RI A6

I thought I'd compile and provide some more data on the RS4 study. The following weblink provides a tabbed spreadsheet with the raw data for 17 Audi RS4 engines and 11 different oil types. The only oil that I identify explicitly is Renewable Lubes BioSynthetic 5W-40, which is highlighted in blue. The link provides the raw data, plus multiple sorted tabs, which generally sort worst to best from top to bottom.

For some of the data, I've normalized it to 1K miles. However, I have not removed the initial oil change wear from the data. So remember that as mileage increases, the average will go generally go down. Also, on some engines samples were taken at multiple mileages on the oil before change. On other engines, only one sample was taken at the oil change. So don't assume that a low oil mileage number indicates the point at which the oil was changed.

All but a couple of samples were processed by Dyson Analysis. All of the samples come from myself and your fellow RS4 enthusiasts, for which I am grateful.

Use your own knowledge, common sense, and research to determine what is best for your engine and warranty. I will be happy to answer reasonable questions.<ul><li><a href="http://www.teraspeed.com/RS4/RS4%20Oil%20Analysis%20Results.htm">RS4 Oil Analysis Results</a></li></ul>

Bob W.

...the thing that stands out to me is the copper content in the blue-highlighted samples...why is it on average approximately 10-fold higher than the others? That's the only consistent trend that stands out at first glance.

RI A6

Copper and Antimony are the patented advanced anti oxidant and anti wear additive in this oil.

Feel free to statistically process the data in any way you see fit. In some cases I have enough data points per change to compute steady-state wear rates. In others I do not. Instead, I show average wear rates, which have an offset to them from the initial conditions after oil change.

I've just updated the data to include a tab that shows an incremental differential wear rate comparison between Motul 8100 Etech 0W-40 and RLI BioSyn 5W-40. Initial wear rates have been removed, and I have corrected for oil loss due to consumption.

Bob W.

which I assume is fuel dilution, in %.

first, I should note that this analysis is ignoring the fact that the engines all have different mileage (and a bunch of other confounding variables). I created 2 groups to compare:

Group 1: RL Biosyn 5w-40
Group 2: everything else

Using the fuel% data, I ran an unpaired Student's t-test to obtain the following:

Group 1:
Number of items= 14
Mean = 1.19
95% confidence interval for Mean: 0.9633 thru 1.408
Standard Deviation = 0.242
Hi = 1.60 Low = 0.850
Median = 1.14
Average Absolute Deviation from Median = 0.203

Group 2:
Number of items= 35
Mean = 1.40
95% confidence interval for Mean: 1.260 thru 1.542
Standard Deviation = 0.463
Hi = 2.37 Low = 0.00
Median = 1.37
Average Absolute Deviation from Median = 0.324

Comparing Group 1 with Group 2:

t= 1.65
sdev= 0.414
degrees of freedom = 47

In this sample, the 2-tailed P value is 0.106, which by conventional criteria, is considered to be not statistically significant.

In conclusion, given the available data, there is no difference in fuel dilution between RL Biosyn 5W-40 and the other oil samples tested.

S4@last!::Now B7 RS4

nice!

RI A6

Good comments for Fuel. Unfortunately, fuel dilution is the hardest to sample and measure, since it is always changing, due to power level and burn off. Elemental metals are stable, accumulate and do not fluctuate.

One example are the following sets of data points on the same engine. The first set is Motul. The second and third are RLI BioSyn

Fuel
9510 537 1.72%
10059 1086 1.65%
10606 1633 1.70%
11896 2923 2%

15738 672 0.85%
16554 1488 1.07%
18061 2995 1.29%
19452 4286 1.60%
20226 5160 1.20%

21339 1113 1.55%
25291 5065 1.03%

Something interesting is happening in the data. Motul never seemed to recover from fuel dilution. Once the dilution value raised to a high level, it would never burn off all the fuel. This might be due to the esters in the oil bonding with the fuel. Biosyn, on the other hand, seems to be able to recover from dilution. I don't have enough multiple sample runs for other oils to know for sure how they might be reacting and fluctuating.

Clearly in this case, BioSyn is slowing the onset of fuel dilution. Since it is my car and engine, I know that the driving style, fuel additives and climate have not changed. And, for the full data set, BioSyn shows a lower mean value and standard deviation. That the data set is not yet statistically significant, is most likely IMO, and that of an expert, a matter of the data set size, not the trends.

Metals are another thing and should show much more clearly defined trends. I reposted the raw data tab of the spreadsheet with the original raw numbers. You might just try a bit of statistics on Iron. You'll need to use an average, since wear is cumulative.

daveak05

as an anti-wear agent? are they adding or reducing?

also, any concern with too much copper adhering/coating on a negative charge?

Bob W.

most of your RL Biosyn data points for iron, for example, are from one engine with higher mileage...which one would expect to show less iron regardless. I'd like to see more data points from engines with similar mileage and a different oil before comparing the two.

once you have 30 or so samples for each oil type, you can be fairly confident the numbers won't change much from there, but again, there are confounding variables at work here that are going to be very difficult to control....any thinking person would see these as a weakness in any conclusions you may draw from these data.

Bob W.

I hear it's good for colds.

daveak05

use of copper radiators used to wreak havoc with early alum engine designs, under a negative charge(system), there was transfer of material.

has adding copper to this oil blend been researched thoroughly?...for potential unforseen problems, as opposed to simply being favored for it's soft metal "lubricating" qualities?