Viscosity refers to very specific physical property, in this case, that of a liquid. Oil with a high viscosity is generally referred to as 'thick' oil, with a lower viscosity as 'thin' oil. When cold-starting an engine, too thick an oil is hard to get moving, with a thin oil the pump cannot build up the necessary pressure because, e.g., the oil in the bearing-cups flows too easily back into the sump. In this case, using customary oil, mixed friction is probable, with all the negative results for the engine mechanics.
Too thick an oil may not even reach the lubrication points, but leaves the oil channel through the pressure relief valve, to relieve the oil pump. It may not even be sucked in at all. Thus, it is possible that oil that is too thick, and oil that is too thin, can cause wear and tear. The oil should maintain a certain viscosity within a certain temperature range.
What we're looking for, is a unit for a force that is to be applied, to shift both extremes of an oil film against each other, taking the time and the contact surface area into consideration. This is the dynamic viscosity, which is expressed by the unit 'Pa * s', whereby, the force is expressed by the pressure (Pa). 1 Pa*s corresponds to 1 Ns/m² (outdated: 10 Poise).
The kinematic viscosity stands for the internal friction between the engine oil layers and is calculated by dividing the dynamic viscosity by the density:
Ns/m²/kg/dm³ = Ns/m²/Nm/s²/dm³ (N is replaced by kgm/s²)Ns/m²/Nm/s²/dm³ = m²/s (with conversion factor)
In 1911, the viscosity classification, SAE J-300, was first agreed on by the Society of Automotive Engineers and later, frequently modified. From the original 6 catagories, there are now more than double that amount. The system is nonetheless, so simple, that still today, it is widely used as an orientation aid. What, in this context is possibly important, is that this standards commitee is, to a certain extent, independent from the USA. We'll see later, that this is not valid for all standards.
The basis is the fluidity of the oil to be tested at 100°C. Originally based on the flow-time, from 1926 onwards it was measured in the, now outdated unit, cST (centi Stokes). In 1952, a measuring point of -17,8°C was added to the 100°C point, because up to then, the cold- starting and running requirements were not sufficiently taken into account. Thus the description numbers were doubled, and the first number was given a 'W' which stood for 'Winter'.
With the development of better additives, the temperature played an ever decreasing part in the fluidity of oils. Indeed, the residues left by these additives in the beginning turned out to be quite a problem. Sedimentation, oxidisation and weaknesses at high temperature were responsible for the retracting of the already issued approvals. In the 1950s there were already low-viscosity oils, they were however, not really reliable.
In the beginning of the 1970s, another problem was solved with the help of the J-300-scale. Due to the further development of the engines, Multi-grade oils were showing irregularities in temperature ranges for which there were no set standards. The results were major engine damages. So now, the additional measuring of the High-Temperature-High-Shear viscosity at 150°C was introduced.