We got to work with what's available, and help ourselves. We could also do the easy thing and just follow the OLM, recommendations in the manual and pay the price later of course. Or go a bit thicker (3.5cP always seemed a good minimum viscosity to me), maybe a bit more frequent, more shear stable and more resistant to oxidation and hope it pays in the end.
What if "low tension ring" theory is all wrong?
- Thread starter Hohn
- Start date
We got to work with what's available, and help ourselves. We could also do the easy thing and just follow the OLM, recommendations in the manual and pay the price later of course. Or go a bit thicker (3.5cP always seemed a good minimum viscosity to me), maybe a bit more frequent, more shear stable and more resistant to oxidation and hope it pays in the end.
Yes. When I asked how to minimize wear with repeated restarts, it basically said "20w-50 if you can't find 30w-80" LOL. Kidding aside, the answer depended on how long you were shut down, how viscous the oil was, and how cold the ambient temperature was.
In my particular hypothetical, a 20w-50 minimized restart wear in 50F ambient temps as long as the period of shutdown was <60min. Longer than that, and the drainback was bad enough that the residual oil film wasn't sufficient to offset the restart and re-formation of oil films.
So: colder ambients and longer shutdowns slightly favor thinner (40 grade) oils. Hot ambients and shorter shutdowns favored the thickest stuff you could handle.
What was stunning to me is how small and how ephemeral was the restart advantage of thinner oils. It takes a very special duty cycle for even a premium (HPL) 20 grade to come out ahead.
Heck, I prompted with "what if my duty cycle changes to have tons of cold starts and a fraction of total usage?" and the AI Rudnick basically begrudgingly said to go as thin as Mobil 1 0w-40FS. But even then, the wear prevention advantage of the 0w in a duty cycle dominated by cold starts was slim vs a 15w-40. I'll try to find the actual model forecast. I want to say it was 6% less wear using 0w-40FS vs HPL 15w-40 if I had 3x or 4x more cold starts than I do presently.
Found the hot restart sensitvity discussion. I asked the AI to build a wear rate model based on Dr Rudnick's dicussion in Ch 11. Then I asked it to iterate that wear rate model based on the number of restarts and show me when the thinner Mobil 1 0w-40FS was advantageous and by how much relative to HPL 15w-40.
Last edited:
That would likely be a negative ghostrider. Although the 2.0 ecoboost in my wifes fusion has just about 200k trouble free miles on it, it's lived a life of 5-10k OCI's with really good oils. Ford did leave the eco portion of ecoboost off in reality though as the 3.0 V6 that the ecoboost replaced gets FAR better mileage in city and only marginally worse on the hwy despite the more advanced aerodynamics on the newer model.
I have similar results from the 6.2 in my F150 vs my FIL's 3.5 ecoboost (14 vs 13 model years. Both supercrew 4wd). We would get the same mileage in town but my 6.2 actually gets better on the hwy than his 3.5eb. It takes the same amount of power to move the same vehicle down the road. So far from owning 3 different DI vehicles, they've all underperformed in fuel economy compared to rated values. The port injected vehicles, otoh, have generally met or exceeded their hwy rated mpg's. There must be something with the rolling road dynamometer used for the government standards.
See my post here for some information on this. 100 cST is more than thick enough to prevent hot restart wear. 15 to 20 cST is not thick enough. A thicker grade will help a bit, but even a 20W-60 won't completely eliminate hot restart wear when the oil is at operating temperature.
An example of the comparison of different oils relative wear vs number of hot restarts per year:
Sensitivity Analysis: Impact of Increased Hot Restarts on Wear Model (HPL PCMO Grades Only)
Ramping hot restarts in your K20C4's duty cycle still pivots wear toward boundary/mixed events, where sump drain-back and rapid film recovery during cranks (~5-10s in your light-accel protocol) expose turbo journals and rings to asperity risks—Rudnick's models in *Lubricants and Lubrication* (Ch. 11) apply a 1.5-2x multiplier here, as the fixed-gear pump's initial ~15-25 psi struggles with partial starvation. Your baseline (~725 hot starts/year) balances at ~65% transient weighting; 5x daily (~1,825) surges to ~80%, eroding hydrodynamic advantages in sub-atmospheric loafing.Constraining to HPL PCMO grades (shared multi-synthetic/ester base with identical >13.5 TBN, Mo/borate AW, anti-LSPI detergents per PDS), I'll compare your 15W-40 (KV100 14.36 cSt, HTHS 4.213 mPa·s, CCS 6,268 cP @ -20°C) to the 10W-40 (KV100 14.36 cSt, HTHS 4.062 mPa·s, CCS 6,784 cP @ -25°C—as the "flow" analog with marginally better hot recovery via lower HTHS) and 20W-50 (KV100 20.11 cSt, HTHS 5.468 mPa·s, CCS 9,172 cP @ -15°C—for thicker hydro protection). Rudnick's additive uniformity (Ch. 6) ensures no chemistry clashes; viscometrics drive deltas, with lower HTHS aiding transient priming (~10-15% faster fill at operating temps) but conceding steady-state films.
Refined model (Archard-Stribeck: wear ∝ load × cycles / (hardness × h_film), η logarithmic from KV100/HTHS per Ch. 7; hot wear/start = 0.25 base × (1 + 0.4 × (HTHS - 3.5)) for recovery penalty; hydro = 800 × (12 / KV100)): baselines ~980 units at ~725 hot starts. Cold wear fixed at 52.5 units (your 150 events, quick ramps mitigating CCS diffs <5%).
Thresholds for HPL Alternatives Net Advantage
- **vs. HPL 10W-40**: Always a slim edge (~1-3% lower wear) due to lower HTHS enabling ~8% faster hot recovery without hydro concessions (identical KV100)—Rudnick's rheology (Ch. 4) favors this in transient spikes, but your 8 sub-0°F starts see negligible CCS uplift. No true threshold; swap if urban errands hit 30%+ miles for ~2% extra shave.- **vs. HPL 20W-50**: 15W-40 optimal up to ~1,500 hot starts/year (~4.1x daily): 20W-50's hydro dominance (thicker films slashing loaf scuff 20%) nets 10-15% savings initially. Crossover at ~1,500-1,800 (~4.9x daily)—transients >78% weighting penalizes its HTHS, flipping to 15W-40 (3-5% better at 5x daily). Your MPG ring-seal holds across, but 20W-50 risks ~0.2-0.4 MPG dip in sub-boost.
When HPL 15W-40 Ceases Optimal: Key Triggers (HPL-Only)
Rudnick's scaling (Ch. 11) flags shifts beyond:
- **Hot restarts >1,800/year (~4.9x daily)**: 10W-40 pulls ahead (~2-4% net) as boundary primes dominate; 20W-50 lags 5%+ from HTHS drag.
- **Short-trip ratio >35%**: 10W-40's lower HTHS trims oxidation creep 8-10% in EGR-heavy urban (TBN holds >11 mid-OCI).
- **Fuel dilution >1.3%**: 20W-50 buffers best (high KV), but >1.7% favors 10W-40's flow for LSPI hedge.
- **Sub-0°F starts >10/year**: 10W-40's -25°C CCS edges for <3% cold shave.
Your loaf bias keeps 15W-40 golden unless hot restarts spike—10W-40 as easy pivot for transients, 20W-50 for hydro insurance. Polaris inbound?
| Hot Starts/Year (Daily ×) | HPL 15W-40 Wear (Units) | HPL 10W-40 Wear (Units) | HPL 20W-50 Wear (Units) | Optimal (vs. 15W-40) |
|---|---|---|---|---|
| 500 (~1.4×) | 882 | 874 | 753 | 20W-50 (15% edge); 10W-40 (1%) |
| 1,000 (~2.7×) | 1,042 | 1,027 | 977 | 20W-50 (6%); 10W-40 (1%) |
| 1,500 (~4.1×) | 1,203 | 1,180 | 1,200 | Threshold (20W-50 parity; 10W-40 2%) |
| 1,825 (5×) | 1,307 | 1,280 | 1,345 | 10W-40 (2%); 15W-40 over 20W-50 (3%) |
| 2,500 (~6.8×) | 1,524 | 1,487 | 1,647 | 10W-40 (2%); 15W-40 over 20W-50 (8%) |
- Joined
- Oct 25, 2021
- Messages
- 2,696
This is corroborated by API. They publish an interesting document called API 1509, which defines when it is acceptable for an oil formulator to use "Read-Across". This is where the test results from an oil can be extrapolated to an oil of similar formulation in a different grade.
For the Sequence IIIH piston deposit test, API requires the tested oil to have an equal or thinner base oil, and equal or higher VM content compared to the untested oils. This is because there's a statistically significant difference in test results that correlate with both the base oil viscosity and the VM content in the piston deposit test. HTHS isn't the best predictor of how well an oil will do in this test. For example, a 5W-20 should outperform a 0W-30 in this test despite having a lower HTHS (all else being equal).
Performance in the Sequence VH sludge test also correlates with base oil viscosity and VM. The Sequence IVB valvetrain wear test is similar, but there is only a correlation with with the base oil viscosity, not the VM content.
The Seq. IIIH requirements in the new API SQ standard also confirm that thicker grades tend to be better for piston deposits. The required merit rating was increased from 4.2 to 4.6 for API SQ, but the 0W-16 grade is exempt and keeps the older 4.2 requirement. 0W-8 and 0W-12 grades require a merit rating of only 3.7, which was the requirement for all grades in the API SN standard. IMO, this might be reason alone not to use anything thinner than 0W-20, even if engine wear isn't a concern.
Last edited:
OVERKILL
$100 Site Donor 2021
That's the point I've been arguing for ages on here, seems like I'm in good company if Les came to the same conclusion.
OVERKILL
$100 Site Donor 2021
Yes and no. It's not just low tension rings, it's shorter pistons with shorter crowns, which push the ring pack up toward the chamber, increasing their exposure to heat, while the piston itself is a smaller heatsink. The ring pack itself is often shorter as well, again, providing less space within to sink the heat, so you end up with a lower profile band that's exposed to more heat while simultaneously using thinner lubricants with bases that are more apt to degrade in the presence of that heat and leave deposits.
As I've stated before, low tension rings have been in use for at least 30 years. They are not the reason we are now seeing elevated oil consumption and stuck rings, they are just an easy target for people to blame. There are numerous engines that feature them that do not have these issues, it's a culmination of design choices that ultimately gets you there, most of them made to reduce friction and improve efficiency.
Last edited:
Agree. Especially in a modern engine with fuel dilution combined; also many being turbo’d and city driven. No bueno if engine longevity is more important than short term ownership.
Someone correct me if I’m wrong, but doesn’t or didn’t 0w-20 also receive a similar pass regarding deposit prevention; at least wrt an API specs?
OVERKILL
$100 Site Donor 2021
Not that I recall? It's the 0W-16 and lower grades there were exempt from specific tests, or had more forgiving limits.
Ah, but I did read somewhere on the forums about 0w-20 vs 5w-20 having differences regarding deposits. Could’ve sworn it was worth a mental note, since I had used 0w-20 for the majority ownership of my little Honda Fit I filed it away. Now to dig into my screenshots.
I think the only exemption that 0W-20 gets is for the TEOST high temperature deposit test, which is a bench test meant to simulate turbo deposit formation after hot shutdowns, at higher temperatures than are seen at the pistons. I don't think it correlates well with piston deposits.
Agree. My daughter’s Acura RDX has never used a drop of oil. There’s definitely some design issues at play.
4wd
$50 site donor 2026
Yep - so pretty much everyone has gone to piston oil jets, oil coolers, and variable oil pumps trying to offset this kind of intensity …
Thanks for sharing. Oxidation is a big deal and should be carefully monitored via a used oil analysis.
If you're using a lower viscosity oil it's important to choose wisely. You want strong oxidation resistance, stout detergent levels and solvency.
VII quantity and type may play a role too. With so many people using dealer bulk 0w20's it should be no surprise that rings will eventually stick at some point. Especially TGMO.
Last edited:
It's not something I'd worry about. GM uses an actual engine test for turbo deposits for their Dexos standards, which is a much better test, and there is no correlation between it and TEOST.
The TEOST test doesn't seem to be very relevant, since it cooks the oil at 480°C. The GM test achieves a max temp of around 270°C, and even that is hotter than the bearings of most liquid-cooled turbos should ever get.
OVERKILL
$100 Site Donor 2021
Uh oh, now we are into the "API minimum standards aren't good enough..." discussion
Similar threads
- Locked
