Viscosity: Engine protection Vs. enhanced mileage

[Carlostrece]

Warranty repairs start with dealer or Carfax OCI. They don't have a way of knowing brand/viscosity unless they take your manual receipts to them. They will then have to prove the viscosity or oil brand was the cause of the failure.

What if you are a DIY oil changer? Are they going to ask for your Walmart or NAPA receipts? Carfax won't know what you purchased at Walmart, Amazon, or autostores.

In any case, they won't be able to prove that going from Xw20 to 5w30 caused a problem because it won't. Up one grade isn't going to cause a lubrication problem, but it might prevent one.

 

[Carlostrece]

One reason I like to go up one grade is safety margin. My car is driven easy, but what if a radiator hose blows and we're in heavy traffic and can't pull over for a while? Those things can happen. If fact, that did happen to one of the Jeep 4L that I owned.

I had to keep driving (on a hot day with no coolant) until there was a safe place to pull over. No engine damage occurred. In that Jeep, I had previously stepped up the oil from 10w30 conventional to 5w40 synthetic which might have saved the day.

Even if you drive easy and your climate is moderate, unexpected things can happen. Exceptionally hot days can happen. Blown radiator hoses can happen and you might not be in a safe place to pull over. So then you're driving for a while with no coolant.

At that point you'll be wishing you had one grade higher oil in it.

 

[WeyounDS9]

My 2022 Lexus ES350 calls for 0w – 20 oil in the 3.5L V6. The manual does say that if 0w – 20 is not available, 5w-20 may be substituted, but it must be replaced with 0w-20 “at the next regular oil change.”

To protect my remaining warranty I continue using Pennzoil Ultra Platinum 0w-20, but I wince every time I pour that watery-looking stuff into the engine.

It is my gut feeling that 0-20w oil is specified solely on the basis of delivering optimum fuel mileage, and that 5-20w (or even perhaps 5w-30 in a warm climate) would likely provide better engine protection and longer service life under "normal," commuter car use.

I live in North Florida; the engine “might” see a cold start near freezing temperature 2 or 3 days a year. Long summers are HOT. I buy my cars new and keep them for a very long time so optimizing service life is paramount to me. I change oil/filter at 4000 mile intervals and am a very conservative driver.

While these are the usage parameters for my particular vehicle, but I suspect my general question applies to many late model vehicles with very lightweight oil recommendations.

Since my gut feelings are worth about as much as a noisy stomach growl, does the empirical evidence suggest compromises have been made in engine protection in favor of boosting fuel efficiency?

This is a discussion we have all the time here.
And there are 2 camps.
Those who follow the evidence ( which is that thicker oils can establish a better film to protect moving parts) and those that observe the party-line crafted to allow manuafactuers to use thin oils in EPA mpg tests, (which are biased for short trips)
A 5w-40 would be a fine choice for your engine year around.

 

[ChrisD46]

Your engine can't differentiate between the 0 and the 5. If your oil of choice meets the spec in your manual, you are good to go. If that were my 3.5, it would be seeing 5-30, but that's just me. I'm not a must-use thicker oil person, but if the engine allows, I would. That's the best way to explain it.

To add , what does the Lexus Australian version , Malaysian , Saudi Arabian version or some other warm countries owners manual call out for the same engine ? If 5W30 is allowed in other warm temp countries then that is what I would be using . If you have any concerns , you could move up to PPPP 5W30 which is a thin 30 weight of 9.6 cSt .

 

[rstcso]

What if you are a DIY oil changer? Are they going to ask for your Walmart or NAPA receipts? Carfax won't know what you purchased at Walmart, Amazon, or autostores.

As a DIYer, Carfax allows you to go into their system and add records. You can upload receipts, make notes, etc. This becomes part of the vehicles record. You can also edit or delete records.

 

[rstcso]

This is what the Carfax dashboard looks like. You can change the intervals to match your planned service schedule. I haven't changed the "favorite shop", so it's showing the dealership where it was purchased.

Carfax will email you if a recall is found on your vehicle. When I had the Audi, I learned of the recall days ahead of Audi even showing it on any of their websites. It was the second engine recall for my car, so I dumped it and got the Honda.

 

[rstcso]

I also enter DIY data into Honda's MyGarage. When I need to take it in a Honda dealer for servicing, their system will pull up these records in addition to ones performed at a dealership.

 

[trd1zz]

What if you are a DIY oil changer? Are they going to ask for your Walmart or NAPA receipts?

Yes, they could. It's pretty standard warranty language that service receipts / invoices need to be kept demonstrating maintenance was followed per factory recommendations.

Put it a different way. Would you buy a used car with over 100k miles from an owner who had zero maintenance records? No receipts, no spreadsheet with dates of service / mileage at service / oil & parts used? And then when you ask about the maintenance they just said "oh, I DIY'ed all of the fluid changes, don't worry about it".

I wouldn't.

 

[rstcso]

Yes, they could. It's pretty standard warranty language that service receipts / invoices need to be kept demonstrating maintenance was followed per factory recommendations.

I'm typically not the best record-keeper, but in one of Car Wizard's YouTube videos, he had a customer whose car had over 200k miles in less than two years. The customer kept a three-ring binder with receipts for everything ever done to that car related to maintenance.

I started doing that with the Audi and now the Honda. When I order something online, e.g. Amsoil (switching to HPL), filters, crush washers, etc., the emails showing the items, purchase dates, prices, etc. gets printed, I'll write the mileage and other notes as I think may be helpful. then put the printouts in "The Book" in chronological order. If something is purchased OTC, I make a copy of the receipt to go in the book and put the original receipt somewhere dark where hopefully it won't fade. I also take pictures of the odometer and the products before installing/changing. The EXIF metadata can prove when the pictures are taken (yes, it's possible to edit this data ). I've become rather retentive about keeping this book updated.

 

[kschachn]

As Carlostrece points out in post #20 above, PUP is on the very thin side of a 30 grade. If I don’t see a meaningful decrease in wear metals with this -30 grade, to pursue a tiny bump in fuel economy I will consider switching down to a 5w-20, perhaps to a different, premium formulation with viscosity on the higher end of the -20 grade versus the PUP. After an oil analysis on that third option, I should have enough comparative data to decide on a long-term product that best meets my vehicle needs and driving conditions.

I realize this whole, exploratory process is picking nits. Any premium, 0w-20 on 4000 mile drain intervals isn’t going to trash my engine; I just wonder if it is the very best I can do for long term reliability. My other vehicle is a 20-year-old, Town and Country minivan I’ve had since new. Currently using 5w-20 Pennzoil Platinum High Mileage, it’s never required any internal engine work, it runs like a top and consumes very little oil. As I said, I keep ‘em for a long time so expending the effort to find the best engine protection possible seems worthwhile.

As much as you may wish to believe it, a $35 spectrographic analysis isn't going to be able to discriminate what is the "best oil" for your engine. It's nowhere near the proper tool for generate comparative oil quality data - especially in the manner you're using it. You're not even close.

 

[Nessism1]

This is a discussion we have all the time here.
And there are 2 camps.
Those who follow the evidence ( which is that thicker oils can establish a better film to protect moving parts) and those that observe the party-line crafted to allow manuafactuers to use thin oils in EPA mpg tests, (which are biased for short trips)
A 5w-40 would be a fine choice for your engine year around.

As long as the oil is thick enough to avoid metal to metal contact, anything thicker is not adding value. In fact, it can be a detriment.

 

[Patman]

Yes, they could. It's pretty standard warranty language that service receipts / invoices need to be kept demonstrating maintenance was followed per factory recommendations.

Put it a different way. Would you buy a used car with over 100k miles from an owner who had zero maintenance records? No receipts, no spreadsheet with dates of service / mileage at service / oil & parts used? And then when you ask about the maintenance they just said "oh, I DIY'ed all of the fluid changes, don't worry about it".

I wouldn't.

I actually would. With most of the used cars that I have bought from dealers I never really get any kind of service records at all. I’m going by the condition of the paint and the interior to see if the car has been taken care of or not. That’s my biggest concern. I think it’s unusual to find a higher mileage used car in mint condition that would have been neglected on the maintenance side.

 

[Sunshine Guy]

As much as you may wish to believe it, a $35 spectrographic analysis isn't going to be able to discriminate what is the "best oil" for your engine. It's nowhere near the proper tool for generate comparative oil quality data - especially in the manner you're using it. You're not even close.

Okay, fair enough, but can you guide me then towards a better, more scientific process? Again, my goal is to make the right oil VISCOSITY choice for maximum engine protection based on facts rather than opinion.

I was thinking of the oil analysis showing me 1) how badly the oil has sheared down or otherwise meaningfully deteriorated in its protective ability, and 2) how much wear metals are in the oil. I figured the $35 oil analysis would shed light at least on those simple analytics and thereby help suggest whether a step up in viscosity might actually reduce engine wear.

I realize I’m probably both ignoring some factors and making some assumptions, but it seems there should be a better way than flipping a coin to decide whether to obey the CAFE-driven, 0W-20 factory mandate or move up to a 5w-30…probably PUP 5w-30 because, as others here have pointed out, is on the lower end of the viscosity scale for that grade.

I am now ruling out a switch to 5w-20 based on advice here that it would provide no meaningful difference in protection for my engine and its usage. So, at least I’m a troll with ears…LOL. I really am here to learn.

My thinking was, if an oil analysis showed no improvement in wear metals with a switch to 5w-30 PUP, I might consider trying and analyzing the 0w-20 QS Ultimate Protection Synthetic recommended previously in this thread because it is considerably higher in actual viscosity than the PUP 0w-20.

The only reason I’d do that (versus just sticking with PUP 5w-30) is to stay with the factory recommended weight. I care for my cars very well and always (quickly and easily) dispose of them in a private sale. I have the thought that a future buyer of this car years down the road might worry that I strayed from Toyota’s mandate to use 0w-20. I suppose most people wouldn’t even think to ask beyond oil change intervals, although some (sad) people are like me and might worry about it. But if the choice is between a future buyer’s potential worries and better actual engine protection, I adamantly choose the latter.

Recalling the facts and some very reasonable assumptions:

1. I drive quite conservatively (in the Lexus anyway) and never tow or carry a heavy load
2. Car is almost never started below freezing temperature; North Florida climate
3. I change oil and filter every 4000 miles which is also within about 6 months
4. I use Pennzoil Ultra Platinum which most would agree is at minimum a “good” oil
5. I have never observed any increase in oil level that would suggest fuel dilution

Regarding viscosity, I better understand now that there seems to be here a thicker camp, a thinner camp, and perhaps a makes-no-meaningful difference camp. So, if there is nothing in a $35 oil analysis that would help with choosing the best viscosity, is there a better way to make this decision other than “picking a camp?”

 

[JAG]

As long as the oil is thick enough to avoid metal to metal contact, anything thicker is not adding value. In fact, it can be a detriment.

Metal-to-metal contact is unavoidable. That’s what makes engine tribology interesting, challenging, and frustrating.

 

[trd1zz]

I actually would. With most of the used cars that I have bought from dealers I never really get any kind of service records at all. I’m going by the condition of the paint and the interior to see if the car has been taken care of or not. That’s my biggest concern. I think it’s unusual to find a higher mileage used car in mint condition that would have been neglected on the maintenance side.

Buy a higher mileage used car from a dealer? No thanks. Private party sales only. And if the seller doesn't have any maintenance records I'll just wait for the next one to come along that does.

The only car I'm buying from a dealer is either brand new or CPO'd.

 

[Patman]

Buy a higher mileage used car from a dealer? No thanks. Private party sales only. And if the seller doesn't have any maintenance records I'll just wait for the next one to come along that does.

The only car I'm buying from a dealer is either brand new or CPO'd.

I guess I’m just super lucky because I have never had a bad car that I bought from a dealer

 

[trd1zz]

I guess I’m just super lucky because I have never had a bad car that I bought from a dealer

Could be.

Dealers will take anything on trade/sale. That game is a race to the bottom. "We'll beat any written offer!" With many places you are lucky if they even give it a once over in their service department and maybe a wash before attempting to resell. Or they'll just wholesale it off.

 

[Patman]

Could be.

Dealers will take anything on trade/sale. That game is a race to the bottom. "We'll beat any written offer!" With many places you are lucky if they even give it a once over in their service department and maybe a wash before attempting to resell. Or they'll just wholesale it off.

My current Civic is so far the best used car purchase ever. I paid $11,400 ($8300 US) for it in March 2021, with 80,000 miles on it and it hasn’t cost me a single dime in repairs. It just keeps plugging along reliably and averaging close to 50 MPG as well! My previous Civic (an 06) was also incredibly good, I bought it with 80,000 miles on it too, for $9,300 ($6800 US) and drove it for ten years. I had to change the AC compressor so that was a $1000 repair but the only other repair was an oil pressure sending unit for $200. And going one step further, the 98 Corvette that I bought used in 2004 only cost me about $2500 in repairs in the 9 years that I had it.

 

[Hohn]

This is a discussion we have all the time here.
And there are 2 camps.
Those who follow the evidence ( which is that thicker oils can establish a better film to protect moving parts) and those that observe the party-line crafted to allow manuafactuers to use thin oils in EPA mpg tests, (which are biased for short trips)
A 5w-40 would be a fine choice for your engine year around.

The official weighting in the EPA test cycle is 55% city, so not hugely biased towards city driving.

But, based on the description of the elements of the test cycle, major portions of the city cycle are with cold oil. So the OEMs are often targeting the viscosity at oil temps significantly less than full hot-day-equilibrium.

I'm certain that the exact nature of this test cycle figured *heavily* in the final selection of oil viscosity grade and that OEMs aren't too concerns whether the engine wears out in 150k miles or 400k miles. They know that the real world variability is so large between the best and worst owners and duty cycles that it's almost impossible to ascribe engine life (or the lack of it) to a particular oil viscosity grade being used or not used.

These debates on viscosity only matter really in the rarefied air of BITOG. The OEMs are just running hurdle tests and if it passes, it passes. Most engines can tolerate a VERY large range of viscosity-- as they must, since oil viscosity is so extremely variable with temperature ranges that are normal for every engine.

Lots of long-lasting engines have been run on thin oils as well as thick oils. The famous example of the man who put a million miles on each of two different Tundras did so using TGMO 0w-20 on 10k change intervals suggests that there are conditions under which a "thin" oil is perfectly sufficient.

Engines wear out with thicker oils and thinner oils alike, just for different reasons.

Since engines vary so wildly in factors like how they heat oil and how much they shear it in different areas, what's "thin" for one engine might not be very thin at all for another. Film loads are not the same for all engines.

It's possible a 0w-8 is perfectly acceptable in actual running viscosity if the engine is 1) engineered with very small clearances in bearings, 2) runs a lower oil temperature of 70C or so, and 3) is designed with low film loads. Low film loads mean things like larger bearing area relative to load. Or lower cylinder pressure for a given piston ring diameter-- that kind of thing.

Note the very thin 0w-8 oils are only used in hybrids with very low average oil temperatures and frequent start/stop where speed of oil pressure arrival is perhaps more important than a point or two of viscosity.

 

[Sunshine Guy]

Great points made by Hohn in the previous post #119!

I previously recommended the video linked in post #4 of this thread. For those like me who are really interested in this viscosity subject, I’d like to add an additional video recommendation, here:

It is by the same person as the previous video but deals more specifically and deeply with the “thick versus thin” oil controversy. He demonstrates and explains much of the same science already mentioned by knowledgeable members in this thread. However, in my opinion, he does a particularly good job of tying it all together and suggesting possible, empirically-based conclusions that I and probably many others are seeking.

One of those several conclusions is that you shouldn’t stray from your manufacturer’s recommended oil viscosity unless you have a specific reason to do so. After thinking about it, my reason would be this: I drive my 3.5L Lexus very conservatively, very light throttle 99% of the time.

When cruising down the highway in pancake-flat Florida, this engine is doing a tiny amount of work; it is grossly under-stressed. I have to believe the recommended 0w-20 is fine for this as the rpm is somewhere near at least 2000 rpm and the load is next to nothing. Brisk bearing speed and low load pushes the Stribeck curve (which for some strange reason everyone seems to think a newbie can’t understand) to the right, assuring optimum lubrication.

As for MY REASON to possibly move up in viscosity, I do a lot of stop and go, local/suburban driving. With my light throttle inputs, during acceleration the engine is pulling most of the time between 1100 – 1900 rpm. Under light throttle, the electronic controller for the 8 speed automatic is more than happy to skew its shifts towards very low rpm, max-gas-mileage shifting. (It works; I get a great gas mileage from this full-size, V6 luxury barge…around 25-26 mpg city and 37-38 mpg highway).

So, I think,…actually I can feel…that the engine is pretty heavily loaded when accelerating from every stop and mostly staying well below 2000 rpm. It does this smoothly as silk, but a V6 engine has to be generating some serious cylinder pressures to accelerate a large sedan at this very low rpm. So, this is one scenario where I feel the heavier viscosity might be warranted, i.e., frequent low rpm operation is shifting the Stribeck curve to the left.

Carrying that conclusion even further, one might go so far as to suspect it might possibly even be easier on the engine if I had a “slightly” heavier throttle foot which would prompt the transmission to allow the engine to rev a little more but work with slightly less cylinder pressure…although this might put more wear on the transmission and other driveline components (NONE of which is likely to ever be proven one way or the other).

Bottom line, I think...counter-intuitively...that moving up slightly in viscosity might be warranted for a light-footed driver who does a lot of stop and go driving.