What oil?

[Tbonem3]

Interesting. What Motul did you use in the trans? And I'm assuming Gear Competition 75W-140 in the diff, yeah?

What was your username on m3forum?

[IamFODI]

Also used Motul in the last tranny and diff oil changes.

Interesting. What Motul did you use in the trans? And I'm assuming Gear Competition 75W-140 in the diff, yeah?

[Altaran]

Motul 8100 X-Power 10W60 for the Motor. Also used Motul in the last tranny and diff oil changes. Same will happen once I refresh my steering.

[stash1]

Still running straight 20wt olive oil and all is well!

[IamFODI]

Basically means nothing - interesting, that’s a fairly definitive statement. Is that from the view point of a certified tribologist, a petro-chemical engineer, or one with none of these qualifications and who is a car enthusiast like most here

Of those options, definitely the last one.

FWIW, here's a keyhole glimpse at what real criteria sets for engine lubricants look like: Afton Chemical - Oil Specification Handbook.pdf

Scroll to your favorite API, ACEA, or OE spec and look at the criteria. Few things you'll notice:

1. There's a lot of testing in actual engines;
2. There are many different criteria covering many different aspects of the oil's performance, virtually none of which involve additive PPMs or starting TBN; and
3. All of the tests are rigorously defined (equipment, procedure, thresholds, tolerances, etc.) and backed by extensive research that shows how they relate to actual performance in real engines in real scenarios.

In other words, the people who do this stuff for a living don't like to choose engine oils based on an incomplete list of ingredients and/or a very small number of un-validated tests on homemade rigs. They want to look at a lot of different aspects of performance, using extensively validated methods and actual engines.

The "lubricity"/"friction"/"film strength" rigs Piotr Tester and Project Farm use share their overall design with the "one-arm bandit," a machine that sorta-kinda-mimics the Timken OK Load test. The real Timken OK Load test is well defined and accepted for extreme pressure (EP) performance, which is very important for greases and gearbox lubes. But there's almost nothing in an engine that sees similar kinds of loads, and formulating a lubricant for very high levels of EP performance involves compromises that are not acceptable for an engine oil (which is a major reason why engine and gearbox lubes are different). So, even if this method had been validated (which it hasn't) and were performed in a rigorous way according to accepted standards (which it isn't), it would only tell us one aspect of an oil's performance, and that one aspect would be largely irrelevant to engines.

PF's evaporation loss test is okay for what it is, and the thought of testing aged oil is great. But at no point in an engine does the oil sit at that temp under atmospheric pressure for 2 hours, and real oil aging methods involve more than just high temps (e.g. bubbling NOx compounds through the oil and then loading it with soot). All he does to justify this method is name-drop the Noack volatility test, which is interesting but very narrow in its utility and works differently from how he did it.

PF's cold flow test also is not meaningful because cold oil doesn't just passively flow through an engine. It gets pumped, and pumping generates shear forces which change the apparent viscosity of the oil. That's why real cold viscosity tests, like the MRV and CCS testing that defines an engine oil's W rating (e.g. the 10W in 10W-60), involve shear forces.

PF describes TBN as the oil’s ability to neutralize acids deposited in the oil from combustion. Like all the tables, the greater the value, the better the oil will perform.

TBN retention is also important. Different oils and applications can deplete TBN at different rates, so the starting number alone doesn't say much. Also, like any oil additive, the ones that underpin TBN have both good and bad effects. Most TBN comes from calcium- and/or magnesium-based detergents, which have obvious benefits -- but they also compete with anti-wear additives and friction modifiers, and increase ash formation.

Same deal with anti-wear additives: there's more than one kind, they don't all work the same, and their performance depends on the rest of the formulation, so the "total" amount itself doesn't tell you anything about overall performance -- and high levels come with downsides as well as upsides. And even if the amount itself could tell you something, PF's "total antiwear additive PPM" numbers aren't reliable. The real numbers are proprietary, and he couldn't figure it out with oil analysis because not all additives show up, and the ones that do (e.g. ZDDP) don't always show up in the same way.

I appreciate that we're all trying to make decisions based on the information we have. But if we were to lean on criteria like PPM numbers, TBNs, etc., here's an example of where we'd end up: MPT Thirty-K. 100% PAO and ester base stocks, with ZERO group III or non-synthetic base stock -- not even as additive carriers. Hysterical amounts of ZDDP and detergents; they no longer list the numbers on the site but they were all 1000-1500+ ppm ZDDP and 2k-3k+ ppm detergent. Massive starting TBN, e.g. 12.5 for the 10W-60. These are all the things the Internet leads you to think you want. And yet, not a single OEM, big-boy engine builder, or major race team wants a piece of it. A lot of them use TWS, Helix Ultra Racing, Mobil 1, etc. -- the stuff Internet warriors and one-arm banditeers like to scoff at for weak "test" results and low PPM numbers.

Also, if you browse the spec handbook I linked above, you might notice there's very little in there about additive PPMs or starting TBNs. Virtually everything is about actual performance, not ingredients.

What limited data is worth basing a decision on? Well, for one thing... the people who developed the engine literally told us what oil they think we should use.

Beyond that, we're unfortunately in a crappy situation because BMW never published an actual lubricant spec for these engines. They and their lubrication partners (Castrol, then Shell) are the only ones who really know the details of what the engine needs, and they're never going to tell. Everyone else is just guessing.

Barring an OEM recommendation, and in the absence of an actual spec, the (distantly) next best thing is a track record of apparent success in real performance engines, ideally including yours. Liqui Moly is the obvious go-to here. If you like big additive PPM numbers on oil analysis, Motul 300V and Red Line's non-OE-approved lineup are no-brainers; Red Line might have more of a rep in S54s per se, while 300V has way more of a track record in high-end racing, for what that's worth. If impressive specs and credible promises of cutting-edge tech are more your thing, Ravenol seems like a good call.

[Tbonem3]

What was your name on m3forum Fodi?

[Speed Monkey]

Unfortunately, these "tests" are basically meaningless for engine oil.

Basically means nothing - interesting, that’s a fairly definitive statement. Is that from the view point of a certified tribologist, a petro-chemical engineer, or one with none of these qualifications and who is a car enthusiast like most here, save for a few who have university degrees in this field. No matter where your knowledge base stands, there is some good information posted in Project Farm’s (PF) video. This detailed information is below, and PF’s data is derived from sending oil samples to a lab for analysis.

PF describes TBN as the oil’s ability to neutralize acids deposited in the oil from combustion. Like all the tables, the greater the value, the better the oil will perform.

So glancing at this lab derived information, would you buy Lucas Oil, or Amsoil based on the data below, or is it like you say, “...”tests” are basically meaningless...”. Whether you like data or not, I do like data, and the data below indicates the better performing oils would be Redline, Amsoil and Pennzoil.

[Dash1]

I stated my point clearly and you quoted it.

It's kinda stupid to design oil around additives but hey, people like adding additives for piece of mind when they can just buy good oil.
I should've said "better" to be less absolute. It's all trash we must sift through. What makes an oil better would be the MOS2 included in the base oil without the additives. Everything else is pretty much the same. I believe it makes it more shear resistant and is just a lubricant that sticks to the metal surfaces. I don't claim to be an expert, I'm just a oil keyboard warrior with an opinion.
Seriously though, I'm done with Liquimoly. It wasn't all the oil's fault I had a piston exploded into at least 100 pieces, but it just left a bad taste in my mouth. I also used Ceratec. Still went boom. I'm sure my car will be fine without it. I didn't notice any smooth coating on anything and it was in there for 5,000 miles. I'm just not a believer. They have really good marketing. They have a potent engine flush.

If your bearings went out after 5k then I'm more inclined to think that there was an improper install. Of course, there's no way to prove it and you're being biased toward an oil that many others are using with no complications and without using additional additives.

[ViN]

I am using Castro Supercar 10/60. The previous change I did switch to Pennzoil 10/60 and I don't know whether it was real or perceived, but the car didn't seem to run as well and I thought there was a bit more noise (rattling sound) coming from the warmup. I have been very pleased having switched back.

[IamFODI]

That’s my dilemma every oil service. Do I keep with the liqui moly which I put 120k hard redlined daily miles on (changed my bearings at 185k, and they looked great) which burns no oil. Or...do I go with the shell. I’m in an if it ain’t broke don’t fix it kind of situation.

As a BMW 10W-60 user currently... If I had your history of success with Liqui Moly, I'd stick with it.

Only use castrol edge 10w60 or shell helix racing 10w60. The rest (liqui moly bmw 10w60 etc) do no have the additives needed.

Not to necessarily defend Liqui Moly vs. Castrol or Shell, but what's important here isn't the additives; it's the performance. Formulations are complex and it's entirely possible to have similar performance with a different additive package.

I've spent hours with tribologists of several oil testing companies and my takeaway each time is: 10w60 is 10w60 and 10w40 is 10w40 REGARDLESS of brand!

If they said that, I'd bet good money they were oversimplifying on purpose and/or speaking within a very specific context -- OR they weren't actually tribologists (maybe they were some other kind of professional).

AFAIK, the truth is this (maybe this is what they meant):

1. Viscosity -- not just the viscosity grade, but the actual viscosity in service -- is generally considered the most important feature of a lubricant.
2. Most of the time, there won't be night-and-day differences among oils within a viscosity grade.
3. If you look at two oils with the same performance classification, they will perform similarly.

Additives can make a big difference, though. Deposit control, anti-wear performance, friction modification, catalytic converter poisoning... Lots of things.

Blackstone will tell you that we S54 guys are the lionshare of their independent customers so they have a huge database on the results of the oils we run. MOST of us, whose oil they test, use 10w60, but they also have a healthy number of those owners who DON'T. I asked what other weights they see allot and they said: 0w40, 5w40 and 10w40. I asked; "do you see any difference in those weight oils versus 10w60 in how those alternate weights test out?" and they say NO.

Agree completely with the overall message, i.e. that there's no evidence that screwing around with viscosity will make the S54 or any other engine blow up or last much longer. Just wanted to add that what Blackstone sees isn't actually wear. It's "wear metal" numbers on an oil analysis, which aren't the same thing.

I've been told several times when talking to Redline Oil engineers who know the S54s well, that 10W60 was 'not the original factory fill' but it was a 'bandaid' and said; IF 10w60 was the fix for thin rod bearings then why do so many engines running it still spin their rod bearings(???)

👍

Oh and BMW oil = Penzoil

Shell owns Pennzoil (and Quaker State).

Lead and molybendum are both lubricants.... why would you want less? What defines better?

Lead isn't a lubricant. In this engine it's a wear metal.

Moly-based friction modifiers are good, though like any component of a lubricant, they do have downsides (e.g. ash, competition with anti-wear and detergent molecules).

Found this guy on who tests pretty much every oil ever made. Ravenol 10w-60 performed exceptionally well. His spreadsheets listed under the video are pretty interesting.

Amsoil wins? Who da thunk it... Penzoil/Twin Power had a respectable showing. Motul and Liqui Moly, not such a good showing, at least when it comes to 5w-30 weight oil specifically. Though 5w-30 is not 10w-60 oil, I would say this is at the very least, it’s a window, or indicator on how a manufacturer’s oil will perform on a different grade oil.

Unfortunately, these "tests" are basically meaningless for engine oil.

[Obioban]

Hmm. Saturday I put TWS in the M3 coupe, but then I put BMW/Shell 10w-60 in the M3 wagon and M5. Hard to know which way my vote falls in the era of FCP

[TexaZ3]

LM 10w60 in my E46, E39, and E63
BMW 5w30 in my E36 and M Coupe.

[Slideways]

I think terra posted this before on M3F, but the MSDS of BMW TwinPower 10W60 and Shell Helix 10W60 is the same. Definitely looks like the same product.

[SliM3]

Not recommended for stock obviously, but Valvoline VR1 Racing 20w50 full synthetic.

[Feffman]

Here's a JPEG conversion of my 10W-60 synthetic oil Excel spreadsheet with manufacturer published test numbers for as many synthetic oils as I could track down. I hope this helps. I'm happy to e-mail the Excel sheet to anyone.

Feff