CMI Resources | Condition Monitoring International, LLC

Lube Tips by Jack Poley

The following article was initially published at Reliability Web.

1. Viscosity: Always important, but must be evaluated in context in a Used Oil analysis...

Viscosity of oil is always regarded as its most important property. This phrase, however, makes reference to fresh or new oil. In testing used oils one looks at viscosity with importance, but with the caveat that a normal viscosity could very easily mask one or more problems signified by other tests.

Consider a diesel engine with a fuel dilution problem and a dribbling injector or two. The fuel dilution, of course, lowers viscosity, but the dribbling injector results in over-fueling, causing excessive soot formation, which tends to thicken oil. Because viscosity is driven in two directions by these simultaneously occurring problems, it is possible that viscosity could test quite normal, masking the issues at hand.

2. Silicon (Si): A ‘chameleon’ test result possibility...

Silicon is frequently a difficult element to assess, and should not be categorized as abrasive every time it appears. It has numerous sources and forms, e.g.:

Abrasives: commonly as silicon dioxide (“sand”)
Seals: as a silicone
Oil Additive (defoamant): as a siloxane in polymer form
Coolant chemistries in diesel engines: as a silicate

The first order of the day is to ascertain which of the above, or combinations of the above, exists.

Wear that occurs proportionate to silicon definitely suggests abrasives as one of the significant forms of silicon present, however, this conclusion should not be automatic. Abrasion in reciprocating engines usually occurs upon entry of ‘dirt’ through the air intake. At that time the dirt particles are fairly large and not necessarily detectable via standard spectrometric methods. Until the particles are ground into smaller sizes one could see significant wear increases (Fe, Al, Cr) from the cylinder region, but only token increases in silicon. Caveat, silicon!

3. Zinc (Zn): another ‘chameleon’ element...

Like silicon, zinc has a ‘chameleon-like’ presence because it can have several forms:

Oil Additive: as roughly half of zinc dithiophosphate, a very common additive
Galvanizing (passive parts erosion, e.g., plumbing or plating)
Wear (brass alloying agent)
Contaminant (when required to be <10ppm, e.g., EMD diesel engine)

Phosphorus in proportionate quantities will suggest an additive source, the most common occurrence. When phosphorus is missing one must then be alert to ‘b’ and ‘c’ above. When Zn is a contaminant (d), it is usually in the form of zinc dithiophosphate, representing addition of incorrect lube or a contaminated source.

4. Copper (Cu): yet another ‘chameleon’ element...

Copper is yet another element that can have several forms or sources, dependent on the scenario:

Most of the time Cu is sourced as a moving part wear metal: brass or bronze, from bushings, thrusts or bearing cages
Copper can also be found as a result of fuel system problems in diesel engines: injector seals or tubes, and internally-mounted fuel pump seals.
The cooler core of a diesel engine is another possible source of copper. Most industrial cooling cores, such as found in diesel engines, are made of nearly pure copper.
Though not popular anymore, copper has been used (in the form of an oleate) as a passenger car motor oil additive. Levels around 150ppm are typical, interfering with precision.

5. AES: Atomic Emission Spectrometers are not chemists…

Spectrometers (instruments that inspect for wear, contaminant and additive metals) do not offer any reliable indication of the nature, i.e., the chemical composition or alloy, of the metal detected – only that the metal was detected, along with its concentration. Thus, if silicon is detected, there is no indication as to whether the silicon is from abrasives, seal material, anti-freeze additive or anti-foam additive, among other possibilities.

Because the silicon could be a combination of all sources mentioned above, dependent on the nature of the sample, there is always a level of uncertainty in assessing the its source. It is, therefore, important to utilize clues from other tests performed on the oil to deduce the best estimate of the silicon’s source.

Additive metals (magnesium, calcium, phosphorus, zinc, e.g.) pose the same problem. How does one discern available additive from spent additive? One doesn’t. Again, if the oil seems to be ‘holding up’ ok, the additive package is likely sufficient at the time of sampling.

It stands to reason, then, that the purpose of determining additive metals in used lubes is to obtain some reassurance that the correct product is in the sump. The exception is a new lube, wherein it is reasonable to assume that any additive metals detected are, in fact, functional and fully available.

6. The silicon (Si) abrasive conundrum...

Silicon, when seen to increase from baseline (anti-foam additive) levels, is probably abrasive in nature far more often than not. But it is usually appropriate to look for wear metals increases to support such an assumption. When Si and wear metals go up in near lock-step, the assumption of abrasives seems highly likely, and most experts would draw this conclusion.

Is it possible for wear metals to increase from abrasives trauma, but not see SiO2? If your first inclination is to say, ‘sure, if titanium or other mineral is being mined so that the mined product is the abrasive, rather than Si’, that’s a good answer, but also consider the following scenario:

A reciprocating diesel engine or compressor is operating in a dusty environment, and the air cleaner is compromised to allow a fair amount of the dust into the intake. One surely expects wear to take place in such a case. It may, however, be possible that the abrasive particles may be well above 2-3 microns (the detection limit of many spectrometers, when a refractive like SiO2 is involved). In this reasonable scenario, wear metals may be generated ‘immediately’ (as the abrasives pass by the rings, pistons and cylinders), but the abrasives may not all have been ‘ground up’ in the process to allow them to be sufficiently small enough to have been detected. Maybe they show up later, if they get subsequently ground down, or maybe some of them get filtered in one or two passes.

What the oil analysis test results may indicate, then, is that wear metals are ‘high’, but Si has not proportionally increased. In this situation, a small increase in Si may be indicative, and the Evaluator may have to nuance this phenomenon to achieve the proper assessment.

Carrying this notion to an extreme, it is also possible that the silicon (even as an abrasive) has ‘just appeared’ in the lube, not having had time to cause any detectable wear via metals analysis, but about to do so. If the sample is secured at precisely THAT point in time, one has simply been unlucky. The wear won’t show until the subsequent sample.