Originally Posted By: JAG
In the paper the Ca/Mg ratios were 6 and 3, which is larger than all of the d1G2 oils I recall seeing. Mobil 1’s ratio is around 1.5 to 2. The one UOA I looked at of Castrol synthetic had a ratio of around 0.8. Those ratios are significantly lower than the ratios tested, except for all Mg.
In terms of TAN and lead corrosion, performance (not all present in every test) from highest to lowest was: all Calcium, 6:1 ratio Ca:Mg, 3:1 ratio Ca:Mg, all Mg. Lead corrosion is what is most important and even oils having same TAN can cause different amount of corrosiveness. Copper also matters but they did not test for it. TBN has a moderate to high degree of deceptiveness/irrelevance. I’d always choose TAN test over a TBN test. That, in combination with lead and copper values tells a lot. The potential interference is rubbing wear of lead and/or copper bearings. If a person gets a UOA on the same oil at different mileage’s across the way, he/she can look at the curve of lead and copper vs. mileage, and if it shows a significant and increasing rate (PPM/mile) over multiple sucesssive UOAs at OCIs long enough really tax the oil, it would likely indicate acid-induced corrosion has started to occur. To conclude that, it must be determined that excessive dirt ingestion is not the cause of increased rubbing wear.
Mobil 1 Ca/Mg ratio is 4/3, not 3/4. You got it mixed.
https://www.bobistheoilguy.com/forums/ubbthreads.php/topics/4389559/Mobil_1_0W-20_Annual_Protectio
https://www.bobistheoilguy.com/forums/ubbthreads.php/topics/2243763/VOA_with_TBN_and_TAN_for_Mobil
That paper's study is
extremely flawed. Their mostly Ca oil is high-TBN versus all-Mg oil is low-TBN. If you look at the results, the main indicator of lead corrosion is remaining TBN. Ca is working better only because it has a much higher starting TBN. When TBN goes below a certain point, lead corrosion increases dramatically. Once the TBN is depleted, the oil no longer protects regardless of Ca or Mg and the engine starts experiencing great wear rates.
That's why TBN retention is crucial. The only way you can have TBN retention is either a high-SAPS oil (ACEA A3/B4) or a Mg-containing detergent. Since the former isn't allowed for most modern engines, you are stuck with the latter.
Also, the very final sentence in the paper is "The best formulating approach utilizes higher TBN and a mixed calcium/magnesium." (What's up with all the typos in this paper?)
When I was going over the paper again, one thing struck me though:
Mg-containing detergents are greatly increasing the oil oxidation. I don't understand this but if this is true, it would certainly be a significant downside of Mg. This also reminds me Delvac, which is Ca/Mg-based, failing the Volvo T-13 oxidation test according to Rotella, which is Ca-only-based. Could it be because of Mg? It just arises some suspicion after reading this.
So, apart from the caveat regarding oil oxidation, a well-balanced Ca/Mg detergent is clearly superior to an all-Ca detergent in terms of neutralizing acids throughout an extended OCI. The oil-oxidation concerns of Mg need further thought and investigation.