oil.

Eighteen months of testing produced many pages of graphs and tables of data, which I will not bore you with here. Instead, I will relay my impressions of what this data showed me.

The data obtained during camshaft break-in showed several interesting results. A major API CJ-4 SAE 15W 40 diesel engine oil actually failed 5 cam lobes during the break-in (a fail was determined to be over 30 microns cam lobe wear).

This isn’t really as surprising as I initially thought.

We know that typical API SN gasoline oils don’t have sufficient ZDDP to protect cam lobes, but API CI-4 and CI-4 plus diesel oils usually had sufficient ZDDP. However, with the advent of the new API CJ-4 and CK-4 diesel oil specifications, a limit of 1.0% wt sulfated ash (SA) was established for emissions reduction. The sulfated ash of a diesel oil is the sum of sulfated ash contributed by both detergent and ZDDP.

In essence, this places an upper limit on ZDDP content since diesels need higher quantities of detergents. We now know the new CK-4 oils cannot adequately protect flat tappet cams and lifters when new. Good info.

Interestingly enough, a new API CK-4 oil was tested with added ZDDP. It also failed the break-in test with two bad cam lobes. This indicates, as I have said on many occasions, that adding ZDDP does not guarantee cam and lifter protection. ZDDP must be added to the oil under controlled conditions to prevent its premature activation. No other oils failed the break-in test limits.

Figure 3 – Post durability power. Photo courtesy of Lake Speed, Driven Racing Oil.

Interestingly, the validity of using break-in oils was verified. Even the poorest performing break-in oil provided a quarter of the wear seen with the best racing oils not utilizing a break-in oil. One break-in oil provided a third of the wear of the other two break-in oils. When every 0.001˝ of valve lift produces additional horsepower, break-in and operational wear should be minimized. This can mean a lot of horsepower after significant engine operation (see Figure 3, above).

Now to the good stuff. I am familiar with virtually all of the racing oils tested. I noticed that two of the racing oils, which I know to be arm-chaired formulations, produced roughly five times as much average cam lobe wear as did the racing oils developed using science and laboratory engine testing. Marketing claims don’t guarantee good racing oils. In fact, one of these oil marketers has since requested that their additive supplier switch them to a more scientifically developed formulation.

I won’t tell you who this is, but if you conduct elemental analysis on new oil samples, you will observe changes in the elemental composition (Calcium, Magnesium, Phosphorus and Zinc) of this racing oil before and after the switch in additive packages.

It was also interesting to observe that heavier oils (e.g., SAE 15 W-50) provided more wear protection than the lighter racing oils.

Although the differences weren’t catastrophic, the engine builder must always balance up-front horsepower with long-term longevity. This is why NASCAR engines don’t run oils as thin as NHRA Pro Stock racers do. In one case, you need over 500 miles of valve train durability. In the other case, you only need 5-10 miles of valve train durability.

All-in-all, this was very good oil testing data to peruse, and I was able to verify a lot of things I had always felt to be true. Be very careful when selecting racing oils for your engines – remember, “Engine Lives Matter.” 

You May Also Like

Shop Solutions July 2024

Those who submit Shop Solutions that are published are awarded a prepaid $100 Visa gift card from Engine Pro.

Engine Builder and Engine Pro present Shop Solutions in each issue of Engine Builder Magazine and at enginebuildermag.com to provide machine shop owners and engine technicians the opportunity to share their knowledge to benefit the entire industry and their own shops. Those who submit Shop Solutions that are published are awarded a prepaid $100 Visa gift card. Submit your Shop Solution at [email protected]. You must include your name, shop name, shop address and shop telephone number. Submitted Shop Solutions not published will be kept on file and reevaluated for each month’s new entries.

Dealing with Deposits to Boost Engine Life

Traditionally the role of engine oil has been to slow down engine deposit development, but in concocting its latest engine-protecting blend, the Valvoline team stumbled upon something… surprising.

Shop Solutions June 2024

Those who submit Shop Solutions that are published are awarded a prepaid $100 Visa gift card from Engine Pro.

pin grooves
Volvo Penta’s D17 Industrial Engine

The 17L diesel engine uses dual-stage turbochargers and heavy-duty steel pistons, making it optimally designed to power a wide range of generator sets.

All About Diesel Fuel Lines

There’s more than might meet the eye when it comes to diesel fuel lines.

fuel lines
Other Posts
Randy Breaux, Group President, GPC North America, Talks to AMN Drivetime

At NAPA, “Breaux Knows” business relationships, ABCs to avoid, and serving the automotive professional.

Randy Breaux
Fire Rings for a Better Sealing Diesel Engine

As the typical diesel engine continues to see more performance, a simple job like fire ring installation can go a long way toward keeping your engine together and working properly.

fire ring block
Diesel Transmissions for Performance Applications

Today’s diesel crowd is using their truck like a truck, and like a toy too, which requires a robust transmission.

Firepunk diesel transmission
The Impact of Diesel Fuel Injectors

Diesel fuel injectors are an essential component of any diesel engine and play a crucial role in performance and efficiency.

diesel injectors