More at home oil analysis- Acidity

[sciguyjim]

I thought I'd ask for opinions on a simple pH measurement method I've tried on some used oil samples. I haven't had any oil show up as acidic so I'm not 100% sure the method works.

Acidity (pH) is normally measured in water solutions, but it is also monitored in used oils. Oils have acid neutralizers added which can be used up over time, especially with extended mileage oil change intervals. I've never done any "true" ph measurements on used oil but this was a method I came up with for a simple at home analysis.

I took a small quantity of used oil and put it into a small container. I added water and shook the two liquids vigorously. Acidic compounds would have a tendency to migrate out of the oil and into the water layer. Then I took a piece of common pH measuring paper and dipped it into the water layer. The presence of acidic compounds would be indicated by a color change on the paper. I've never gotten a positive test, either because the test method is flawed or because there were no unneutralized acids in the oil. Is anyone familiar with the proper method of pH measurement in oil? Does this sound like a valid test method?

 

[ivymike]

I expect that the oil will remain slightly basic, or perhaps neutral, until all of the "acid neutralizers" are used up. What you should probably do is measure the amount of buffering agents left in the oil, perhaps by a method similar to the following:

1) do what you did before with the water (not sure if this will capture the buffers or not, but it's worth a try)
2) decant the water into a small-diameter graduated cylinder, and make a note of the volume
3) add an indicator that will tell you if the solution is acidic
4) slowly drip an acid of known concentration into the solution while swirling the solution
5) when the indicator starts to turn color, stop and measure the total volume of the solution

from the above you should be able to estimate how much buffering capacity was in the initial solution. Remember, with a buffered solution, pH will remain fairly constant until all the buffer is exhausted, and then it will rapidly change. SLOW addition of the acid is therefore necessary to achieving accurate results.

 

[ivymike]

try searching online for things like "field test total base number" or "field test TBN" or "testing TBN," and you may find more info.

 

[sciguyjim]

Ivymike,

Your suggestions are good and could work. I kept the analysis as simple as possible because it was done at home where my equipment is limited. I used to work in the petroleum industry long ago, but at my last job there wasn't equipment I could borrow. I believe my method could give a quick yes/no answer about the acidity in just a few minutes which is what I was after. Thanks.

 

[ivymike]

Except that finding out that the oil is not acidic doesn't really tell you how much longer it will stay that way...

 

[sciguyjim]

True, but it tells you if it needs changing right then and there. If I were experimenting with extended drain periods, this would be the fastest way to tell when I reached this limit. Monitoring the TBN of the oil can only indicate when you might be due for an oil change if you have graphed out a trend of TBN vs time. Extrapolation of the graph would pinpoint the oil's maximum allowable lifetime. Soot dispersal is the same. You can monitor soot buildup over time but you get a definite change in the oil spots when it is all used up. Then in the future, just change the oil before the final date determined from the tests. Overall, it's less work than determining remaining TBN or comparing suspended soot particles to those setteled out. The more complex the analysis, the less people will be able to do it.

 

[ivymike]

That's not the only way to determine when you might be due for an oil change. You could also pick a cutoff value and change whenever you reach it. This is the method you're proposing, if I'm not mistaken: changing the oil when TBN is zero, or worse, when the oil becomes acidic.

 

[Chumley]

sciguyjim: Yeah, pH really only has meaning in water. Adding water to oil, then mixing and decanting would dilute any acids present so you wouldn't get an accurate pH reading. It would depend on how much water you added. I wonder if it would be possible to use some type of easily oxidized material like a shiney piece of zinc, tin, or aluminum foil or something where you could note the change in the surface reflectance as a measure of corrosion potential in the oil. Lots of metals etch fairly easily in the presence of weak acids. In fact, I think I'll try that just to see what happens.

I don't know how they calculate TAN or TBN. Anybody know?

Chumley

 

[romke]

I wonder whether this will get you anywhere in terms of being capable in deciding whether the oil has to be changed or not.

In the oilindustry the äcidity can be measured by various methods, all based on titration in some form or another. Neither of these methods were designed to be used in used oils, but were solely designed as a measure of quality control ("have we put the right amaount of additives in&quot. Depending on the method used a Total Acid Number or a Total Base Number is thus calculated.

Unfortunately the values derived by testing used oils do not indicate whether the oil is "worn" or not, because the figure you get is the result of various changes in the composition of the oil and no clear relationship exists between the outcome and the suitability for further use.

Moreover, with modern engine oils for eample you will never get to the point where you measure acidity, because these products are highly buffered. Far before all base material has been used up, other components have deteriorated to a level that renders the lubricant unfit for further use. In engine oils it is usely the anti-wear and anti-oxidant that give up first and the TBN gives no clue to this.

One of the few things you can easily do is compare the oil in terms of viscosity with the fresh oil. Just a slanted board will make comparative measures possible, as long as you see to it that both the fresh and the used oil have the same temperature.

An increased viscosity indicates soot, oxidation, thermal cracking or water, a decreased viscosity indicates fuel dilution or shearing of viscosity improving components.

Water can easily be detected by means of a crackle test, fuel dilution can usually be spotted by the smell.

One more remark: oil should be changed long before it actually gets acidic: some machinery components (copper) are highly sensitive to hot and acidic oil.

 

[Azmio]

Guys,

Interesting, if I can add, what is the sulfur content of the fuel used in your area? BMW, Aprilia, Jaguar, etc. used to have problem with their Nikasil coating on the block bore surface. The sulfur from the combustion by product in the blowby gas as well as the sulfur absorbed by the oil on the bore surface during induction stroke will finally rest in the oil sump. This will form acid when mixed with water which also comes from the combustion product. Together it will eat up the nickel in the nikasil during engine start up water vaporization and causes the Silicone particle to detach from the bore.

However, we also noticed that the Nikasil is problem free in the market like Germany and Japan where the sulfur content is low. Could it be that your region is running low sulfur fuel?

 

[25362]

A note on romke's correct comment on viscosities. When speaking of marine lubricants, changes in viscosity in service of +/- 20-25% are generally tolerated, assuming there are no other adverse symptoms.

Viscosities are reduced by contamination with lighter distillate fuels; contamination with heavy fuel may increase viscosity. On the other hand, the flash point is likely to be reduced in both cases. If increased viscosity is due to carbonaceous insolubles, both viscosity and insolubles can be reduced by centrifuging.

As for automotive engine oils, reserve alkalinity in used (or in-service) oil is there to neutralize acids formed during combustion protecting the engine components from corrosion. TBN is measured, among others, by ASTM and IP methods mostly by electrometric, instead of color, end-points because of the nature of the oil.

 

[jmw]

The limits for marine lubricants seem rather high compared to other lubricants. I recall 10% as a warning and 15% as a critical change for lubricants (from a Noria article some time ago) but the following link suggests lower levels:

http://www.practicingoilanalysis.com/article_detail.asp?articleid=429&relatedbookgroup=OilAnalysis

There are two things you may want to know; the trend so as to predict when the oil needs to be changed and to detect sudden change due to contamnation e.g. coolant contamination due to seal failure. Sudden change requires on line measurement and viscosity is a very effecive indicator.

JMW

http://www.viscoanalyser.com