Confessions of a Recovering Thickie

[dnewton3]

Out of curiosity, if viscosity does not matter for wear rates, what is your reason for solely running 5W-30 regardless of the specification?

Edit: To clarify, what benefit does the greater viscosity have when running extended OCIs, which you indicated is you case for the 5W-30?

I have several vehicles; some spec 5w-20 and some 0w-20 and one is 5w-30.
So I choose to have just one vis; 5w-30. I'm not worried about the minimal fuel economy loss.
My decision is simply that of convenience; one vis to stock in the garage.

 

[dnewton3]

I will disagree, respectfully, with some of what is said below ...
I use my training in statistical analysis and over 32k UOAs I've got in my database to make my argument.

UOAs aren't good for comparing wear between different oils.

They most certainly can be, but you've not delineated between micro and macro sample methodologies. I've harped on this hundreds of times over my 20+ years here.
I agree that a few samples cannot provide enough data to make a good conclusion.
Most certainly, if one had 30 samples of oil A and 30 samples of oil B, used in the same application, then UOAs can provide enough information to make a reasonable conclusion. The problem is that 60 (sixty!) samples represents hundreds of thousands of miles and hundreds of dollars of UOAs. Pretty much no one here has the time and money to do a PROPER statistical analysis of two or more lubes.

UOAs test the serviceability of the oil. After 100,000 miles of 2 oils (of very different quality and/or viscosity) in identical engines with identical use, both could show consistent 1 ppm/1k miles, but one of them could have far more piston deposits, ring coking, blow-by, oil consumption, seal degradation/leaks, sludge, and varnish than the other.

I would agree. But those other parameters you mention (deposits, etc) aren't measured by a UOA anyway. UOAs tell us two things:
- how is the lube holding up; this is a direct view of the oil's condition
- how is the engine reacting to the lube; this is an indirect view of the equipment's condition (taken as an inference)
The UOA can tell us about what comes out of the crankcase, but it cannot tell us about what stayed stuck to the internal parts. And absent of a tear-down of the engine, no other wear tool can tell you that either.

It could also have higher wear that won't show up in UOAs because of the range of particle size the ICP captures and some wear metals getting trapped in carbonaceous deposits as they form. Factors like magnets and bypass filters will skew results.

True as well. But UOAs are a tool which provides a partial view. When things are otherwise running well, the vast majority of wear is small stuff. If you've got metal particulate in the lube which is too large to be seen by a UOA, then you've probably got big problems afoot.

UOAs also cannot determine the source of wear. Copper could be bearing wear or could just merely be chelation from an oil cooler or brass fitting. Iron could be wear from rings, valvetrain, crank journals, etc... or could be from rust of an iron block.

I would agree. But UOAs can narrow the choices down, and with some investigative understanding of the equipment (such as does it have a cooler with Cu or not), you can make some educated inferences.

Other forms of analysis like ferrous spectroscopy could better determine these things but now you're getting well outside the scope (and cost) of a UOA.

And here, you've hit the nail on the head. UOAs are a very cost effective and easy tool to employ. Whereas other tools to measure wear do exist, they are WAY past the time/money/technology limits of most causal users. UOAs are not as telling as an engine autopsy; pulling out bearings and cams/cranks/pistons will certainly give a more accurate indication of wear. But who's got the time/money to tear down and reassemble an engine every 5k miles to check wear???? The answer is obvious ... no one. Other tools like ferrous spectroscopy are also way outside of the reach of Joe Average; as you acknowledge.

The only way you can accurately measure wear between 2 oils is with extensively controlled conditions on a dyno with before and after measurements with a profilometer for peaks and valleys on metal surfaces (particularly cylinder walls and rings), adcole machine for measuring cam wear down to a millionth of a inch, and so on. Simulation testing like Te-77 and SRV can give a good idea of that oil's performance, alongside rust, copper corrosion, and other tests.

Again - I disagree.
UOAs most certainly can be done to distinguish similarities and differences in wear. But the key is to employ the tool in the balancing act of time/money/effort.

The better way to use UOAs is to take your singular samples and compare/contrast them to Marco data norms. That way, you understand how well your one engine is fairing relative to well-established trends and variations.

Singular UOAs are NOT to be used to compare/contrast two products; that is bad methodology.
Singular UOAs can be used to judge how your equipment (engine, trans, diff, etc) is performing relative to the rest of its peers. That's where reasonable cost and effort meet up with sensible conclusions.


I offer this for a more detailed explanation and understanding:
https://bobistheoilguy.com/used-oil-analysis-how-to-decide-what-is-normal/

The bottom line is that UOAs can most certainly be used to measure wear, relative to two or more lubes. It's just that people don't generally understand the proper method to use the tool, and don't have the time/money to do it right. Rather than saying UOAs can't do the job, which is inaccurate to claim, it's more appropriate to say that most folks don't have the understanding and resources to do it properly.

Most simply stated ...
Don't blame the tool for the limitations of the user.

 

[ZeeOSix]

Piston rings live in a pretty harsh environment, especially the top piston ring due to the combustion heat. Below are results from an engine wear study showing how the oil viscosity impacted the top ring wear. Test was done at the Southwest Research Institute using an irradiated top ring and a radioactive tracer detector to measure wear rate in an actual running engine under different load conditions. A sophisticated controlled test like this tells a whole lot more than a $30 optical emission spectroscopy (OES) UOA.

 

[dnewton3]

Interesting, ZeeO6 -

Two things I would comment on (not that I have an argument to purport, just observations to note). The graph is interesting, but it's not really telling enough for me to understand some of the nuances which might affect the presumption of conclusion(s).

First, this chart would indicate that a 5w-30 would be a better choice for start/stop driving (stop/start, and stop/start very cold). I find that odd given that most hybrid engine cars call for thinner lubes due to the cyclic nature of their drive systems. This info seems to be counter intuitive to the commonly seen practice in most OE hybrid applications. What is the nature of the "stop/start" conditions in the test? Is this portion of the test when the engine is always running, and the test cycle is simply representative of the vehicle in urban traffic (starting and stopping the vehicle motion, but the engine is constantly running)? Or is this "stop/start" representative of the type engine operation when the engine is frequently stopped and restarted at each stop light? Or is it representative of the start/stop/start/stop cycles of a hybrid drivetrain?

Some of those tests are run at oil temps of 115C. That indicates the engine and lube are fully warmed. I would think that a hybrid-drive system with an engine and lube that isn't nearly as warm would perhaps give some very different results. There are a few tests which hint at the fact that cooler conditions may favor the thinner lube.



Second, this "top ring" test is obviously only looking for wear in one specific spot of the engine. There are bearing journals to consider, piston walls, cam surfaces, etc. This graph should be accepted as proof of only one location within the engine, not all of them.

My comments don't distract from the validity of results as shown on the graph; they are what they are.

Many hybrid engines are spec'd for 0w-8 or 0w-16 because the OEMs understand the real world operational temps may not favor the thicker lube. And maybe they have other tests which indicate the thinner oils are preferable for some specific expected conditions?

I want to point out to the novice/noob that these results cannot be applied across the board as a one-size-fits-all situation.
I would no sooner run a 0w-16 in a TT v-6 (say a 3.5L EB used for towing), than I would use a 15w-40 HDEO in a hybrid driven in urban city life.

 

[ZeeOSix]

The bottom line is that UOAs can most certainly be used to measure wear, relative to two or more lubes. It's just that people don't generally understand the proper method to use the tool, and don't have the time/money to do it right. Rather than saying UOAs can't do the job, which is inaccurate to claim, it's more appropriate to say that most folks don't have the understanding and resources to do it properly.

A history of UOAs on the same engine can eventually show some trends. A good example is the UOA history that @TiGeo had going when there was an internal engine issue going on. The UOA history did give an indication, but it was some pretty heavy wear going no before the UOA data gave an indication. The small wear differences between different oils is going to be pretty much in the noise of the UOA data. Maybe TiGeo can post the link to those that thread for reference. Other articles I've read on UOA also showed that by the time the UOA started showing increased wear that raised a red flag, the machinery was well into major wear to the point when opened up there were large wear debris particles all over the place, magnitudes larger than any UOA could detect.

Here's a graph that's been shown in a few other threads discussion UOA data. As can be seen, a optical emission spectroscopy type UOA can only see around 5u and less particle size, so when major wear starts to occur you only get a relatively small increase in wear metal in the UOA. This is basically what TiGeo saw going on in his UOA history when major wear was going on.

 

[ZeeOSix]

Second, this "top ring" test is obviously only looking for wear in one specific spot of the engine. There are bearing journals to consider, piston walls, cam surfaces, etc. This graph should be accepted as proof of only one location within the engine, not all of them.

I think SwRI did a top ring test because of the harsh environment that component lives in. Yes, all engine components react differently to a change in oil viscosity. And of course every engine isn't quite designed the same, so some engines may tolerate a thinner oil better then the next engine. But like mentioned many times in these thick vs thin threads, a higher HTHS viscosity is going to result in more film thickness between moving parts, and that's been the basic physical property of Tribology that helps prevent wear, and that never changes. I looked for the whole SwRI study on my computer, but didn't find it right now, might take some more digging around. So I don't know the exact details of the various engine test sequences. I think this study was also discussed in a few old threads here on BITOG, probably where I got the study PDF download link from.

 

[dnewton3]

A history of UOAs on the same engine can eventually show some trends. A good example is the UOA history that @TiGeo had going when there was an internal engine issue going on. The UOA history did give an indication, but it was some pretty heavy wear going no before the UOA data gave an indication. The small wear differences between different oils is going to be pretty much in the noise of the UOA data. Maybe TiGeo can post the link to those that thread for reference. Other articles I've read on UOA also showed that by the time the UOA started showing increased wear that raised a red flag, the machinery was well into major wear to the point when opened up there were large wear debris particles all over the place, magnitudes larger than any UOA could detect.

Here's a graph that's been shown in a few other threads discussion UOA data. As can be seen, a optical emission spectroscopy type UOA can only see around 5u and less particle size, so when major wear starts to occur you only get a relatively small increase in wear metal in the UOA. This is basically what TiGeo saw going on in his UOA history when major wear was going on.

View attachment 279015

That graph above speaks to particle size (presumably from a PC analysis) by %, but doesn't give any indication to understand composition.

ICP UOAs can discern composition, but not size.
PCs can discern size, but not composition.

I do generally agree that UOAs are not fool-proof and only tell a portion of the story of wear. But the same can be said of PC analysis, as well as other tools.

There are lots of anecdotal stories which can show any one analysis tool succeed or fail. There are plenty of UOAs which caught the onset of issues before catastrophic failure. There are also examples of UOA completely missing the mark.


I am aware of one such example going on right now, but I'm not at liberty to discuss due to respect for privacy. The ICP UOAs are indicating impending doom, but the vehicle is still serviceable and being operated. Yet because the vehicle is under warranty and concerns exist for the OEMs potential for investigation, the story is not being presented here on BITOG, yet ...

 

[ZeeOSix]

That graph above speaks to particle size (presumably from a PC analysis) by %, but doesn't give any indication to understand composition.

ICP UOAs can discern composition, but not size.
PCs can discern size, but not composition.

The gist of the graph was to show the particle size distribution associated with different levels of wear. That graph was out of another study. So when looking at that graph of just the particles 5u and smaller that a typical UOA can only see (note x-axis is a log scale), then the rise in the 5u particles and smaller is relatively low compared to the particles above 5u as the level of wear increases. The level of wear would have to be pretty large to see any significant up-tick in the UOA numbers. That graph could apply to any particles that the UOA can detect. Obviously someone should be tracking ppm/1000 miles to normalize the date and look for wear trends because the longer the OCI the higher the ppm counts even if the wear rate was constant.

 

[dnewton3]

I think SwRI did a top ring test because of the harsh environment that component lives in. Yes, all engine components react differently to a change in oil viscosity. And of course every engine isn't quite designed the same, so some engines may tolerate a thinner oil better then the next engine. But like mentioned many times in these thick vs thin threads, a higher HTHS viscosity is going to result in more film thickness between moving parts, and that's been the basic physical property of Tribology that helps prevent wear, and that never changes. I looked for the whole SwRI study on my computer, but didn't find it right now, might take some more digging around. So I don't know the exact details of the various engine test sequences. I think this study was also discussed in a few old threads here on BITOG, probably where I got the study PDF download link from.

I think one of the more accurate and poignant phrases from years past has been lost in the explosion of membership here on BITOG.
There is a wonderful concept regarding vis ... "As thick as necessary, as thin as tolerable." But that's only part of the story ...

For any given piece of equipment, in any given application, there will be a minimum safe threshold for a desired wear rate. That rate is supported not just by vis, but also by the add-pack and other considerations. Each characteristic of a lube doesn't operate in a vacuum irrespective of the other characteristics.

It's entirely conceivable that lube A (being of thinner vis), may outperform lube B (being of thicker vis) in terms of wear control, because of other considerations such as base stocks and anti-wear additives. When other things are held equal, increasing vis may (or may) not help. But it is unlikely to hurt (unless operational temps are subdued such as a hybrid).

Which is why these "thick vs thin" topics never have an end to them. It's silly to focus on one aspect of the lubricant when other factors are also in play.

I've repeatedly said for years that I don't care about what's in the bottle nearly as much as what comes out of the crankcase.

I can attest, with absolute certainty, that real world data (one cannot ignore over 32k UOAs) that once a reasonable wear rate is achieved, adding more vis doesn't show any correlation to reducing wear further.


And then there's filtration, but that's an argument for another thread ...

 

[ZeeOSix]

I think one of the more accurate and poignant phrases from years past has been lost in the explosion of membership here on BITOG.
There is a wonderful concept regarding vis ... "As thick as necessary, as thin as tolerable." But that's only part of the story ...

Yeah, but nobody has a real accurate way to measure wear so nobody knows for sure what "thick as necessary" and "thin as tolerable" really is except that their engine didn't blow-up or wear out at 100K miles. Going up a grade from xW-16 or xW-20 takes the guess work out of what those two statements mean. And of course going up a grade or two for extreme use cases like track use, etc is the way to go for obvious reasons.

The typical UOA is like looking through a straw - it's a pretty narrow view of wear. It's too insensitive to accurately see wear changes in the benign wear regime shown in the graph above - that's why people won't see much if any change in wear metals by just switching brands in the same viscosity. But put some 0-5W racing oil in a twin turbo and you're most likely going to see some wear metal increases in the UOA (was done and seen here on BITOG).

UOAs are useful if many routine UOAs are done on the same engine, and the wear tread uptick can clearly be seen when the wear starts increasing above the benign regime. The wear would have to deviate into the higher wear regimes before a UOA would start raising a caution flag. That's what the graph is showing for the paricles 5u and smaller that the UOA can only see.

I can attest, with absolute certainty, that real world data (one cannot ignore over 32k UOAs) that once a reasonable wear rate is achieved, adding more vis doesn't show any correlation to reducing wear further.

A bit more viscosity is about added wear protection headroom. Why possibly operate on the ragged edge of zero MOFT depending on operating conditions when you can add some film thickness between moving parts to give some added wear protection.

Re: Filtration ... cleaner oil results in less wear. Just like more film thickness results in less wear. Go for both for the win.

 

[dnewton3]

A bit more viscosity is about added wear protection headroom. Why possibly operate on the ragged edge of zero MOFT depending on operating conditions when you can add some film thickness between moving parts to give some added wear protection.

I agree in principle. But the topic gets distorted when the 'Murican concept of "more is always better" comes into play. Some headroom is good. More headroom isn't proven to be any "better". I am going to limit my comments to normal engines operating in DD conditions; we're ignoring racing applications, abused engines, etc.


Example 1:

Thread 'Ford 3.5L at 31k miles / SuperTech 5w-20 at 11.8k miles'

Dec 17, 2021

Here's the first UOA on my 2018 Taurus; naturally aspirated 3.5L Cyclone v-6.
Mix of city/hwy driving. Was in the crankcase for about a year.

Veh Miles 31k
OCI miles 11.8k 
filter TG
oil brnd ST
oil grade 5w-20
oil base conv

Alum 5
Chromium 0
Iron 9
Copper 6
Lead 0
Tin 0
Moly 39
Nickle 0
Manganese 3
Silver 0
Titanium 0
Potassium 0
Boron 42
Silicon 38
Sodium 3
Calcium 1396
Magnesium 435
Phos 758
Zinc 876
Barium 0

Vis at 210F 54.0
Vis at 100C 8.49
Flash Point 440
Fuel  Antifreeze 0
Water 0
Insolubles .2

As you can see, conventional ST 5w-20 is more than up to the task. Nearly 12k...

• dnewton3
• Replies: 68
• Forum: Used Oil Analysis - Gasoline

• 
• 

I ran conventional 5w-20 in that Cyclone v-6.
I've got over 600 UOAs where other folks have run anything from syn 0w-20 to 10w-40 in that engine series. A 20 grade has shown that, in macro data, wear rates are VERY desirable. Would a 16 grade work? Maybe ... maybe not. But what is known with total clarity is that thicker lubes (above a 20 grade) don't reduce the wear rates; not one bit. Once the reasonable wear rates are achieved (grade is proven), moving up the vis scale has no discernable effect.


Example 2:

Thread '5w-20; Ford 4.6L engines; UOA testing'

Apr 26, 2017

Here are two UOAs from our cars at home. These are both 4.6L MGM cars; one an '05 and the other an '07, both identically equipped. My wife and I both drive nearly identical routes to work; about 32 miles each way and we only diverge in the last mile (she goes east and I go west). Because she occasionally works from home, I even drive her car to balance out the miles. One was run on dino oil; the other car was run on synthetic.

Tell me, if you can, which sample is the syn and which the dino oil? Which is a brand-name lube and which a house brand? Tell me, if you think you can discern from...

• dnewton3
• Replies: 53
• Forum: Used Oil Analysis - Gasoline

Same concept ... I ran 5w-20 in the 4.6L engines we had.
I have nearly 800 UOA on this engine series. People run anything from a 20 grade to a 40 grade in these engines. Doesn't matter one single iota. There is zero proof (none; zip; nada) that indicates thicker lubes reduce the wear rates.



I'm not saying this is always true; unique applications can always show an anecdotal failure or concern. But macro data will indicate, with a high degree of certainty (95% CI), what minimum vis is warranted. Once that determined vis gives generally excellent wear, and someone wants to jump up a grade, that's fine. But there's often no proof that it's going to reduce wear; in fact, there's proof of just the opposite. What the data really proves is that the "headroom" is already present.

How much headroom is needed? If a 20 grade gives very good results, then is a 30 grade "better". And if that 30 grade is "better", is a 40 grade "more better"? And if a 40 grade is "more better", is a 50 grade "many more better" ???? The clear answer is "NO!"

My point isn't that a 20 grade is always OK. That's not what my data indicates. What I'm saying is that macro data will indicate what grade will give very desirable results, and anything thicker doesn't really show any statistically significant improvement.

The increase in vis shows no correlation once the minimum threshold of desirable wear rate is achieved.
And without correlation, there can be no causation.

Just like with filtration, vis only has an effect to a certain point, and then after that desired effect is achieved, adding "more better" doesn't mean much at all.

 

[D1dad]

My grandad was in WWii, stormed a beach, became an engineer for the railroad after laying the track from WV to Ann Arbor where he settled. He eventually became a diesel mechanic and could just about fix, or rebuild anything that moved. He had a 78 bonnevile that he took to almost 500k, which was unheard of back in that time. TOTAL RUST BUCKET. I can still hear him preaching to me in the garage…”clean air, clean oil and good fuel” pretty much what I do to this day. I replace oil and air filters at 1/2 of the manufacturers recommendations minus my 18 which only calls for 5k on the oil. I give away my Kroger fuel points because it’s not top tier. If a UOA reveals anything bad, if I ever decide on one, I’m blaming the manufacturer.

 

[TiGeo]

A good example is the UOA history that @TiGeo had going when there was an internal engine issue going on. The UOA history did give an indication, but it was some pretty heavy wear going no before the UOA data gave an indication. The small wear differences between different oils is going to be pretty much in the noise of the UOA data. Maybe TiGeo can post the link to those that thread for reference. Other articles I've read on UOA also showed that by the time the UOA started showing increased wear that raised a red flag, the machinery was well into major wear to the point when opened up there were large wear debris particles all over the place, magnitudes larger than any UOA could detect.

Here are my data. Background was that in early 2022 I heard a noise when starting my car one morning that was the timing chain tensioner bolts shearing off. The tensioner got lodged between the lower timing cover and the block and fully extended to maintain tension on the chain. I had no idea beyond the noise I heard was at the time, but in hindsight it sounded as you'd expect (chain rattling against a metal pan). I would get the noise on start up occasionally. I continued to drive the car that year including multiple drag race events and a trip to the GA mountains for a VW event that included the obligatory Tail of The Dragon runs etc. The next oil change that May Blackstone noted an increase in iron beyond the "normal" levels for my car - indicator of the issue in UOA. I also noted excessive metal "glitter" in my oil filter when I cut the filter open after that change based on a hunch - indicator of the issue. Kept driving. Late that summer before a trip to the track, the noise got more consistent and I took it to my shop where they found the issue and repaired it thankfully before full failure which would have caused catostrophic engine damage.

One note on Si in the UOAs below, for all HPL data, I backed out 8ppm SiO2 (anti-foaming) based on my VOA data to normalize it against the other data.

For the table below, orange highlighted data are not included in the stats due to break-in. Red ("bad")/green ("good") highlight are below/above one SD depending on the analyte/variable. For rate/1K miles, I used the LSJR 5ppm/1K as a benchmark for red ("bad") bold font. Should be self-explanatory for the brain-trust here on BITOG.

 

[Samurai1225]

My grandad was in WWii, stormed a beach, became an engineer for the railroad after laying the track from WV to Ann Arbor where he settled. He eventually became a diesel mechanic and could just about fix, or rebuild anything that moved. He had a 78 bonnevile that he took to almost 500k, which was unheard of back in that time. TOTAL RUST BUCKET. I can still hear him preaching to me in the garage…”clean air, clean oil and good fuel” pretty much what I do to this day. I replace oil and air filters at 1/2 of the manufacturers recommendations minus my 18 which only calls for 5k on the oil. I give away my Kroger fuel points because it’s not top tier. If a UOA reveals anything bad, if I ever decide on one, I’m blaming the manufacturer.

FYI.. Kroger fuel points work @ Shell stations

 

[buster]

UOA is good for trending and viewing results in a holistic manner. Even if you had a case where oil (A) showed 15ppm of Fe higher than oil (B) after 50k miles, you couldn't ultimately determine if the oil (A) was allowing for more wear. You'd have to tear down the engine. In theory you could say it was giving up more wear, but you could also say oil (A) may be controlling deposits better. You could also ask the question, would a difference of 15-30ppm of Fe over the lifetime of the engine even ever be noticeable in service?

Something I learned recently was the TAN/TBN crossover. It's not as simple as condemning the oil if TAN>TBN. Goes back to the overall health of oil - oxidation, wear metals, viscosity etc.

 

[kschachn]

UOA is good for trending and viewing results in a holistic manner. Even if you had a case where oil (A) showed 15ppm of Fe higher than oil (B) after 50k miles, you couldn't ultimately determine if the oil (A) was allowing for more wear. You'd have to tear down the engine. In theory you could say it was giving up more wear, but you could also say oil (A) may be controlling deposits better. You could also ask the question, would a difference of 15-30ppm of Fe over the lifetime of the engine even ever be noticeable in service?

Something I learned recently was the TAN/TBN crossover. It's not as simple as condemning the oil if TAN>TBN. Goes back to the overall health of oil - oxidation, wear metals, viscosity etc.

Or worse, you’re blaming the oil when in reality there are other uncontrolled variables which might cause the increase. All the way from improper ICP calibration to a leaky air filter to human error in sampling.

This is the base problem with real-world UOA and making grand pronouncements about cause. It’s not just the oil.

 

[D1dad]

FYI.. Kroger fuel points work @ Shell stations

No shell stations in my town. There were, but they were bought out and turned into Marathon. I did use them when they were shell but could only use 10 cents at a time. I give them to my son and he uses them for diesel and his contractor business.

 

[ZeeOSix]

I agree in principle. But the topic gets distorted when the 'Murican concept of "more is always better" comes into play. Some headroom is good. More headroom isn't proven to be any "better". I am going to limit my comments to normal engines operating in DD conditions; we're ignoring racing applications, abused engines, etc.

I'm not saying this is always true; unique applications can always show an anecdotal failure or concern. But macro data will indicate, with a high degree of certainty (95% CI), what minimum vis is warranted. Once that determined vis gives generally excellent wear, and someone wants to jump up a grade, that's fine. But there's often no proof that it's going to reduce wear; in fact, there's proof of just the opposite. What the data really proves is that the "headroom" is already present.

How much headroom is needed? If a 20 grade gives very good results, then is a 30 grade "better". And if that 30 grade is "better", is a 40 grade "more better"? And if a 40 grade is "more better", is a 50 grade "many more better" ???? The clear answer is "NO!"

My point isn't that a 20 grade is always OK. That's not what my data indicates. What I'm saying is that macro data will indicate what grade will give very desirable results, and anything thicker doesn't really show any statistically significant improvement.

The increase in vis shows no correlation once the minimum threshold of desirable wear rate is achieved.
And without correlation, there can be no causation.

Just like with filtration, vis only has an effect to a certain point, and then after that desired effect is achieved, adding "more better" doesn't mean much at all.

To know all of this everyone would basically have to do a long controlled UOA tracking history to ensure they are trying to run oil that is "thin as possible" while still ensuring "acceptable wear rates". Based on what I know about Tribology and lubrication, I elect to go up a grade on anything that is specifying a xW-20 or less. Even Ford and a few others decided to now go thicker on some of their engines, and it obviously wasn't for anything positive associated with CAFE. That does ensure better film thickness and added wear protection.

All someone has to do grab the jug off the self, and then it's not necessary to do a controlled UOA self test program to prove to themselves that going up a grade is good or not. Every wear study comes to the conclusion that more viscosity creates more film thickness between moving parts and that reduces wear - viscosity is the backbone of Tribology ever since the science was discovered. It doesn't matter to me if I have more or "too much" wear protection headroom above possibly running on the ragged edge of zero MOFT and causing more wear in some operating conditions.

 

[Silverado12]

Someone mentioned top compression ring wear. I'm thinking that the wear metal from the top ring might blow out the exhaust rather than be detectable with an oil analysis. FWIW I'm using 5W-30 in my wife's Jeep 2.4 Multiair Cherokee which calls for 5W-30. It definitely quiets down the motor. I will switch to the same for my 2023 Cadillac 2.0 Turbo motor after the warranty runs out. I only use the NAPA branded oil.

 

[ZeeOSix]

Here are my data. Background was that in early 2022 I heard a noise when starting my car one morning that was the timing chain tensioner bolts shearing off. The tensioner got lodged between the lower timing cover and the block and fully extended to maintain tension on the chain. I had no idea beyond the noise I heard was at the time, but in hindsight it sounded as you'd expect (chain rattling against a metal pan). I would get the noise on start up occasionally. I continued to drive the car that year including multiple drag race events and a trip to the GA mountains for a VW event that included the obligatory Tail of The Dragon runs etc. The next oil change that May Blackstone noted an increase in iron beyond the "normal" levels for my car - indicator of the issue in UOA. I also noted excessive metal "glitter" in my oil filter when I cut the filter open after that change based on a hunch - indicator of the issue.

That's a good example of when wear becomes much higher than the "benign" realm in the graph in post 125, and a UOA only seeing 5u wear particles and below won't give a good indication of a problem until the wear is pretty extreme where there's also a bunch of metal glitter in the oil filter. It was easier to see because you had a lot of UOA history on the engine. If there was no mechanical noises with a wear issue like that going on, and someone just did a random UOA or two, they may not have even been that alarmed that something was really going on because they didn't have enough UOA history on the engine to see it.