Why the increase in Calcium, Phosphorus and Zinc?

[skyship]

My used oil analysis after 10K km indicates that most of the detergents and anti-wear additives except Boron have increased from the Virgin oil OA. Eg Zinc from 984 to 1049, Calcium from 2586 to 2751 and the Boron fell from 71 to 62.
I did not add any oil between services, so is this increase because the additives were being concentrated as some (Less than 10%) of the oil was burnt or evapourated, or do additives tend to collect in the sump?? Hard to figure why those figures did not decline.
If it is because the volume of oil has reduced, this might give me a way of tracking oil consumption, as that is very difficult with a diesel that burns so little oil.

 

[Rand]

testing error

i've seen blackstone off 20% or more before...

 

[volk06]

It was probably residual from the previous oil. There is usually +/-1qt left in the engine during an OCI even though you might have drained out the standard 5qts. You can't/don't get every drop out during an OCI.

 

[dparm]

Originally Posted By: volk06
It was probably residual from the previous oil. There is usually +/-1qt left in the engine during an OCI even though you might have drained out the standard 5qts. You can't/don't get every drop out during an OCI.


Bingo. A good lab can typically spot this and put the numbers in context. That's why I use Blackstone for every analysis -- they will comment on how my latest oil choice did compared to the previous one.

 

[skyship]

I cross checked that Blackstones are accurate as I already knew from Liqui Moly what the correct figures were for the virgin oil analysis and their results were very accurate, so this is not an error.
The previous oil was of the same type and not much is left behind anyway, so that is not the reason, BUT I did add a can of LM pre change flush before the previous oil change. The flush might have some very high detergent levels and even a small amount might have had an effect on those results. I can't figure out how the Zinc increased though, so I presume it must be due to the reduction of volume as the figures fit.

 

[ARCOgraphite]

VOA is notoriously innaccurate.

 

[edhackett]

Plus/minus 10% is about all you can expect. Any number within that is "the same".

I usually work to 1-2% and every number comes with an uncertainty value. Each element will cost you about $66.00.

Ed