Marine Engine Crankshaft and Main Bearings

The crankshaft is the single most expensive component inside a marine diesel engine. A fully machined slow-speed crankshaft for an engine such as the MAN ME-C 12-cylinder series weighs over 300 tonnes, costs several million US dollars, and takes around 18 months to procure through forging, heat treatment, and finish machining. It’s also the component whose failure is most catastrophic: a crankshaft fracture in a running engine throws connecting rods through the crankcase at forces that can breach the ship’s structure. The classification rules and IACS unified requirements that govern crankshaft design exist because the failure modes are that severe.

This article covers crankshaft construction (semi-built, fully-built, monobloc, welded), the IACS UR M53 fatigue-strength calculation in enough detail to follow an approval drawing, the main and crankpin bearings that carry the journals, hydrodynamic oil-film theory, bearing materials from babbitt to thin-shell tri-metal, failure modes, the deflection check that detects misalignment before bearings wipe, and the service practices (journal grinding, undersize bearings, oil-mist detection) that keep slow-speed engines running between their overhaul cycles. For the sibling topics that overlap here, see engine alignment and bedplate flexure and engine torsional vibration analysis, which this article cross-links at the relevant points but does not duplicate.

Crankshaft construction types

Three main construction strategies exist for marine diesel crankshafts, and the choice is driven by engine size rather than designer preference.

Monobloc forging

Medium-speed four-stroke engines (bore 200 to 640 mm, cylinder counts 6 to 20) use a monobloc crankshaft: a single continuous forging that runs from the free end to the flywheel flange. The forging is carried out in a large hydraulic press from an ingot of low-alloy steel (typically 34CrMo4 or 42CrMo4 per EN 10083), then rough-machined, heat-treated to quench-and-temper condition for tensile strength in the 700 to 900 MPa range, and finish-machined to the final journal and pin diameters. The crankshaft for a Wartsila 16V32 is about 4.5 metres long and weighs around 12 tonnes; a single-piece forging of this size is feasible.

The journals and pins are induction-hardened to 50 to 60 HRC surface hardness, producing a case depth of 3 to 8 mm. Fillet radii between journal and web are rolled under high compressive load to introduce residual compressive stress, which cuts the fatigue crack initiation risk at the highest-stressed point in the crankshaft. Without fillet rolling, the stress concentration factor at a sharp fillet can reach 2.5 to 3.0; with rolling and an adequate fillet radius, effective stress concentration values fall toward 1.3 to 1.5 (per IACS UR M53 Rev.6 tables).

Semi-built crankshaft

Large slow-speed two-stroke engines (bore 320 to 960 mm, stroke/bore ratios of 3.0 to 4.0, cylinder counts 4 to 14) cannot use a monobloc forging. A 14-cylinder MAN ME engine crankshaft is roughly 17 metres long; no press in the world produces a single forging at that scale with uniform mechanical properties throughout. The solution used by MAN Energy Solutions, WinGD, and their licensees since the mid-20th century is the semi-built crankshaft.

In a semi-built design, the main bearing journals are machined as a continuous journal piece running through several bearing positions. The crank throws (each comprising one crank pin and two webs) are forged separately, machined including the bore at the journal seat, and then heated to expand the bore. The journal section is cooled with dry ice or liquid nitrogen. The throw is assembled onto the journal at the prescribed relative angle, and the temperature difference reverses: the throw contracts onto the journal, producing an interference (shrink) fit. The journal diameter is typically 0.6 to 1.0 per mille of the journal diameter larger than the bore before assembly. Torque capacity of the shrink fit is confirmed by calculation under IACS UR M53 Section 7.

Properly designed, the interference produces a contact pressure high enough that the fit does not slip under the combined bending, torsional, and axial forces at the coupling surface. The IACS M53 calculation verifies this: the minimum required safety factor against slip at the coupling is 1.5 at the maximum combined stress state, accounting for bending from firing loads, torsional load from the power stroke, and any additional axial force from thermal expansion.

Fully-built crankshaft

A fully-built crankshaft takes the semi-built concept further: every journal and every throw is a separate piece, individually heat-treated and hardened, then assembled by shrink-fit one at a time. This approach was used in older large Sulzer two-stroke engines and is retained in some licensee constructions. The advantage is that each journal piece can be replaced independently if damaged, rather than replacing the entire shaft section. The disadvantage is more shrink-fit joints per shaft, each requiring individual verification.

Welded crankshaft

Welded crankshafts join forged throw segments to journal stubs by circumferential butt welds, typically electron-beam or submerged-arc welds, with full non-destructive testing of every weld. This construction allows the use of different materials in the throw and journal sections. Mitsubishi and some European licensees produced welded crankshafts for certain medium-bore engine families. IACS UR M53 addresses welded crankshafts in Section 6, requiring that the weld zone fatigue strength be evaluated separately from the parent-metal fillet strength, and the weld geometry (including any backing rings or weld-root conditions) must be accounted for in the stress concentration factor.

Comparison of crankshaft construction types

Crankshaft geometry: the critical dimensions

The fillet radius is the single most consequential geometric dimension in crankshaft design. The transition from the cylindrical journal or pin surface into the flat web face creates a stress concentration. The stress concentration factor $K_t$ is a function of three non-dimensional ratios: $r/d$ (fillet radius to journal diameter), $D/d$ (web thickness diameter to journal diameter), and $b/d$ (web width to journal diameter). IACS UR M53 Rev.6 provides charts and polynomial regression equations for $K_t$ as functions of these ratios, derived from finite-element studies validated against photoelastic and strain-gauge measurements.

The fatigue-critical locations in the crankshaft are:

1. The fillet between crank pin and web (crankpin fillet), under the combined bending from the pin load and the torsional load from the adjacent throws.
2. The fillet between main bearing journal and web (journal fillet), under bending from the main bearing reaction and residual torsional stress.
3. The oil hole in the crank pin or journal, where a drilled passage creates a stress concentration in addition to the fillet.

The oil hole (drilled from the journal to the pin to feed oil to the bottom-end bearing via the throw) is addressed by a separate stress concentration factor in M53. The hole-edge stress concentration factors for a drilled hole in a shaft under bending and torsion are tabulated, and the resulting effective alternating stress at the hole edge must satisfy the same acceptability criterion as the fillet.

IACS UR M53: fatigue strength calculation

IACS Unified Requirement M53 (current revision: Rev.6, 2017, applicable to all vessels classed or re-classed from that year) prescribes the method by which crankshaft dimensions are approved for marine diesel engines. Every classification society that is an IACS member (including DNV, Lloyd’s Register, ABS, ClassNK, Bureau Veritas, and seven others) enforces M53 as the minimum standard; their own rules may add requirements but cannot be less stringent.

The calculation verifies that the safety factor against fatigue fracture exceeds the prescribed minimum at every stress-critical location in the crankshaft. The procedure is as follows.

Step 1: gas and inertia forces at the crank pin

The tangential gas force on the crank pin is derived from the cylinder pressure diagram (P-V or P-crank-angle indicator diagram). The maximum cylinder pressure $P_{max}$ and the mean effective pressure $P_{mep}$ from the engine specification are the primary inputs. The radial and tangential components of the pin force at each crank angle are:

where $D_{cyl}$ is the cylinder bore, $P(\theta)$ is the cylinder pressure at crank angle $\theta$, $\phi$ is the connecting-rod obliquity angle, $m_{rec}$ is the reciprocating mass (piston + piston rod + crosshead + fraction of connecting rod), $r$ is the crank radius, $\omega$ is the angular velocity, and $l$ is the connecting-rod length. Both gas and inertia contributions are included. M53 specifies the crank angle increments and the minimum number of harmonics to include in the Fourier decomposition.

Step 2: bending moments at the fillet

For each cylinder, the pin force is resolved into bending moments at the adjacent journal fillets. M53 treats the crankshaft as a statically determinate beam between the two main bearings on either side of the throw, with the pin load as the applied force. The bending moment $M_b$ at the journal fillet on the firing side is:

where $F_R$ is the radial pin force, $a$ is the distance from the loaded pin to the near journal, and $c$ is the bearing span. M53 specifies the exact reference cross-sections where this moment is to be evaluated.

Step 3: torsional stress

The torsional moment at each crank station accumulates from the tangential forces at all throws inboard of the station. The alternating component of the torsional moment is taken as half the difference between the maximum and minimum torsional moment over a complete cycle, computed for the most adverse angular position. IACS UR M68 governs the full torsional vibration analysis for the shafting system; M53 borrows the torsional stress amplitude at each crankshaft station from the M68 torsional vibration calculation. The two URs are therefore interdependent: an M53 approval presupposes a completed M68 torsional vibration report. The companion calculator at /calculators/iacs-ur-m68 carries the M68 torsional stress computation.

For engines with known torsional vibration issues in barred speed ranges, M53 requires the fatigue calculation to be conducted at the resonant condition as well, using the maximum permissible torsional stress amplitude from M68 as the input.

Step 4: additional bending from bearing misalignment

This is the step most frequently omitted in desk calculations but explicitly required by M53 Rev.6 Section 5.5. When a main bearing drops below its design elevation (due to wear, chock settlement, or hull deflection), the crankshaft acts as a continuous beam on settling supports. The misalignment produces an additional static bending moment superimposed on the cyclic gas-and-inertia bending. M53 requires that the crankshaft calculation includes the additional bending moment corresponding to the maximum permissible deflection from the engine builder’s manual. This additional moment is static (mean stress contribution) and shifts the mean stress state, reducing the allowable alternating amplitude. This is the mechanistic link between the crankshaft deflection measurement described later in this article and the fatigue margin of the crankshaft itself: a crankshaft operating at the deflection limit is a crankshaft operating at the edge of its M53 safety factor.

For the full alignment and bedplate context, see engine alignment and bedplate flexure.

Step 5: acceptability factor

The combined alternating stress at the fillet or oil-hole is:

where $K_b$ and $K_t$ are the bending and torsion stress concentration factors from the M53 charts, $\sigma_{a,b}$ and $\tau_{a,t}$ are the alternating bending and torsional stresses, $\sigma_{m,b}$ is the mean bending stress (including misalignment contribution), $\sigma_{-1,W}$ is the material’s rotating-bending fatigue limit, and $\sigma_{0,W}$ is the pulsating-bending fatigue limit. The acceptability factor $Q$ is:

where $\sigma_{-1,W}^*$ is the fatigue limit corrected for surface finish, size effect, and actual surface treatment (induction hardening, nitriding, fillet rolling), and $Q_{min}$ is the minimum required factor, which M53 Rev.6 sets at 1.15 for the fillet and 1.05 for the oil-hole location. An engine builder’s design will typically target $Q$ in the range 1.25 to 1.50 to provide margin for the unknown scatter in material fatigue properties.

The companion calculator for IACS UR M71 crankshaft scantling (the related scantling check) is at /calculators/iacs-ur-m71-crankshaft.

Torsional vibration cross-reference

The torsional stress amplitude fed into the M53 calculation comes from the torsional vibration analysis of the complete shafting train. The shafting train modelled in that analysis includes the crankshaft mass-elastic model, the propeller entrained water mass, and any torsional vibration dampers. The analysis identifies critical orders and determines barred speed ranges where vibration amplitudes would exceed M68 limits. See engine torsional vibration analysis for the mass-elastic modelling approach and the relationship between class-approved damper settings and the M53 torsional stress input.

Shrink-fit coupling integrity

For semi-built and fully-built crankshafts, the shrink-fit coupling between throw and journal is a distinct structural element governed by M53 Section 7. The contact pressure $p_c$ at the interface under the interference fit is:

where $\delta$ is the diametrical interference (typically 0.6 to 1.0 per mille of joint diameter), $E$ is Young’s modulus, $d_i$ is the joint diameter, and $D_H$ is the outer diameter of the hub (web). The contact pressure must generate sufficient friction force to resist slip under the full combination of bending moment, torsional moment, and axial force (from firing and inertia) that the joint will experience in service.

The assembly of a semi-built throw requires controlled heating of the throw to typically 150 to 200 degrees Celsius above ambient to open the bore by the required amount, followed by rapid assembly within the thermal window. The relative angular position of the throw is fixed to the prescribed firing order angle by a key or by precision alignment jigs that maintain position until the assembly has equalized in temperature.

Disassembly at major overhaul uses controlled hydraulic oil injection into the shrink-fit interface via radial oil holes (the Pilgrim nut method or oil-injection method per the engine builder’s workshop manual). The oil injection reduces the effective contact pressure to near zero, allowing the throw to be pushed off the journal hydraulically without heat damage.

Main bearings: construction and materials

The main bearings support the main bearing journals of the crankshaft. The crankpin (bottom-end) bearings support the crank pins and are split through the connecting rod big-end cap. Both are thin-shell half-shell bearings in modern practice, though older engines used thick-shell poured white metal in a machined housing.

Thin-shell construction

A thin-shell bearing half is a steel-backed shell stamped or machined to the housing bore, with a total wall thickness of 4 to 12 mm depending on journal diameter. The shell wall-to-diameter ratio is 0.006 to 0.010. The steel back provides the structural stiffness; the bearing material on the inner face provides the running surface. Retention is by interference fit between the shell outer diameter and the housing bore (the crush), supplemented by a locating tang at the parting face to prevent circumferential rotation.

The thin-shell approach replaced poured-metal construction from the 1950s onward. A poured-metal bearing requires in-situ scraping to fit the journal, skilled labour for the pouring, and long overhaul periods. A thin-shell bearing is replaced by a fitter in 4 to 8 hours per journal; no machining or scraping is required.

Bearing material selection

Tin-based white metal (babbitt) has been the standard material in slow-speed marine main bearings since the late 19th century. Its properties are well-matched to the operating conditions: high embeddability (debris particles are absorbed into the soft metal rather than scoring the journal), good thermal conductivity for heat removal, and the ability to conform to minor misalignment and journal imperfections. Its fatigue strength is low (15 to 30 MPa) relative to the peak pressures in the oil film (often 30 to 60 MPa for highly loaded main bearings), which means the white metal depends on the hydrodynamic film to prevent direct loading. If the film breaks down, the white metal yields rather than fracturing, giving the bearing a degree of grace before catastrophic failure.

Tri-metal bearings use the copper-lead or aluminium-tin layer for load capacity and the thin lead-tin overlay (10 to 30 micrometres) for embeddability and running-in. Once the overlay has worn in service (typically within the first 200 to 500 running hours), the intermediate layer is exposed and carries the load at higher fatigue strength. Tri-metal bearings are specified in medium-speed engines where unit loads (force per unit projected area) exceed the fatigue capacity of white metal, typically above 12 to 15 MPa peak film pressure.

Oil supply and clearance

Lubricating oil reaches the main bearings through drillings in the engine block from the main oil gallery. Each main bearing housing has a circumferential groove machined in the upper half that distributes oil around the journal. The oil exits the bearing at the parting faces and at the axial ends, forming the bearing’s flow circuit. A proportion is directed through the crankshaft oil drillings (from the journal through the throw to the crankpin) to lubricate the bottom-end (crankpin) bearings.

The diametrical clearance between journal and bearing (the running clearance) is typically 0.10 to 0.15 per mille of the journal diameter. For a 750 mm journal on a large slow-speed engine, this is 0.75 to 1.125 mm clearance. Too little clearance restricts oil flow and produces high bearing temperatures; too much clearance reduces the oil film load capacity and produces knocking at start and stop when the journal is insufficiently supported.

Clearance is measured at survey by lifting the journal with a hydraulic jack (or in older practice by feeler gauge across the parting faces). The jack method, used on slow-speed engines, lifts the journal vertically and reads the rise on a dial gauge; the lift equals the upper clearance. The feeler gauge method measures the gap at the horizontal parting face. Classification societies’ survey records require the clearance to be within the engine builder’s tolerance band; outside the band requires bearing replacement.

Hydrodynamic oil-film theory

The oil film in a journal bearing is governed by the Reynolds equation. In its two-dimensional form for a finite-length bearing:

where $p$ is film pressure, $h$ is the local film thickness, $\mu$ is the dynamic oil viscosity, $U$ is the journal surface speed, $x$ is the circumferential coordinate, and $z$ is the axial coordinate. This equation has no closed-form solution for the full three-dimensional, dynamically loaded case, which is why engine bearing design has used numerical finite-difference and finite-element solvers since the 1970s.

The film’s ability to support load depends on the Sommerfeld number:

where $N$ is journal speed (rev/s), $P$ is load per unit projected area (N/m$^2$), $R$ is journal radius (m), and $c$ is radial clearance (m). High $S$ (low load, high speed, low viscosity) gives a thick film with the journal running near the bearing centre. Low $S$ (high load, low speed, or thinned oil) drives the journal toward the bearing surface. The minimum film thickness $h_{min}$ falls as $S$ decreases. Marine main bearing design codes (from engine builders) set an acceptance threshold of $h_{min} \geq 3 \times$ the combined RMS roughness of journal and bearing surfaces, typically 2 to 5 micrometres in practice.

The bearing design problem for a marine main bearing is complicated by the dynamic nature of the load. The journal does not sit at a fixed eccentricity: the gas-force cycle in each cylinder drives the crankpin reaction through a full orbit during each revolution. Orbital path analysis, solving the Reynolds equation at each time step through the cycle, reveals whether the journal approaches the bearing surface at any point in the orbit. Class society approval of new engine designs now routinely includes submission of the bearing orbital path calculation from the engine builder’s software, alongside the M53 fatigue calculation.

The companion calculator for the hydrodynamic film is at /calculators/bearing-hydrodynamic-film.

Crankshaft deflection measurement

Crankshaft deflection is the primary in-service alignment check for slow-speed two-stroke engines. The measurement is taken with the engine cold and stopped, with the crankcase fully cooled to ambient temperature (at least 12 hours after shutdown for accurate results on a large engine).

A dial gauge with a curved bridge is clamped between the faces of two adjacent webs at one throw. The bridge is set at the crankpin centre height. The gauge is zeroed at the bottom position (turn the engine to approximately 20 degrees before bottom dead centre, to avoid the connecting rod obscuring the gauge). The engine is then turned by the turning gear through 360 degrees and the reading is recorded at four positions: bottom dead centre (BDC), starboard (or fuel-pump side), top dead centre (TDC), and port (or exhaust side). A fifth position returning toward BDC verifies that the gauge has not shifted.

The deflection value is defined as the difference between the TDC and BDC readings divided by two (half the total web opening). A positive value indicates the shaft is bowing downward (the web opens at TDC, meaning the bearing below that throw has dropped). A negative value (unusual) indicates upward bowing.

MAN ME-series service documentation (ME Engine Service Manual, Chapter 709) gives acceptable deflection limits of approximately ±0.15 mm for typical bore ranges, with tighter limits for longer stroke engines. WinGD specifies similar limits in their TB-X72-01 bulletin. A bearing that has worn sufficiently to produce a deflection at the limit requires replacement before the deflection forces the crankshaft to operate beyond its M53 safety factor, as discussed in the misalignment section above.

Deflection should be measured at commissioning, after every major repair or drydocking, after any grounding incident, and at the intervals specified in the planned maintenance programme under the applicable continuous survey of hull and machinery schedule. The data are trended: a bearing wearing steadily produces a deflection that increases at roughly 0.01 to 0.02 mm per 1,000 running hours in normal service. A sudden step change in deflection without corresponding bearing temperature change suggests a thermal event (hot spot in the chock or bedplate) rather than bearing wear.

The deflection measurement and its interpretation are covered in the context of the full engine mounting system in engine alignment and bedplate flexure.

Journal grinding and undersize bearings

When a main journal or crankpin is scored, pitted, or worn beyond the maximum allowable out-of-roundness (typically 0.02 to 0.04 mm for main journals on slow-speed engines), the journal is ground to a smaller diameter to restore a smooth, round surface. The grinding removes metal to a standard undersize: 0.25 mm, 0.50 mm, 0.75 mm, and 1.00 mm below nominal are the common steps, though engine builders specify the allowable steps in the service manual.

A ground journal requires a matching undersize bearing. The bearing bore diameter is manufactured 0.25, 0.50, or 1.00 mm smaller than the nominal bore to maintain the correct running clearance on the reduced-diameter journal. Engine builders maintain stock of undersize bearings (or can supply them within a procurement cycle of 4 to 8 weeks); the existence of undersize stock is a normal expectation of the classification survey, as bearings worn to the point of requiring journal grinding are a foreseeable maintenance event.

Journal grinding is performed in-situ on slow-speed engines using portable grinding rigs that mount to the bearing housing and use the engine structure as the reference datum. The grinding machine turns an abrasive wheel against the journal while the engine is rotated by the turning gear. This approach avoids removing the crankshaft entirely (an operation that would require the engine to be partially dismantled and the crankshaft lifted by engine-room crane, a major project). In-situ grinding to 1.00 mm undersize typically takes 24 to 48 hours per journal.

For crankpins, the situation is similar, with the additional complexity that access requires the connecting rod to be removed and the piston withdrawn. The pin is ground using a rig mounted on a crosshead guide or similar datum reference.

After grinding, the journal is polished to a surface roughness of Ra 0.4 to 0.8 micrometres (depending on the builder’s specification) to ensure adequate oil film formation. The polished surface is verified by profilometer before new bearings are fitted and the engine is reassembled.

Oil-mist detection and crankcase safety

Oil-mist detection is the primary early-warning system for bearing overheating in the crankcase. When a bearing approaches wipe temperature, the lubricating oil in the vicinity of the overheating surface vaporizes and partially oxidizes, generating a dense oil mist at sharply elevated concentration. Oil-mist detectors (OMDs) sample this crankcase atmosphere continuously and trigger an alarm when the mist concentration exceeds a threshold, typically 25 to 35 per cent of the lower explosive limit.

IACS Unified Requirement M20 (Rev.2, 2004) requires oil-mist detectors on the crankcases of engines above 2,250 kW, or with bore above 300 mm. The detector either serves the whole crankcase through a multipoint sampling arrangement or each bay individually. The alarm output is connected to the engine slow-down system; a full trip (shutdown) is triggered at a higher concentration threshold.

The importance of the OMD is the crankcase explosion sequence. Oil mist alone is not immediately dangerous; it becomes so when a bearing wipes severely, creating a hot spot that ignites the mist. The resulting pressure wave travels through the crankcase and can cause the pressure-relief valves to lift. If air then enters through the open valve and mixes with residual mist, a secondary explosion (the more destructive of the two) follows. The OMD, by alarming on the mist condition before ignition, breaks this chain.

Modern OMD systems supplement mist concentration with bearing temperature trending, oil return temperature per bearing, and crankcase pressure monitoring to give a fuller picture of crankcase condition. Engine builders such as MAN Energy Solutions and WinGD integrate these signals into the engine control system (ECS/PMI) for continuous condition monitoring.

Bearing failure modes and investigation

Wiping

Wiping is plastic flow or melting of the white metal. The contact surface shows a smeared, glazed appearance in the direction of journal rotation, with displaced metal piling at the edges of the contact zone. Severe wipes expose the copper or steel backing. The cause is almost always oil film breakdown: inadequate oil supply (blocked supply drillings, low system pressure), contaminated oil (water contamination reducing viscosity), journal out-of-round beyond tolerance, or excessive bearing load (engine overspeed, short-circuit of adjacent bearing by misalignment).

Investigation follows a systematic sequence: measure the clearance on the wiped bearing and on its neighbours; take an oil sample for spectroscopic analysis (elevated Sn, Sb, Cu indicate tin-white-metal wear); examine the journal for scoring; check the oil supply drilling is clear; review the deflection history and align with bearing reaction expectations.

Fatigue spalling

Fatigue in the bearing material produces a network of fine cracks initiating at the bond line between the white metal and the steel backing, propagating through the white metal layer, and eventually detaching flakes. The appearance is a pock-marked surface with angular flake craters. Fatigue occurs when the peak oil film pressure persistently exceeds the fatigue strength of the white metal (typically 25 to 30 MPa for tin-based alloys). This can result from operating above the engine’s rated power for extended periods, or from bearing unit loads increased by misalignment.

Corrosion (acidic attack)

Corrosive attack on white metal produces a matt, granular surface with a characteristic orange-brown colouring from lead oxide compounds. The attack occurs when lubricating oil acidity rises above a pH equivalent of approximately 6.5, caused by water contamination, sulphur-acidic combustion products ingressing through piston rod glands, or additive depletion. Regular oil analysis with neutralisation number (TBN) monitoring catches this condition before significant bearing damage occurs.

Fretting

Fretting occurs at the interface between the bearing shell back and the housing bore. If the bearing shell does not seat properly (insufficient crush, contaminated seating surface, or incorrect housing bore diameter), micro-relative motion at the interface produces fretting wear that generates iron oxide debris and erodes the housing bore. The consequence is reduced heat conduction through the bearing back and potential loss of dimensional accuracy of the housing, requiring housing boring and line boring on reassembly. The remedy is correct crush (measured by strip-in-strip-out dimensional check) and clean, burr-free housing surfaces.

Cavitation erosion

Cavitation in the oil film produces a characteristic pitted appearance on the unloaded zone of the bearing surface, typically at the parting line region. The pressure fluctuations in the oil film, driven by the cyclic journal load, cause the film to drop below its vapour pressure in the low-pressure zone, generating vapour bubbles that collapse on the bearing surface. Cavitation is more prevalent in high-speed engines and in bearings where the oil supply is restricted. Remedies include increased oil supply pressure, optimised groove geometry, and in some cases a change to a more cavitation-resistant bearing material.

Connection to shafting alignment

The crankshaft’s aft end connects to the propulsion shafting through an output flange, or in direct-coupled two-stroke installations through a coupling to the intermediate shaft. The alignment of this connection affects the mean bending moment on the aft journal and aft main bearing of the crankshaft, and through the M53 additional bending term, the crankshaft fatigue margin.

A correct shafting alignment calculation specifies the target bearing reactions at the aft crankshaft bearing as one of the constraints. The engine builder’s alignment manual for each engine type gives the acceptable range of this reaction. Where the intermediate shaft sag or the stern tube bearing position deviates from the design, the aft crankshaft bearing reaction changes and can exceed the acceptable range, adding mean bending stress to the aft-most crank throw.

This is the intersection with marine propulsion shafting and stern tube alignment calculations and with the engine alignment and bedplate flexure analysis. All three topics share the same root calculation: the influence coefficient matrix relating bearing elevations to bearing reactions across the complete shafting train.

Classification society survey requirements

Classification society surveys of the crankshaft and main bearings follow a defined schedule. Under IACS Unified Requirement S22 and the individual society’s survey programmes, the crankshaft and main bearings are surveyed at each Special Survey (every five years for a vessel on continuous survey), which requires partial or full disassembly of the crankcase for visual inspection of bearing surfaces, journal condition, and fillet areas. Magnetic particle or dye-penetrant examination of fillet areas is required at intervals specified by the society.

DNV Rules for Classification (Ships, Part 4, Chapter 3, 2024 edition) require that crankshaft deflections be measured and recorded at each docking and that the records be available to the surveyor. Deflections outside the engine builder’s limits require investigation and justification before the survey certificate is endorsed.

ABS Rules (Part 4, Chapter 2) require bearing clearances to be measured and recorded. Lloyd’s Register Rules (Part 5, Chapter 2) require that the condition of bearing surfaces is assessed against LR’s published wear criteria.

ClassNK Guidelines for Planned Maintenance Surveys allow extended interval surveys (6.5 years) where the vessel has an approved condition monitoring programme, including continuous deflection trending via shaft monitoring systems and continuous oil analysis.

Monitoring systems in service

Modern slow-speed engines carry a full suite of crankshaft and bearing monitors:

A bearing temperature sensor (thermocouple or Pt100 RTD) is embedded in each main bearing housing within 5 to 10 mm of the bearing back. The signal feeds the engine control system and triggers an alarm at typically 80 to 90 degrees Celsius and a slow-down signal at 95 to 100 degrees Celsius. The exact thresholds are set by the engine builder and are published in the specific engine manual.

Crankshaft deflection autolog systems (fitted as an option on new-build engines by MAN Energy Solutions and WinGD) mount a non-contact displacement sensor between the webs of selected throws and log the deflection continuously during operation, including the variations between hot running and cold stopped conditions. The autolog data are trended over the vessel’s service life and the trend used to schedule bearing replacements before the deflection reaches the limit.

Oil-mist detectors serve the crankcase as described in the oil-mist section. Lubricating oil spectroscopic analysis (LOSA) samples from the crankcase oil at intervals of 250 to 500 running hours and reports on wear metals (tin, antimony, copper, iron, lead) allowing trending of bearing wear rate independently of temperature signals.

Limitations

This article covers the crankshaft and main bearing principles applicable to marine diesel engines from around 500 kW to over 100,000 kW, with an emphasis on large slow-speed two-stroke designs because these dominate merchant shipping. The following limitations apply:

The IACS UR M53 calculation described here follows Revision 6 (2017). Any revision issued after this article’s lastmod date (2026-06-06) may introduce changes to the acceptability factor, stress concentration factor tables, or the misalignment bending treatment; always verify against the current M53 text via the IACS publications portal.

The hydrodynamic theory section treats the Reynolds equation in its classical form. Real bearing analysis uses full three-dimensional, thermo-hydrodynamic models with temperature-dependent viscosity and elastic compliance of the bearing shell and housing (elasto-hydrodynamic or EHD analysis). The qualitative conclusions on film thickness, Sommerfeld number, and minimum film thickness hold; the specific numerical thresholds are engine-specific and must be taken from the engine builder’s bearing design data, not from the general treatment here.

Journal grinding undersize allowances and the specific deflection limits cited here are taken from published MAN ME-series documentation as representative values. For any specific engine type, always use the manufacturer’s service manual for the actual limits; values differ between engine families (MC, ME, ME-C, ME-B, ME-GI) and between cylinder bore sizes.

The crankpin (bottom-end) bearing in the connecting rod big-end and the crosshead bearing in crosshead-type slow-speed engines are related topics not covered in depth here. Their failure modes (primarily wiping in the crankpin bearing and fretting in the crosshead bearing) and their inspection procedures follow the same principles described for the main bearing but with different load patterns and accessibility constraints.

See also

Related wiki articles

• Engine Alignment and Bedplate Flexure
• Engine Torsional Vibration Analysis
• Marine Diesel Engine
• Two-Stroke Marine Diesel Engine Fundamentals
• Four-Stroke Marine Diesel Engine Fundamentals
• Marine Engine Cylinder Liners and Pistons
• Marine Engine Camshaft and Valve Train
• Marine Lubricating Oil Systems
• Marine Propulsion Shafting and Stern Tube
• Engine Bedplate Construction
• Continuous Survey of Hull and Machinery
• IACS: International Association of Classification Societies

Related calculators

• IACS UR M71: Crankshaft Scantling Check
• IACS UR M68: Propulsion Shaft Torsional Vibration
• Bearing Hydrodynamic Film
• Bearing Temperature Alarm Threshold
• Shaft Torsional Critical Speed
• System: Main Engine Slow-Speed 2-Stroke
• System: Auxiliary Engine Medium-Speed 4-Stroke