Engineering Failure Analysis 18 (2011) 18841888

Contents lists available at ScienceDirect

Engineering Failure Analysis

journal homepage: www.elsevier.com/locate/engfailanal

Failure analysis of a gear wheel of a marine azimuth thruster

M. Fonte a,, L. Reis b, M. Freitas b
a
ENIDH Escola Superior Nutica, Pao de Arcos, Portugal
b
IST Instituto Superior Tcnico, Technical University of Lisbon, Portugal

a r t i c l e i n f o a b s t r a c t

Article history: A failure analysis of two helical gear wheels of a ducted azimuth thruster is presented. The
Available online 19 July 2011 research work consisted of a fracture examination of the material in order to determine the
damage root causes. The samples for the failure analysis were obtained from two broken
Keywords: teeth of two helical gear wheels. An analysis through the scanning electron microscope
Crack initiation (SEM) was carried out close to the crack initiation. It was found that the damage in the
Fatigue failure analysis bevel gears were by the fatigue fracture mode. The SEM analysis showed that the gear teeth
Helical gear wheels
were under severe contact stress during the operation aggravated by an inappropriate
Marine azimuth thrusters
lubricating. A possible misalignment between the pinions and the gear wheels teeth could
also contribute for the premature failure.
2011 Elsevier Ltd. All rights reserved.

1. Introduction

Ducted azimuth thrusters [1] are currently used for propulsion system of AHTS (Anchor Handling/Tug/Supply) ships,
namely cable laying and oceanographic researching, drilling vessels, etc. Nozzle propellers are also used in many types of
ships for dynamic positioning of offshore platforms. Computational methods are currently developed in order to take exactly
interaction between propellers and nozzles [2]. Ducted azimuth thrusters are generally used in seismic research ships. How-
ever there are dramatic differences between ducted azimuth thrusters units and conventional open propellers, namely the
shape complexity of propellers, nozzles, gears case housing, stays, etc., which results in a strong hydrodynamic interaction
between propellers and nozzles.
The present failure analysis is based on a vessel which propulsion system is obtained by three ducted azimuth thrusters,
draw-up installation type, having each one a maximum continuing rating (input) of 1839 kW at 900 rpm. The ship where the
damage occurred is a survey ship, being originally built as a cable layer vessel during 10 years. The azimuth thrusters have
been overhauled because the vessel was converted for a new function. The total running hours of the three azimuth thrusters
are about 20,000, being about 10,000 after the converting in the last two years.
Propellers of this vessel are of the controllable pitch type and each one of them is tted in a duct, steering through 360
direction and with a Z-drive gear connection. The output propeller revolutions are about 270 rpm, and each one of azimuth
thrusters is able to develop a bollard pull of approximately 30 tons under 100% of load. Both gearbox and clutch use a forced
lubricating system, the oil ows to each bearing and bevel gears in the gearbox, being adjusted by the orice in the lube oil
manifold provided at the upper side of the gearbox.
Helical gears are extensively used in numerous engineering applications including gearboxes which are key components
of machines. One of the most common causes of gear failure mode is due to tooth bending fatigue. The maximum tensile
stress occurs at the root radius on the active ank of the gear tooth, while the maximum compressive stresses occur at

Corresponding author. Fax: +351 21 4429546.

E-mail address: fonte@enautica.pt (M. Fonte).

1350-6307/$ - see front matter 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.engfailanal.2011.07.009
 M. Fonte et al. / Engineering Failure Analysis 18 (2011) 18841888 1885

Fig. 1. (a) Ducted azimuth thrusters, and (b) one of helical gear wheels.

the root on the passive ank radius [3]. Under cycling loading of gear operation, these regions become preferential sites for
fatigue crack initiation. The causes of gear failure are numerous such as faulty designs, incorrect assembly or misalignment of
gears, overloads, improper applications and eventually poor machining or faulty heat treatments [3,4]. Surface pitting can be
the main failure modes of mechanical elements due to stress concentrations which act as crack initiation sites, and this gen-
erally governs the service life of the components [5]. Damage due to contact fatigue in gear teeth usually occurs along the
pitch line, in the addendumdedendum, and the pits formed on the surface lead to stress concentrations which serve as ini-
tiation sites for the fatigue cracks and eventually reach the failure [6,7].
Failure analysis is a process to determine the cause of a component failure in order to prevent it from happening again. If a
failed component is replaced without determining its root cause, there will be a recurring failure and thus nothing is more
disturbing than a repeat failure.
This work aims to identifying the failure cause of two gear wheels of a vessel in order to prevent or minimize the reoc-
currence of similar damages in the future, providing a better understanding of the fatigue failure for this type of gear wheels
which operate under severe loading conditions.

2. Material and experimental procedures

All relevant damage thruster parts were packed and stored at the shipyard workshop ready for inspection and data col-
lecting. Samples for the micrograph analysis were obtained from two broken teeth of the gear wheels nos. 2 and 3 (600 mm
diameter and 160 mm gear wide) of ducted azimuth thrusters, see Fig. 1a, which damage has resulted in two broken teeth,
one in each gear wheel, see Fig. 1b. The failed gears were inspected visually and macroscopically, and an overall dye-check
was also done in all teeth of gear wheels.
A close-up view of the fractured bevel gear wheel no. 2, showing a fractured tooth and the contact patterns left by the
reduced contact area on the back side of the teeth surfaces, and the bevel gear no. 3 with a broken tooth are shown in
Fig. 2. The samples for the analysis were cut from one of the two damaged teeth, see Fig. 3: one from the gear wheel no.
2 (a), where the contact marks are visible, and another one from the gear wheel no. 3 (b). The two broken teeth present very
similar features and it seems clear that the damage root type of the two helical gear wheels are the same.

3. Results and discussion

The general appearance of the failed helical gears, nos. 2 and 3, is as shown in Fig. 2. Each one shows a broken tooth,
Fig. 2a and b. Arrows in Fig. 2a shows the contact patterns where any crack could be initiated. After visual examination

40 mm 40 mm
(a) (b)

Fig. 2. (a) Fractured bevel gear wheel no. 2, showing the contact pattern on the convex side of the gear teeth, and (b) the bevel gear no. 3 with a broken
tooth.
1886 M. Fonte et al. / Engineering Failure Analysis 18 (2011) 18841888

Fig. 3. Two fractured teeth: (a) from the bevel gear no. 2; (b) from the bevel gear no. 3.

Fig. 4. Micrographs of a sample with 500 X magnication, according to x (a), y (b) and z (c) axes directions.

of the gears, the morphology of the fractured surface of the teeth shows typical features of fatigue phenomena, see Figs. 2 and
3. The micrographs of the teeth material were prepared at metallographic laboratory. Two fractured teeth, Fig. 3, were taken
as samples and were prepared for metallographic examination. The microstructures were obtained in the (a) x, (b) y and (c) z
axes directions, see Fig. 4. The metallurgical analysis of the material showed a typical microstructure for these helical gear
alloys. Heat treatments on the teeth surface layer such as nitriding and carburizing were not found.
A scanning electron microscope (SEM) examination was carried out close to the crack initiation site of teeth. The SEM
observation allowed identifying an extreme contact pressure region, an indication of the lost of lubricating layer effect. It
was found a lot of sliding signs, corresponding to local shear stress effects, as is shown in Figs. 5 and 6.
Fig. 7 shows the pinion gear and the affected areas of helical teeth where are visible the wear marks due to high pressure
contact.
Results do not reveal any material defect of the bevel gear. The two gear wheels show the same type of damage and the
broken teeth have the same feature and size, see Figs. 2 and 3. The attack surface between the active teeth of the gear wheel
and the teeth of the drive side of pinion has the same localized marks left on the surface. Being the same type of damage in
the two gear wheels, the probable root cause seems to be the same. The damage of two gear wheels was clearly by fatigue
due to shear stresses on the tooth surface contact which led to crack initiation. Therefore it seems that the ducted azimuth
thrusters worked above the nominal power and the lubricating oil layer was destroyed.
Observing the micrographs, defects were not found. The microstructure is a typical microstructure for this type of steel
alloy. There was not found any pitting on the active contact surfaces of the bevel gear teeth. Also it was not found any heat
treatment on the teeth surface. The contact marks left on the helical teeth surfaces, as well as the edge of the tooth fragment
shown in Fig. 2, clearly show that there was a high sliding contact between the helical gear wheel teeth and pinion gear
teeth, which is revealed by the observed modication on the tooth root surface where the slant fracture morphology is also
clearly present. This sliding shows an effect of local overstressing. The active contact surface areas among the teeth should be
distributed on a large contact surface and not on a small contact area, as is well presented in Fig. 2a. This type of fatigue
failure shows that crack initiation sites are near the teeth root, and is a consequence of local overstressing due to an incorrect
load distribution on the gear teeth which lead to destroy the lubricating oil layer.

20 mm

Fig. 5. Fractured tooth (bevel gear wheel no. 3) where is shown shear stress effect on the crack initiation zone of the active side of the gear tooth.
 M. Fonte et al. / Engineering Failure Analysis 18 (2011) 18841888 1887

Fig. 6. SEM micrograph observations, close to the crack initiation, where is shown the material sliding on tooth contact surface.

140 mm

(a) (b)

Fig. 7. (a). Pinion gear and (b) the bevel gear teeth where is shown the wear effect due to the high pressure of contact.

In the absence of case-hardening, the resistance to fatigue crack initiation can be reduced and therefore the cracking can
occur prematurely. However the contact pattern left on the convex surface of teeth shows that the loading is localized and
not uniformly distributed, possibly resulting from misalignment, which hypothesis is supported by the examination of the
contact pattern on the matching gear. At the coast side ank of the gear tooth there is a clear indication that during operation
the contact stress has been high enough to initiate fatigue cracks.
The root causes for the helical gear wheels damage seems to be a continuous overloading operation of the thrusters after
the operating change of the vessel to be converted to a seismic research two years ago. For this converting, high power pro-
pulsion was probably required for the vessel operation, associated to a low performance of lubricating oil in the gearbox.
Some misalignment between the pinion shafts and bevel gears arisen, associated with an inadequate thrusters design, for
the required power of this seismic vessel, contributed for the premature failure.

4. Concluding remarks

Two helical gear wheels of a ducted azimuth thruster were examined. The helical gear wheel failure was caused by local-
ized overstress on the teeth surface, probably related with the converting of the vessel, which required additional power. The
fatigue crack initiation started at the root of coast side ank of the gear teeth, followed by the crack growth, and nal frac-
ture. For both gear wheels, the damage root cause identied is clearly the same. The broken teeth have the same size and the
same morphological characteristics. A continuum overloading of the thrusters seems to be the main root cause for this dam-
age, associated to a poor lubricating oil performance in the gearbox. However, some possible misalignment between the pin-
ion and the helical gear wheel, and an inadequate design of the thrusters, contributed to the failure.

References

[1] Funeno I. Hydrodynamic development and propeller design method of azimuthing podded propulsion system. In: 9th symposium on practical design of
ships and other oating structures (PRADS2004), vol. 2. LbeckTravemnde, Germany; 2004. p. 88893.
[2] Funeno I. Analysis of hydrodynamic forces on a propeller blade of tunnel thrusters in bollard conditions. In: Proc. 2nd Pan Asian association of maritime
engineering societies and advanced maritime engineering conference 2006. Jeju Island, Korea; 2006. p. 4538.
[3] Fernandes PL. Tooth bending fatigue failures in gears. Eng Failure Anal 1996;3(3):21925.
1888 M. Fonte et al. / Engineering Failure Analysis 18 (2011) 18841888

[4] Samroeng N, Panya S. Failure analysis of a helical gear. In: The rst TSME international conference on mechanical engineering. Ubon Ratchathani; 2010.
[5] Abhay KJ, Diwakar V. Metallurgical analysis of failed gear. Eng Failure Anal 2002;9(3):35965.
[6] Alban LE. Systematic analysis of gear failures. 2nd ed. USA: ASM; 1985. ISBN:087170-2002.
[7] Johnson KL. The strength of surfaces in rolling contact. J Mech Eng Sci 1989;203(C3):1513.

All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately.