What It Means for Real Reliability — and What Happens When It Falls Short
A bearing fails on a paper machine line. The post-mortem finds frosted, glazed raceways — classic surface fatigue. The maintenance team orders the same replacement bearing, applies the same grease, and expects a different outcome. Six months later, the bearing fails again. The root cause was never addressed: the elastohydrodynamic (EHD) film separating the rolling elements from the raceways was chronically too thin to do its job.
EHD film thickness is one of the most consequential — and least visible — variables in rolling element bearing reliability. It cannot be seen, is rarely measured directly in the field, and is almost never discussed during work order reviews. Yet according to ISO 281:2007, film quality directly drives the life modification factor that determines whether a bearing achieves its rated L10 life, exceeds it by a factor of four, or fails catastrophically before it even gets close.
This article explains what EHD film thickness actually is, what governs it in practice, and — critically — what maintenance managers and reliability engineers can do to ensure their bearing contacts operate in the right lubrication regime.
Most maintenance personnel interpret bearing lubrication failure as a lack of lubricant. In reality, a bearing can be fully packed with grease and still fail due to inadequate EHD film thickness — because the base oil viscosity is too low at operating temperature, the speed is too slow to build a film, or the load crushes the film before it forms. Insufficient film is not the same as insufficient lubricant quantity.
Section 1 — The Physics of EHD Film Formation
Why a Film Only 1 Micron Thick Determines Bearing Life
Rolling element bearings operate in a contact regime that most engineers were never taught in school. When a ball or roller presses against a raceway under load, two things happen simultaneously that seem contradictory: the pressures at the contact zone spike to several gigapascals (GPa), and yet the surfaces remain separated by a thin layer of lubricant film — often just 0.1 to 2 micrometers thick, less than the width of a human hair by two orders of magnitude.
This phenomenon is called elastohydrodynamic lubrication, or EHL/EHD. The prefix captures two mechanisms working together: the hydrodynamic action that builds oil pressure as lubricant is squeezed into the narrowing inlet zone ahead of the contact, and the elastic deformation of the metal surfaces themselves, which flattens the contact geometry and extends the high-pressure zone long enough for the film to persist.
The lubricant also undergoes a remarkable transformation under these conditions. As pressure in the contact zone rises — often exceeding 1 GPa — the oil’s viscosity increases dramatically. A lubricant that is 100 centistokes at atmospheric pressure may behave as though it were tens of thousands of centistokes under contact pressure, momentarily transitioning from a flowing liquid to a near-solid state. This pressure-viscosity effect is what allows an extremely thin film to support the full bearing load without rupturing.
The Three Variables That Build the Film
EHD film thickness is governed primarily by three field-controllable variables:
ISO 281:2007 introduced the life modification factor aISO, which adjusts L10 life based on lubrication quality (kappa), contamination level (ec), and bearing material limits. This was a fundamental shift from earlier standards that treated bearing life as a purely fatigue-driven function of load. The standard establishes kappa as the ratio of actual operating viscosity to the minimum reference viscosity required for adequate film formation at the bearing’s speed and geometry.
Key Takeaway — Section 1: EHD film thickness in rolling element bearings is typically sub-micron to a few microns. It forms through a combination of speed-driven hydrodynamic pressure and the pressure-viscosity transformation of the lubricant under contact load. The three primary field variables are speed, viscosity at operating temperature, and load — with viscosity being the most controllable by maintenance and reliability teams.
Section 2 — The Lambda Ratio and What It Tells You
Lambda (Λ): The Number That Tells You Where You Actually Are
Knowing the absolute EHD film thickness in nanometers is not, by itself, enough information to assess lubrication adequacy. A 500 nm film on a mirror-polished bearing surface might provide full separation; the same 500 nm film on a rougher surface might mean significant asperity contact and accelerating wear. The metric that links film thickness to real-world surface conditions is the lambda ratio (Λ, also called specific film thickness).
Lambda is defined as the ratio of EHD film thickness to the composite surface roughness of the contacting pair. Composite roughness combines the individual roughness (Ra or Rq values) of both the rolling element and the raceway surface. In practice, bearing raceways after standard finishing and running-in typically have composite roughnesses in the range of 0.1 to 0.5 micrometers.
The Three Lubrication Regimes and Their Consequences
| Λ Value | Regime | What Is Happening at the Contact | Bearing Life Impact |
|---|---|---|---|
| Λ < 1 | Boundary Lubrication | Full asperity-to-asperity metal contact. Load carried entirely by surface asperities. Lubricant additive chemistry (EP/AW) becomes critical. | Significantly reduced — approaching catastrophic wear potential. ISO 281 kappa < 1 triggers life penalties. |
| 1 ≤ Λ < 3 | Mixed Lubrication | Partial film. Some asperity contact, some fluid load-carrying. Friction elevated. Surface distress and micro-pitting can initiate. | Moderately reduced. Life correction factor below 1.0. Common operating condition in many industrial plants. |
| 3 ≤ Λ < 5 | Full EHD Lubrication | Complete surface separation. No asperity contact. Load carried entirely by EHD film. Maximum fatigue life achievable. | Full rated L10 life or better. Target zone for most industrial applications. |
| Λ ≥ 5 | Thick Film | Over-thick film in some geometries. Risk of roller skidding in lightly loaded zones, viscous drag, heat generation. | No additional life gain. Possible adverse effects from excessive churning in high-speed applications. |
NASA research (Anderson, W.J., 1978) demonstrated that at lambda values below 3, failure modes shift from classic subsurface-initiated rolling contact fatigue to surface- or near-surface-initiated spalling — a qualitative change in failure mechanism, not just a quantitative change in life.
Research has shown the lambda ratio has limitations as a sole predictor of lubrication quality. More sophisticated approaches account for surface topography, inlet conditions, and non-Newtonian lubricant behavior. In practice, lambda remains highly useful as a field-applicable screening tool, but should not be treated as an infallible boundary — some surfaces run-in to establish micro-EHL films even when the nominal lambda suggests boundary conditions.
Key Takeaway — Section 2: A lambda below 1 means asperity contact. A lambda of 3 or above is the target for full film separation and maximum bearing life. Most plants operate a significant fraction of their rotating equipment in the mixed lubrication regime (Λ 1–3) without realizing it — because nobody is calculating it.
Section 3 — Kappa: The Practical Field Metric
Kappa (κ): Translating Film Theory Into Viscosity Selection
For field practitioners, calculating EHD film thickness directly requires data that is rarely available at the work order level: contact geometry, elastic moduli, pressure-viscosity coefficients. The viscosity ratio kappa (κ), codified in ISO 281:2007, provides a more accessible engineering shortcut that correlates directly with film adequacy.
Kappa is defined as the ratio of the lubricant’s actual kinematic viscosity at bearing operating temperature (ν) to the reference viscosity (ν₁) — the minimum viscosity required to form an adequate film for that specific bearing geometry and speed.
According to Pumps & Systems (2025), wrong lubricant viscosity selection accounts for an estimated 20 to 25 percent of all pump bearing failures — not because maintenance teams ignore lubrication, but because they default to a catalog viscosity grade without checking kappa for actual operating conditions.
| ν₁ | Reference viscosity in cSt |
| n | Bearing speed in RPM |
| Dm | Bearing mean diameter in mm |
| k | Constant ≈ 4500 for standard applications |
Key Takeaway — Section 3: Kappa translates EHD film theory into a viscosity selection problem that maintenance engineers can actually solve. Check kappa for every critical bearing application. If kappa is below 2, you have a film problem — not just a lubrication schedule problem. For step-by-step kappa calculation and ISO VG selection, see our guide on how to select the right bearing viscosity (ISO VG).
Section 4 — What Kills the Film in the Field
Five Field Conditions That Collapse the EHD Film Before Failure Begins
Bearing film thickness is never static. Every time operating conditions change — a heat soak after startup, a load surge, a filter change that dilutes the lubricant — lambda and kappa shift. Understanding which field conditions most reliably collapse the EHD film is essential for prioritizing where reliability effort pays off.
The ISO 281:2007 life modification factor aISO is not additive across its inputs. When kappa is below 2 AND contamination is elevated, the combined effect is multiplicative. A contamination level of ISO 18/16/13 combined with a kappa of 1.0 can theoretically reduce bearing life to less than 10% of the rated value. Fixing only viscosity or only filtration without addressing both simultaneously leaves most of the available life gain on the table.
Key Takeaway — Section 4: The five primary film-killers in industrial practice are temperature rise above design, low operating speeds, water contamination, lubricant starvation, and particle contamination. Each can independently collapse the EHD film. When they combine — which they frequently do in heavy industrial environments — bearing life degradation becomes exponential. Reliability programs must address all five, not just grease quantity.
Section 5 — Reading Failed Bearings Through a Film Thickness Lens
What EHD Breakdown Actually Looks Like
Every failed bearing tells a story, but only if the analyst knows what to read. A thin-film or boundary-lubrication failure presents with distinctive surface morphology that differs clearly from classic subsurface-initiated rolling contact fatigue.
Key Takeaway — Section 5: Thin-film and boundary lubrication failures have a characteristic appearance: frosted raceways, fine-grained surface spalling, and heat discoloration at normal loads. These morphologies are distinct from classic subsurface fatigue failures and directly indicate inadequate EHD film thickness. Bearing failure analysis must include a root cause assessment of operating kappa, not just a replacement-in-kind decision.
Section 6 — Improving EHD Film Performance in Practice
Moving the Numbers: How to Improve EHD Film Performance Without Replacing Bearings
EHD film thickness is improvable through lubrication practice changes that do not require bearing replacement, equipment redesign, or significant capital investment. The following actions, applied systematically, can move most bearing applications from inadequate to adequate film conditions.
Recalculate kappa for critical assets at actual operating temperatures. Most lubrication programs set viscosity at ambient or catalog reference temperatures. Recalculating at actual measured bearing operating temperature (from infrared or contact thermometry) frequently reveals kappa values well below the target range.
Upgrade to the correct ISO VG grade. For many slow-speed or high-temperature applications, switching from ISO VG 68 to ISO VG 150 or from VG 100 to VG 220 may be the only change needed to achieve kappa 2–4. This is the highest-leverage, lowest-cost intervention available. See our bearing viscosity selection guide.
Consider synthetic base oils for elevated-temperature applications. PAO and ester-based synthetics have higher viscosity indices than mineral oils — they thin less with temperature. A PAO 100 at 90°C may deliver 25–30% more viscosity than a mineral VG 100, improving kappa significantly without changing NLGI grade or thickener type.
Implement oil analysis or grease sampling to detect viscosity degradation in service. Oxidation, thermal degradation, water dilution, and shear degradation all reduce effective viscosity in service. Oil viscosity at operating temperature is not static; it degrades toward every relubrication interval. For calculated intervals that account for this, see our guide on bearing relubrication intervals.
Address contamination — it multiplies the film thickness deficiency. Improving filtration from ISO 20/18/15 to 16/14/11 in oil-lubricated systems can double effective bearing life at any given kappa value. Contamination and film thickness interact — fixing one without the other leaves substantial reliability gains uncaptured. Full ISO 4406 framework in our article on particle contamination in bearings.
For slow-speed or oscillating applications, evaluate high-viscosity specialty greases with solid lubricant additives (MoS₂, PTFE). These applications may never achieve full EHD film formation. Acknowledging the limitation and compensating with appropriate chemistry is more reliable than expecting the film to form when the physics will not support it.
Key Takeaway — Section 6: Improving EHD film performance is primarily a lubrication engineering problem, not a hardware problem. The highest-leverage interventions are: recalculating kappa at actual operating temperature, selecting the correct ISO VG grade, upgrading to synthetic base oils for high-temperature service, and coupling viscosity improvements with contamination control. These changes can multiply bearing service life by factors of 2 to 5 without any mechanical equipment changes.
Start with your three highest-consequence rotating assets — the ones whose failure causes the most downtime, the most safety risk, or the highest replacement cost. For each, do three things:
- Measure or estimate the actual bearing operating temperature
- Calculate the reference viscosity using the bearing’s pitch diameter and speed
- Calculate kappa by comparing your lubricant’s actual kinematic viscosity at operating temperature to that reference value
If kappa is below 2 on any of those assets, you have identified a reliability improvement opportunity that costs nothing to analyze and very little to fix. A viscosity grade change, a temperature reduction, or a lubrication interval adjustment may be all that stands between current failure frequency and a bearing that outlives the equipment.
The plants achieving top-quartile bearing reliability are not doing so because they buy better bearings. They are doing so because they treat lubrication as an engineering discipline — one governed by calculable parameters like EHD film thickness and kappa, not by habit and historical precedent.
