Ultimate Motor Oil Thread or Why we hate CJ4/SM oils

[rickdm]

Charles, if you could show that we are not giving up horsepower by running the Brad Penn 20W-50 it would be a slam dunk. I have a couple of cases, but I have not put it in yet because it had been mentioned that I could lose 8hp over the Mobil 1 V-Twin that I am currently running. That is a lot in a track prepared 3.2 Carrera. I will anxiously await your results.

[cnavarro]

Actually, the engine made less horsepower running the Mobil 1 10w40 MXT4 and the Mobil 1 20w50 Twin on this motor than the Swepco 306 or the Brad Penn.

All we can attribute to the wild difference is that the engine we just used for oil testing this round had a much higher BHP per liter than the engine we used last year for our testing. The engine was a 2056cc type 4 that made 170HP with 9:1 on pump gas that was blueprinted and other than that, had coated bearings. Last year's dyno mule was a 1914cc type 1 that made 110hp that was pretty much slapped together with mail order parts on the cheap.

The Swepco did make more torque than the Brad Penn, just want to point that out. I know it's expensive and time consuming, but if a few guys could put their cars on chassis dynos and do something like what we've done now twice to compare the different oils, it would be interesting to see. I have had a few shops privately email me stating they have seen gains of 8 hp with just changing an oil, in this case, Royal Purple Max Cycle, from GTX, but there is nothing saying that the same gains could have been found or improved on with another oil too. The only sure fire way it seems is to put the oil in and dyno it, just to be sure.

Give me a week or so to get caught back up from being out and I will have all the results posted, as well as the average HP numbers for each run.

[David E. Clark]

Originally Posted by cnavarro

... I know it's expensive and time consuming, but if a few guys could put their cars on chassis dynos and do something like what we've done now twice to compare the different oils, it would be interesting to see ...

I know that this is not exactly what you are referring to but this article is interesting anyway. *This "shootout" appears to be sponsored by Amsoil (no affiliation) so that possible bias needs to be taken into account when evaluating the results.

A Shootout of a different kind: major oil brands go head-to-head on the dyno and in the lab.

[cnavarro]

Here we go. I've tabulated the averages from every pull and it looks like Brad Penn and Swepco were basically identical in performance, but the Swepco edged out just by a hair with a lower BSFC.



Just for identification, Swepco 15w40 is #1, Brad Penn 20w50 is #4, Royal Purple Max Cycle 20w50 is #5, and #9 and #11 are Mobil 1 4T 10w40 and Mobil 1 V-Twin 20w50 respectively.

#12 and 16 were Mobil 1 0w20 and an unspecified racing 0w20 oil, as we were wondering what effect viscosity would have on our testing. By no means would we recommend runnning anything that light ever!

[WERK I]

Thanks Charles.

The Mobil oils were the only synthetics in the test?

[billybek]

I was wondering (and please direct me with post #) if the new M1 turbo diesel oil has been evaluated? Also was looking at the Royal Purple 10/40 semi synthetic (not cycle max), has that been evaluated as well? Had a look at the local automotive supplier and am looking for something that is easily available to me..

[cnavarro]

Sorry, there just wasn't enough time to do every oil. We tried to hit the popular ones.

There were other semi-syn and full synthetics tested, but out of courtesy to their manufacturers, we have omitted their names. I have only identified the oils most oftenly identified and used throughout this thread.

Here are the peak HP numbers:

[cnavarro]

Originally Posted by billybek

I was wondering (and please direct me with post #) if the new M1 turbo diesel oil has been evaluated? Also was looking at the Royal Purple 10/40 semi synthetic (not cycle max), has that been evaluated as well? Had a look at the local automotive supplier and am looking for something that is easily available to me..

RP 15w40 (the regular royal purple) was #7. Big difference between their Max Cycle and even Swepco 15w40.

Another oil we would have liked to test, but could not get in time was Brad Penn's 15w40 conventional oil.

[cnavarro]

Torque numbers at 4500 rpm:

[Bill Verburg]

Just as an aside it has been reported that EOS is available again. The new GM part number is 88862586

Haven't tried to buy any yet so can't confirm.

[kang]

Two M1 0W-40 oil analysis’s for your consideration. They are a bit old, and I don’t recall if they were the SM version or the previous version. I usually get my oil right before I change it (I don’t stock more than a couple quarts), so perhaps someone can tell from the dates what version of the oil this was. I don’t know why I edited the date off of the second one, I guess I got carried away when I removed my name and address. The date should be 2/16/06.

I used a bunch of fuel system cleaners prior to the first analysis, and a TON of them in the second.

I’d welcome any comments, especially regarding iron. The zinc levels seem OK to me.

[RWebb]

Does each bar represent a single replicate? Or did you do multiple runs on each oil?

If only one run was done on each, then I would suggest doing 4-5 runs on at least one or two oils. As it is, you have a small difference (range ~4 hp) and no apparent intra-class sampling. i.e. no conclusion

[cnavarro]

On the averages, each oil was run at least two times. If there was a discrepancy in any one number, the run was repeated until the numbers were consistent. We also threw in some controls that were tested repeatedly, to make sure the results were also consistent. The peak number was obviously just that, the peak, from the best run.

We started each pull with the oil at 180F and made sure that the oil was at or as close to 220F at the end of each set of sweeps from 3,000 to 6,000 rpm. The cell temperature was also closely controled and we made sure each session started with the air inlet temperature as close to 72F as possible with ending inlet temperatures not experiencing more than a 10 degree delta from start to finish. We've done this kind of testing before, and we know that you have to be consistent and meticulous, or it's not worth doing.

Also, worth noting, we had excellent weather and the barometric and altitute pressures did not change for 8 hours straight!

For not being paid anything and certainly not having the oil provided free for doing the testing, not to mention the cost of thrashing on a customer's engine (with their blessing of course) with the knowledge that it would have to come apart to be checked and freshened up at the end of all the testing is a lot to ask. Luckily Aircooled Technology is as neurotic as I am when it comes to oils and they were willing to take a few days to prep and engine, do the testing, and consume all the costs of thrashing on a motor for hours and hours non-stop without expecting anything in return (other than my time and labor to help them with some of their other projects in exchange).

[rickdm]

Charles, thank you for the hard work, the information you provide is just awesome. I know there are a ton of additional oils out there, but you have given me all I need to be comfortable that I am not leaving horsepower on the table with Brad Penn.

Thank you,

[GH85Carrera]

Charles, I also must give my thanks to your hard work to get the information out to the rest of us. I recently changed to Swepco 301. My "seat of the pants" feeling is a big improvement. No science, just a butt feeling. As a bonus I feel happy that the cams will be happy as well. The information in this post is the most useful I have ever found in all my many years of using computers.

I was just a mis-informed consumer before I saw the light of this post. Thanks for showing me the path to understanding the truth. A simple "Thank You" seems inadequate.

[RWebb]

I don't mean to be overly critical (and I have already decided to run the Brad Penn. oil).

But... loading everything into a multivariate statistical package with a few replicates like I suggested (to do a sparse matrix type analysis as pioneered by Box, Hunter et al.) would really make the study much more robust.

You're in for a dime, why not for 12 cents...

Anyway, thanks again for running all the tests.

[Jeff Alton]

Randy, maybe cough up the extra 2 cents to help.....

Charles,

All of what you and Jake have done on this is greatly appreciated. Thanks very much. Lots of people want to talk about, you stepped up and did it.

Cheers

[RWebb]

Alright - I am not a professional statistician but if you'll give me an Excel sheet with all the data for each run in it, I'll do the stats. that a scientist would do in order to publish a peer-reviewed journal article. On a time -available basis of course - I have to reserve time to attack the guv'mint - my #1 priority.

[Jeff Alton]

That's the spirit! I applaud all who give of their own time and money to educate the rest of us!

Cheers

[cnavarro]

I didn't keep all but the #'s that were repeated and verified, as I was putting them right into excel. Maybe for next time, as I'm sure that the shop that I did this testing with will want to do more testing in the future, to hopefully, repeat the numbers we saw this time.