Infineum- Investigating LSPI

[buster]

Link

Quote:
Our current thinking is that the auto-ignition of oil droplets or deposit particles is probably the major cause of LSPI.


Quote:
In addition to investigating the causes of LSPI, Infineum is studying the effects of lubricant composition to better understand how it might contribute to the suppression of LSPI events. Work has already been undertaken to investigate the effects of using different types and levels of base stocks and additives.

The initial focus of the research has been to evaluate the effect of using different detergent chemistries, Zinc dialkyl dithiophosphate (ZDDP) types as well as varying the levels of other additives.

This research has been very challenging. The destructive nature and random occurrence of LSPI has necessitated the development of sophisticated testing and simulation methods in an attempt to fully investigate the phenomenon. Despite the challenge, the initial findings from this research have identified a number of lubricant and fuel chemistries and hardware and operational conditions that can suppress or increase LSPI activity.


Quote:
Infineum’s position here is clear: achieving performance in an engine test that correlates to the field is the preferred method because arbitrary chemical limits do not always result in relevance in the field.

 

[wemay]

Thanks buster. Good read on a situation that may become more prevalent in the near future.

I remember the SAE paper suggesting their test (in its infancy at the time) pointed towards Calcium additives possibly causing this phenomenon.

 

[Solarent]

Thanks Buster - So this means we have 2 out of the 4 major additive companies looking into this. I assume both Lubrizol and Infineum are involved in the development of the new LSPI test that will be part of GF-6.

Quote:
The destructive nature and random occurrence of LSPI has necessitated the development of sophisticated testing and simulation methods in an attempt to fully investigate the phenomenon.


My understanding of this test development was that there is difficulty in predictably reproducing LSPI events and determining their root cause. I'm not convinced that calcium is the culprit, but it does seem that both oil chemistry and fuel chemistry is the main source of the problem. It may take changes in both fuel and oil to get it solved.

 

[wemay]

http://papers.sae.org/2014-01-2785/

Quote:

According to recent technical reports, auto-ignition of an engine oil droplet in a combustion chamber is believed to be one of major contributing factors of LSPI and its formulations have a significant effect on LSPI frequency. Toyota has reported that the oxidation stability of engine oil can be a dominant factor of its auto-ignition, which is a function of its base oil and additive components, of which an example was found that calcium-based detergents significantly increase LSPI frequency.

Based on these findings, we have developed a new engine oil product which can reduce LSPI frequency to less than 10% of that of conventional ILSAC certified gasoline engine oils. High quality base oils (Group III, Group IV) and optimized additive components were formulated in which the amount of calcium-based detergent was reduced to levels lower than general ILSAC oils, and a sufficient amount of anti-oxidants were added. This newly developed engine oil is scheduled to be introduced for Toyota's new turbocharged gasoline engine vehicles. The new engine was released in July 2014 in the Japanese market. Toyota plans to subsequently launch the engine in Europe and the rest of the world.




http://papers.sae.org/2014-32-0092/

Quote:

In this study, abnormal combustion experiments were conducted to investigate the effect on autoignition of a calcium-based additive that is typically mixed into engine oil to act as a detergent. The experiments were performed with a single-cylinder 4-cycle gasoline engine using a primary reference fuel (PRF 50) into which the calcium salicylate (CaSa)-based detergent was mixed at various ratios.

The experimental results showed that autoignition occurred increasingly earlier with a higher concentration of the CaSa-based engine oil additive, giving rise to severe abnormal combustion. This indicates that the addition of a CaSa-based detergent to engine oil tends to promote autoignition and abnormal combustion.

 

[buster]

The mystery continues....Thanks wemay.

That Toyota finding is interesting. I'd say more of the studies seem to show a correlation of Ca and LSPI.

The one you posted seems to be the only one that did not find this correlation.

 

[buster]

Quote:
Influence of Lubricant Additives Calcium Higher calcium levels correlate with higher LSPI frequency


Afton also sees calcium as being an issue.

http://www.slideshare.net/AftonChemical/...ion-engine-oils

See page 27 of the slide.

 

[A_Harman]

I'd like to see a comparison of the Calcium detergents to Magnesium detergents in LSPI tests. Mobil and Castrol have been increasingly using Mg in their passenger car oils. Some of Castrol's oils are majority Mg over Ca.

 

[buster]

^ that is true. Mobil has moved their entire line up, even the MC oils, to the Mg/lower Ca based additive system. Castrol as well.

Mobil 1's SA is now at .8%, the lowest among all full synthetics.

 

[Virtus_Probi]

Originally Posted By: buster
^ that is true. Mobil has moved their entire line up, even the MC oils, to the Mg/lower Ca based additive system. Castrol as well.

Mobil 1's SA is now at .8%, the lowest among all full synthetics.


Not that this oil is easy to find in the US, but M1 5W30 ESP is at 0.6% SA. VOAs indicate that it is low in calcium but does not appear to have much magnesium at all...TBN is pretty low, as one might expect. This product is targeted towards diesel cars and SUVs, but is still API SM/SN and Mobil also recommends it for gasoline engines.
Believe the -30C viscosity is somewhat higher than M1 5W30, which is not an issue at all for most people but is of some concern to me.

 

[Luisraul924]

Originally Posted By: buster
^ that is true. Mobil has moved their entire line up, even the MC oils, to the Mg/lower Ca based additive system. Castrol as well.

Mobil 1's SA is now at .8%, the lowest among all full synthetics.


I believe Mobil's 0w-40 is still rich in calcium.



So on a side note is the "calcium is bad" strictly speaking for turbo direct injection? or is this all gasoline engine vehicles?

 

[wemay]

No matter how cursory the findings may be at this point; To many studies are showing a correlation between Calcium + T-GDI = potential LSPI for me not to believe there's something to it. It's enough for me to stay away from high calcium oils in my Santa Fe Turbo.

 

[wemay]

Originally Posted By: Luisraul924
Originally Posted By: buster
^ that is true. Mobil has moved their entire line up, even the MC oils, to the Mg/lower Ca based additive system. Castrol as well.

Mobil 1's SA is now at .8%, the lowest among all full synthetics.


I believe Mobil's 0w-40 is still rich in calcium.



So on a side note is the "calcium is bad" strictly speaking for turbo direct injection? or is this all gasoline engine vehicles?


Turbo GDI applications, according to the studies cited.

 

[Shannow]

It would be good (IMO) if the manufacturer's adopted the Myagi defence .... best defence is no be there.

Better oil control, rather than trying to engineer oils that won't destroy stuff in their "quart per 800 miles" mindset would be the actual engineering solution.

If it's oil in the end gasses, and the fact that they have no "regional" octane rating due to having no fuel in those areas, then how come so many diesels aren't breaking pistons ?

It's going to be hard to develop a standardised test when the small number of culprit engines are (likely) at the outlying fringes of their manufacturing tolerances.

 

[Greasymechtech]

Turbo + GDI + HighCompression + low quality non-toptier fuel + low octane fuel recommended...

Seems to me that they are making it more complex that it really is. Either lower the CR or stop the recommended 87.

Let's throw in conventional oils, excessive oil consumption, low oil level, high oil temps, inadequate coolant flow, inadequate airflow thru small radiator or under hood in general, soundproofing/insulating engines/manifolds....

Obviously, someone hates cows and can only relate LSPI to oil calcium level. Really?

 

[wemay]

"Someone" = Toyota, Infineum, Lubrizol and the SAE. I'll take their word for it, thank you. Unless further studies show otherwise, as this is a fluid situation. Granted.

 

[wemay]

Excerpt from a previous poster's contribution:

http://www.slideshare.net/AftonChemical/...ion-engine-oils

"26. Influence of Base Oil Takeuchi et al. [SAE 2012-01-1615] evaluated different base stocks in a prototype engine  Found higher group numbers correlated with lower LSPI frequency (Source: Copyright SAE 2014, 2012-01—1615, Takeuchi et al.)
27. Influence of Lubricant Additives Calcium Higher calcium levels correlate with higher LSPI frequency [Takeuchi et al., 2012-01-1615, Hirano et al., 2013-01-2569] (Source: Copyright SAE 2014, (Source: Copyright SAE 2014, 2013-01-2569, Hirano et al.) 2012-01—1615, Takeuchi et al.)
28. Influence of Lubricant Additives Phosphorous and Molybdenum Increasing molybdenum and phosphorus (through ZDDP) levels decreased LSPI frequency [Takeuchi et al., 2012-01-1615] (Source both graphs: Copyright SAE 2014, 2012-01—1615, Takeuchi et al.) Confidential and Proprietary - not to be copied, shared, or reproduced
29. Lubricant Physical Properties Takeuchi et al. [SAE 2012-01-1615] also evaluated lubricant volatility using the Noack test  There was no clear correlation between lubricant volatility and LSPI (Source: Copyright SAE 2014, 2012-01—1615, Takeuchi et al.)
30. Lubricant Physical Properties Takeuchi et al. [SAE 2012-01-1615] also evaluated lubricant oxidative reactivity inferred by auto-ignition temperature  There was a clear correlation between oxidative characteristic and frequency of LSPI DSC = differential scanning calorimeter (Source: Copyright SAE 2014, 2012-01—1615, Takeuchi et al.)
31. Wear Metals and Older Oils Hirano et al. [SAE 2013-01-2569] also looked at lubricant effects  Simulated the impact of wear metals using iron and copper • Found that in their presence LSPI frequency increased  Similar to Takeuchi, also found that used oil may have higher LSPI frequency"

 

[Solarent]

So I went back and reviewed both the papers that Wemay posted:
http://papers.sae.org/2014-01-2785/
This one seems pretty conclusive - they varied the content of Calcium, Phosphorus and Molybdenum and found that reduction of calcium, combined with increased antioxidant strains containing phosphorus or molybdenum reduced the frequency of LSPI events in their test engine. (I wonder if this is the same Ford Ecoboost that is being considered for GF-6).
I would have liked to see the results of different kinds of detergents (for example is it only metalic detergents - ie calcium sulphonates) or do other detergents also pose similar issues. It appears that reduced detergents in general helps - the best results were obtained with low SAPS oils.

This then has trickle down effects:
Originally Posted By: SAE
A lower Ca formulation may lead lower detergency
performance at high temperatures, since Ca comes from
detergent additives such as Calcium sulfonate. Also as VMs
(Viscosity Modifiers), which are an important additive to
improve fuel efficiency by increasing the VI (viscosity index),
affects high temperature detergency, and therefore VMs in the
formulation should be optimized to maintain sufficient
detergency performance in low a Ca formulation.


http://papers.sae.org/2014-32-0092/
This one is interesting because they are really testing what happens when the additive chemistry mixes with the fuel. This seems to indicate that the mixture of fuel and oil together may also be a contributing factor to LSPI. Although this test method is a little less precise than the previous one, the information they found as well as observing the combustion chamber through a window built into the test rig was very interesting. Again the focus here is on calcium containing ingredients - however little by way of comparison to other types of detergents. So I am left to wonder is it purely the calcium or could their be other factors contributing to the results.

I am still not 100% convinced that calcium alone is the culprit, however the correlation towards lowering the concentration is definitely demonstrated. Thanks for those links Wemay! they were good reading.

 

[wemay]

No, thank you for breaking it down Solarent.

 

[Virtus_Probi]

Originally Posted By: Solarent
So I went back and reviewed both the papers that Wemay posted:
http://papers.sae.org/2014-01-2785/
This one seems pretty conclusive - they varied the content of Calcium, Phosphorus and Molybdenum and found that reduction of calcium, combined with increased antioxidant strains containing phosphorus or molybdenum reduced the frequency of LSPI events in their test engine. (I wonder if this is the same Ford Ecoboost that is being considered for GF-6).
I would have liked to see the results of different kinds of detergents (for example is it only metalic detergents - ie calcium sulphonates) or do other detergents also pose similar issues. It appears that reduced detergents in general helps - the best results were obtained with low SAPS oils.

This then has trickle down effects:
Originally Posted By: SAE
A lower Ca formulation may lead lower detergency
performance at high temperatures, since Ca comes from
detergent additives such as Calcium sulfonate. Also as VMs
(Viscosity Modifiers), which are an important additive to
improve fuel efficiency by increasing the VI (viscosity index),
affects high temperature detergency, and therefore VMs in the
formulation should be optimized to maintain sufficient
detergency performance in low a Ca formulation.


http://papers.sae.org/2014-32-0092/
This one is interesting because they are really testing what happens when the additive chemistry mixes with the fuel. This seems to indicate that the mixture of fuel and oil together may also be a contributing factor to LSPI. Although this test method is a little less precise than the previous one, the information they found as well as observing the combustion chamber through a window built into the test rig was very interesting. Again the focus here is on calcium containing ingredients - however little by way of comparison to other types of detergents. So I am left to wonder is it purely the calcium or could their be other factors contributing to the results.

I am still not 100% convinced that calcium alone is the culprit, however the correlation towards lowering the concentration is definitely demonstrated. Thanks for those links Wemay! they were good reading.

Hmmm...they didn't experiment with varying the magnesium levels?
Seems like M1 and Castrol keep a reasonable TBN by adding magnesium when they drop the calcium. M1 ESP just cuts the calcium, adds no magnesium, and lives with a lower TBN with the benefit of a SAPS lower than M1 (I think).

 

[Solarent]

Originally Posted By: Virtus_Probi
Hmmm...they didn't experiment with varying the magnesium levels?
Seems like M1 and Castrol keep a reasonable TBN by adding magnesium when they drop the calcium. M1 ESP just cuts the calcium, adds no magnesium, and lives with a lower TBN with the benefit of a SAPS lower than M1 (I think).


We really can't know exactly what the formulators did. Here on BITOG we compare VOA and PQIA data and assume we know what kinds of detergents/dispersants are being used. In reality - while a lot of things can be inferred by these kinds of reports oil formulas are much more complex than that.

One thing is certain - if it was as easy as substituting a magnesium detergent for a calcium one then everyone would have done it and there would be no need for developing a new LSPI test for the next category upgrade. There is still lots to learn and as TGDI engines take over the market over the next 5 years, I feel like there are still lots to learn that we probably haven't even thought about yet.