The Lubrication Regimes and Which One You Are In

Why this matters

"Is it greased?" is the wrong question, and it is the one almost everyone asks. A bearing can be full of the correct lubricant, relubricated on schedule, and still be running with metal touching metal, because the regime it operates in is not a property of the lubricant. It is a running condition made from the lubricant's viscosity at the temperature it is actually at, the bearing's size, and the speed. Change the temperature by 15 C on the same machine with the same grease and you can move across a regime boundary in either direction. This article is a worksheet for finding out which side you are on, filled in for one machine.

Before you take a reading or open a housing

An operating temperature reading on a running machine is a non-contact infrared reading taken from outside the guard. Do not reach past a guard, and do not put a contact probe near a rotating shaft. A bearing housing at 70 C will burn on sustained contact, so if you touch one, the machine is isolated and locked out under 29 CFR 1910.147 first, the wheel has stopped turning rather than merely been switched off, and you either let it cool or use gloves rated for that contact temperature. Do not wear gloves near a shaft that can turn: that is an entanglement hazard, and it is a different task from the one gloves protect. Pulling a grease sample means the same lockout, plus a spring-loaded motor base restrained before belts come off, because the tensioned base is stored energy under the same standard. Cleaning components with solvent is an inhalation route, so work with ventilation, follow the SDS you are entitled to under 29 CFR 1910.1200, and use respiratory protection only inside a written program under 29 CFR 1910.134 if the product calls for it.

The regime is a ratio, not a product

There are two ratios in circulation and they are not the same thing.

Lambda, the specific film thickness, is the physically honest one: the minimum film thickness divided by the combined root-mean-square roughness of the two surfaces. Below about 1 the asperities are touching most of the time. Between about 1 and 3 they touch intermittently. Above about 3 the surfaces are separated. Different sources put the upper line at 3 or at 4, so state which convention you are using rather than mixing them.

Kappa, the viscosity ratio, is the one you can actually compute in the field: the lubricant's kinematic viscosity at its operating temperature, divided by the reference viscosity that this bearing's mean diameter and speed require. It is the input to the life-modification factor in the ISO 281 framework.

Kappa is a practical proxy for lambda, not a substitute for it, and the difference matters: kappa contains no roughness term at all. Two bearings at the same kappa with different raceway finishes are in different regimes. The proxy holds because bearing raceway finishes are broadly similar within a class; it stops holding on a reground shaft journal, a worn raceway, or any surface that has already been damaged.

The worksheet

The worksheet filled in

A belt-driven fan shaft pillow block, insert bearing, 50 mm bore and 90 mm outside diameter, running at 1750 rpm on a lithium-complex grease with a mineral base oil.

• dm = (50 + 90) / 2 = 70 mm
• n = 1750 rpm
• v1 from the catalogue chart at dm 70 mm and 1750 rpm reads about 12 mm2/s. That chart is drawn for a mineral oil, for a normally loaded bearing, and the value it gives is the viscosity required at operating temperature, not at 40 C.
• Grade: base oil ISO VG 100, so 100 mm2/s at 40 C, viscosity index near 95.
• Operating temperature: 70 C, from the housing reading corrected upward, because a housing surface reads below the outer ring which reads below the oil entering the contact.
• v at 70 C off the oil's chart is about 26 mm2/s. Do not interpolate linearly between the 40 C and 100 C datasheet points. Viscosity falls far faster than a straight line, which is why the chart is drawn on log scales.
• Kappa = 26 / 12 = 2.2
• Cleanliness: purge shows fine grey discolouration, shaft seal has visible wear.

Reading the answer

Kappa of 2.2 sits in the mixed regime. Below 1, the film is not carrying and the surface chemistry is the whole protection. From 1 to 4 the asperities touch intermittently and both viscosity and chemistry matter. At 4 and above the surfaces are separated and there is little further gain from going higher.

Three things follow directly from 2.2 that would not follow from 4.0.

The additive package is doing real work. Anti-wear and extreme-pressure chemistry only functions where metal is contacting metal. On a full-film bearing the same additives are inert passengers.

Contamination is expensive here in a way it is not at 4.0. In the ISO 281 life-modification approach the viscosity ratio and the contamination factor interact rather than adding: the same particle count costs far more life at low kappa than at high kappa. This is why a worn shaft seal on a mixed-regime bearing is a real finding and the same seal on a full-film bearing often is not.

Temperature is the lever you actually have. Speed is usually fixed by the process and viscosity grade is a change of product. Temperature is often movable, and it moves kappa hard.

The same bearing, 15 C cooler

Nothing changes except that the machine is in a cooler space and the operating temperature is 55 C.

• v at 55 C off the same chart is about 48 mm2/s
• Kappa = 48 / 12 = 4.0

A 15 C difference moved this bearing from mixed into full film. That is the number worth carrying out of this article: on a mineral base oil with a viscosity index near 95, viscosity roughly doubles for a 15 C drop across the 55 to 70 C part of the curve. The multiplier is smaller at higher temperatures and larger at lower ones, and a high viscosity index or a synthetic base flattens the curve so the same 15 C buys less. Take it from your oil's own chart.

Which is why ventilation on a hot mezzanine, a shaded outdoor unit, or a bearing housing that has been insulated by years of accumulated lint is a lubrication decision and not a housekeeping one.

What grease adds to the arithmetic

Grease is base oil held in a thickener. The thickener does not lubricate; the base oil does. So the kappa calculation uses the base oil viscosity and only the base oil viscosity.

The NLGI grade, the number most people quote when asked what grease is in a machine, is a consistency measurement, and it tells you nothing about film thickness. Two greases both labelled NLGI 2 can have base oils differing by a factor of five in viscosity, which is the difference between kappa 1 and kappa 5 on the same bearing. If the only thing recorded about a machine's lubricant is the NLGI number, the record does not contain enough information to compute the regime.

Grease also has to reach the contact. At very low speeds the running track may not be replenished from the surrounding grease reservoir, so a bearing that computes to an acceptable kappa can still starve. That is a supply problem, not a regime problem, and it shows up on machines that idle at low speed for long periods.

The three ways this worksheet lies to you

You compared the 40 C datasheet number to v1. This is the most common error and it flatters the answer badly: in the example above, using 100 against 12 gives kappa 8.3 rather than 2.2. The reference viscosity is required at operating temperature. Both numbers have to be at the same temperature or the ratio means nothing.

You used the NLGI grade as a viscosity. Covered above; it produces confident nonsense.

You assumed a synthetic is thicker where you need it. A synthetic of the same ISO VG has the same viscosity at 40 C by definition. A higher viscosity index means a flatter curve, so it holds more viscosity at high temperature and correspondingly has less at low temperature than a lower-index oil of the same grade. If the problem is a hot bearing, that helps. If the problem is a cold start on an outdoor machine, the high-index oil is not the thicker one, and calling a synthetic better without naming the temperature is not a statement about anything.

The regime nobody escapes

At zero speed there is no entrainment, so there is no film, whatever the grade. Every machine starts in boundary contact and passes through mixed on its way to whatever regime it runs in. It does the same in reverse on every shutdown.

This is why start-stop count is a real duty parameter and not a footnote. A fan that starts twenty times a day accumulates twenty boundary-regime events a day that a continuously running fan does not, and no lubricant selection removes them. Where the count is high, the levers are the ones that shorten the boundary period, meaning a faster ramp to running speed and an additive package chosen for boundary conditions, rather than a heavier grade, which lengthens the time spent below the entrainment the film needs.

One caution on reaching for extreme-pressure chemistry as insurance. Sulphur-phosphorus packages can attack copper-bearing alloys at elevated temperature, and the reactivity depends on the specific package and the temperature rather than on the category, so check the datasheet's copper-strip corrosion result for the actual product against the materials in your machine rather than treating extreme-pressure grease as either universally safe or universally aggressive.

References

• ISO 281 for the viscosity ratio and its role in the life-modification factor, and for the interaction between viscosity ratio and contamination
• ASTM D341 for the viscosity-temperature relationship the charts are drawn to, and ASTM D2265 or the product datasheet for the base oil properties of a grease
• Bearing manufacturer catalogue for the reference viscosity chart against mean diameter and speed, which is manufacturer-specific
• 29 CFR 1910.147 (hazardous energy and stored energy in a tensioned drive), 29 CFR 1910.1200 (SDS availability), 29 CFR 1910.134 (respiratory protection program)
• See related: How a Lubricant Film Carries a Load; The Lubricant and the Materials It Touches