Analysis of fuel properties in the context of the causes of three marine auxiliary engines failure – A case study

Abstract

This article presents the effects of a severe failure of three four-stroke auxiliary engines in a container ship’s power plant. The failure included jammed cylinder valves, fuel equipment damage, a fire in one of the engines and an explosion of its turbocharger. An analysis of fuel as a factor that constitutes the source and common cause of failure of each engine is made. Laboratory analysis of the fuel is performed. Auxiliary indicators describing the fuel properties are determined, including the calculated carbon aromaticity index (CCAI) and the calculated ignition index (CII). Laboratory analysis of the fuel ignition properties, including the determination of the equivalent cetane number (FIA CN and ECN) and the delay of self-ignition in test conditions (ID). The possible causes of non-compliance with quality standards by the fuel used to power the damaged engines before and during the failure are consulted. Recommendations that may improve safety are presented.

Full text

Engineering Failure Analysis 150 (2023) 107362
Available online 27 May 2023
1350-6307/© 2023 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).
Analysis of fuel properties in the context of the causes of three
marine auxiliary engines failure – A case study
Leszek Chybowski
a
,
*
, Jarosław My´
sk´
ow
b
, Przemysław Kowalak
b
a
Department of Machine Construction and Materials, Faculty of Marine Engineering, Maritime University of Szczecin, 2 Willowa Str, 71-650
Szczecin, Poland
b
Department of Marine Power Plants, Faculty of Marine Engineering, Maritime University of Szczecin, 2 Willowa Str, 71-650 Szczecin, Poland
ARTICLE INFO
Keywords:
Marine diesel engine
Heavy fuel oil
Ignition properties
Fuel contamination
Abrasive wear
Cetane number
Jammed valves
Turbocharger explosion
Cat nes
ABSTRACT
This article presents the effects of a severe failure of three four-stroke auxiliary engines in a
container ship’s power plant. The failure included jammed cylinder valves, fuel equipment
damage, a re in one of the engines and an explosion of its turbocharger. An analysis of fuel as a
factor that constitutes the source and common cause of failure of each engine is made. Laboratory
analysis of the fuel is performed. Auxiliary indicators describing the fuel properties are deter-
mined, including the calculated carbon aromaticity index (CCAI) and the calculated ignition
index (CII). Laboratory analysis of the fuel ignition properties, including the determination of the
equivalent cetane number (FIA CN and ECN) and the delay of self-ignition in test conditions (ID).
The possible causes of non-compliance with quality standards by the fuel used to power the
damaged engines before and during the failure are consulted. Recommendations that may
improve safety are presented.
1. Introduction
Incorrect fuel characteristics directly translate into the efciency and durability of engine components. Contaminants cause an
increase in the intensity of the corrosive and abrasive wear, and incorrect self-ignition properties result in chronic combustion or no
combustion in the cylinders [1–3]. Accumulated excess fuel can damage, seize and disable the air intake and exhaust outlet valves and
cause res in the scavenging air receiver and exhaust manifold [3–5]. The occurrence of these factors may, in turn, result in explosions
in the crankcase [6–8], turbocharger explosions [9–11] and res in the marine power plant [12–14]. A number of examples of the
Abbreviations: a (% m/m), ash content; CCAI (-), calculated carbon aromaticity index; CII (-), calculated ignition index; CIMAC, Conseil Inter-
national des Machines `
a Combustion (the International Council on Combustion Engines); CN (-), cetane number; CP (ms), combustion period; DNV,
classication society Det Norske Veritas; ECN, estimated cetane number; EGR, Exhaust Gas Recirculation; f, g, function designations; FCA, fuel
combustion analyser; FIA, fuel ignition analyser; FIA 100/3, apparatus type; FIA CN, equivalent of cetane number measured with the use of FIA 100/
3; G (MJ/kg), gross specic energy; HFO, heavy fuel oil; ID (ms), ignition delay; IFO, intermediate fuel oil; ISO, International Organization for
Standardisation; N
a
(MJ/kg), net specic energy; NOx, nitrogen oxides; ROHR (bar/ms), rate of heat release; ROHR
max
(bar/ms), maximum rate of
heat release; s (% m/m), sulphur content; SCR, Selective Catalitic Reduction; SMC (ms), start of the main combustion; SOx, sulfur oxides; t (◦C),
temperature of measurement; T22, U15 reference fuels from Phillips Petroleum International; ULO, used lubricating oils; w (% m/m), water content;
ν
(mm
2
/s), measured viscosity;
ρ
15
(kg/m
3
), density at 15◦C;
τ
(ms), time during the measurement with the FIA 100/3 apparatus;
τ
ROHRmax
, moment
on the time axis where the ROHR reaches its maximum value.
* Corresponding author.
E-mail address: l.chybowski@pm.szczecin.pl (L. Chybowski).
Contents lists available at ScienceDirect
Engineering Failure Analysis
journal homepage: www.elsevier.com/locate/engfailanal
https://doi.org/10.1016/j.engfailanal.2023.107362
Received 29 January 2023; Received in revised form 9 May 2023; Accepted 26 May 2023

Engineering Failure Analysis 150 (2023) 107362
2
engine damage and problem cases induced by poor quality fuels are known [15–17]. Nevertheless, each case is different and very often
connected with very complex conditions [18,19]. It is why every case can provide a new look on this topic.
The present authors attempt to analyse the factors that affected the severe failure of three auxiliary engines in the power plant of the
34,015 DWT container ship in 2004. Information about the manufacturers of the ship and engines is anonymised since they are
irrelevant to the presented research problem. As a result of the failure, there was a re in the marine power plant, which was
extinguished by the crew. Fortunately, none of the ship’s crew members was injured because of this incident. Despite the time that has
elapsed since the failure, the subject matter remains relevant due to relatively minor changes in the content of normative acts regarding
fuel parameters, for which deviations from the norm are observed, and the fact that the engines of the analysed type are still in use on
many sea-going vessels [8,20,21]. The above statement motivated the preparation of this article and the presentation of the analysis of
the accident in question in the context of the properties of the fuel used to power the engines that failed shortly one by one within less
than 30 days.
The incident occurred during a sea voyage. The failure concerned three out of four auxiliary (AE) diesel engines of the ship’s power
plant. These were 5- and 6-cylinder, four-stroke, medium-speed, trunk piston, turbocharged, single-acting engines with a rated speed
of 720 rpm and a cylinder power of 220 kW. The engines were not equipped with SOx and NOx reduction systems, including, in
particular, scrubbers, SCR systems, or EGR systems.
The fuel is fed from the return tank (i.e., the deaeration unit) to the auxiliary engines by the booster pumps (one duty pump and one
parallel standby pump). The fuel entering each auxiliary engine is ltered into a duplex lter with a ltration accuracy of 50
μ
m, and
then directed to the injection pumps. Each cylinder is equipped with an individual injection pump. A camshaft mechanically drives the
pumps. Injection pumps feed fuel to the fuel injectors through the high-pressure lines. Each cylinder has one fuel injector located
centrally on the cylinder head. Fuel injection is performed directly into the combustion chamber. The fuel not used by the engine is
returned from the injection pumps via the return line to the return tank. The fuel is properly prepared in advance (cleaned in tanks,
centrifuged, ltered, and heated). The temperature of the fuel fed into the engines in the fuel preparation units is maintained at a
predetermined level by the use of fuel heaters. The viscometer system sets the temperature to such a level that the fuel fed into the
engines has the desired viscosity [22]. For the engines in question, according to the manufacturer’s guidelines, this should be 12–18 cSt
(12–18 mm
2
/s) before the injectors. The fuel supply system for each auxiliary engine is equipped with control, measurement, and
safety equipment, including the monitoring of temperature, pressure, and fuel ow, as well as the pressure drop across the fuel lter.
Each cylinder has two air intake valves and two exhaust gas valves. Air is supplied into the valves through the supply air manifold,
which is supplied with air from the turbocharger (T/C) compressor. Each engine equipped with one T/C. The air is sucked in through
the air lter and directed into the air cooler (one cooler for each engine). From there, the air is directed into the supply air manifold.
The air supply to the cylinders is provided by mechanically actuated intake valves that open the air supply to the cylinders. The
camshaft drives the valves via push rods and rocker arms. The exhaust gases are discharged from each cylinder through the exhaust
valves, which are driven similarly to the intake valves. The exhaust gases are discharged to the exhaust gas manifold, from where they
ow into the funnel through the T/C turbine. The exhaust gases expand in the turbine, which drives the T/C compressor rotor located
Fig. 1. Components of the turbocharger damaged because of its explosion: (a) exhaust gas inlet casing, (b) exhaust gas nozzle ring, (c) inside side
cover, (d) twisted turbocharger shaft, (e) compressor rotor remnants, and (f) turbine wheel remnants.
L. Chybowski et al.

Engineering Failure Analysis 150 (2023) 107362
3
on the same shaft. The air intake and exhaust gas exhaust systems have control and measurement systems. The boosting system is
supported by a jet system that enables air to be supplied to the T/C compressor from the engine’s starting system (a T/C boosting
system). The T/C is protected against overspeed, which causes an emergency stop of the engine when the T/C rotor exceeds the speed
of 42680 rpm.
The engines (ship) was 7 years old at the time of the incident. The time of their actual work was in the range of 38 thousand – 41
thousand operating hours. Unfortunately, we do not have the exact information on how much each engine worked. 16 days have
passed since the last bunkering to the time of the start of the incident. At the time of the event, engines 1 and 2 worked approximately
550 h, while engines 3 and 4 worked approximately 150 h since the last bunkering. The crew made no adjustments in the engines
during the voyage. The number of injector operating hours since the last adjustment/replacement was less than 2000 operating hours.
Engines operated at ca. 50–60% nominal load.
The failure occurred due to a common cause [23–25], which was the negative impact of fuel on the engine components. During the
power plant operation, three engines were damaged and stopped, and the turbocharger on one of the engines exploded – the view of
the essential elements of the turbocharger after the explosion is shown in Fig. 1.
The subsequent verication of the technical condition of the engines showed jammed valves, one of the fuel pumps failed and
incorrect operation of the fuel injectors. The list of individual damages is given in Table 1.
The fuel that was used to power the engines just before the crew dened the failure as a mix of two fuels from two bunker tanks,
which contained fuels taken in two different ports from two different suppliers. In this article, we added anonymized data on suppliers
and tank numbering. The fuel was bunkered in the ports of Port 1 - fuel A (to tanks 1, 2, 7, and 8) and Port 2 - fuel B (3, 4, 5, 6, 7, and 8).
Thus, storage tanks 7 and 8 contained fuel mixed from two different suppliers in ratio of fuel A/fuel B equals to ca. 80/20 % v/v.
Before bunkering, the storage tanks were entirely emptied to the extent allowed by technical solutions, i.e., there are always small
amounts of fuel from previous bunkers, which is the so-called dead fuel. Mixing of fuel types, even of the same type, should not occur
and, if necessary, it should be preceded by appropriate tests to check fuel compatibility, which was not performed in this case (tanks 7
and 8).
The laboratory results of fuel A showed full compliance with the requirements of the then-applicable standard 8217:1996 in
relation to the fuel RMG 35 class (formerly CIMAC IFO 380). On the other hand, laboratory tests of fuel B showed a slight increase in
sodium and water content.
During the voyage that preceded the accident, fuel B was used from tanks 3, 4, 5, and 6, i.e., except for tanks 7 and 8, in which there
was a mix of fuels A and B.
All the events occurred over a short period of less than 30 days. We do not provide exact dates due to the anonymization of the
course of events, but only present individual events on a relative scale that are counted against the initial event (day 0), the moment of
fuel bunkering A. The chronology of events was as follows:
•Day 0–1180 MT of fuel A was bunkered in the rst port.
•Day 3–619 MT of fuel B bunkered.
•Day 16 - engine No. 3 failure - both exhaust valves on each cylinder jammed, and one fuel injection pump became blocked.
•Day 26 - engine No. 1 failure - jammed one inlet and both exhaust valves on one of the engine cylinders, re in the scavenge air
receiver and exhaust gas manifold and, nally, the explosion of the turbocharger. These events occurred almost simultaneously
within a few minutes (less than 30 min).
•Day 29 - engine No. 2 failure - jammed both exhaust valves on the two engine cylinders.
2. Materials and methods
Fuel samples used by the auxiliary engines before and at the time of the failure were tested by a certied testing laboratory of DNV.
The samples taken during the accident investigation were taken from the fuel supply manifold that supplies fuel to the auxiliary
engines (the samples were collected in the fuel conditioning unit of the auxiliary engine fuel supply system). According to the sup-
plier’s documentation, the fuel met the requirements of the RMG 35 class following ISO 8217:1996 (formerly CIMAC IFO 380) [26].
In the following years, the requirements of the ISO 8217 standard changed [27]. The individual values of the characteristics of the
tested fuel are compared with the requirements of the valid at the time of incident ISO 8217:1996 standard. Nevertheless, where
applied we extend our comments taking into account later standards [28]. Information specied for the fuel samples is as follows:
•kinematic viscosity at 40 ◦C, 80 ◦C and 100 ◦C (mm
2
/s),
Table 1
A list of damages to the engines of the ship’s power plant resulting from the analysed failure.
Auxiliary engine AE1 AE2 AE3
Fuel injector nozzles leaking Yes Yes Yes
Fuel injection pump blocked up No No Yes
Cylinder valves blocked up Yes Yes Yes
Piston and liner damages No No No
Turbocharger explosion Yes No No
L. Chybowski et al.

Engineering Failure Analysis 150 (2023) 107362
4
•density at 15 ◦C (kg/m
3
),
•micro-carbon residue (% m/m),
•total sediment potential (% m/m),
•ash content (% m/m),
•water content (% m/m),
•aluminium content (mg/kg),
•total content of aluminium and silicon (mg/kg),
•calcium content (mg/kg),
•iron content (mg/kg),
•lead content (mg/kg),
•magnesium content (mg/kg),
•nickel (mg/kg),
•phosphorus content (mg/kg),
•sodium content (mg/kg),
•silicon content (mg/kg),
•sulphur content (% m/m),
•vanadium content (mg/kg),
•zinc content (mg/kg),
•used lubricating oil content (Ca +Zn and Ca +P) (mg/kg).
The measured physicochemical parameters are compared with the ISO and CIMAC standards to determine the deviations from the
norm. Moreover, based on the measured parameters, the lower and upper caloric values of the fuel are calculated, as well as the
computational indices used to determine the self-ignition properties of the residual fuels, i.e., the calculated carbon aromaticity index
(CCAI) and the calculated ignition index (CII) [29].
To evaluate caloric value of the fuel, we used the formulas for determining the net specic energy N
a
and gross specic energy G
for residual fuels indicated in the CIMAC recommendation (Annex 1 of ref. [30]) following the then applicable ISO/TR 18,455
standard. For residual fuels, net specic energy, N
a
(MJ/kg), and gross specic energy, G (MJ/kg), can be calculated with of accuracy
acceptable for normal purposes from the equations [30]:
Na=46.704 −8.802 •
ρ
15
2•10−6+3.167 •
ρ
15 •10−3
•[1−0.01 • (a+s+w) ] + 0.0942 •s−0.024 •w,(1)
G=52.190 −8.802 •
ρ
15
2•10−6• [1−0.01 • (a+s+w) ] + 0.0942 •s,(2)
where:
ρ
15
(kg/m
3
) is fuel density at 15 ◦C, a (% m/m) is the ash content, s (% m/m) is the sulphur content, and w (% m/m) is the
water content.
CCAI is Shell’s calculation of the autoignition capability of residual fuels (HFO), which is calculated based on the measured vis-
cosity
ν
(mm
2
/s) for a given fuel determined at t (◦C) and the density at 15 ◦C
ρ
15
(kg/m
3
). CCAI can be determined from one of the
equivalent formulas [31]:
CCAI =
ρ
15 −140.7•log[log(
ν
+0.85)]− 80.6−210 •lnt+273
323 ,(3)
CCAI =
ρ
15 −140.7•log[log(
ν
+0.85)]− 80.6−483.5•logt+273
323 .(4)
CII is BP’s calculation of the autoignition capacity of residual fuels (HFO), which is calculated based on the measured viscosity
ν
(mm
2
/s) for a given fuel determined at t (◦C) and the density at 15 ◦C
ρ
15
(kg/m
3
). The values obtained from the calculation of CII for
residual fuels are interpreted analogously as CN for distillate fuels. CII is determined from the formula [31]:
CII = (270.795 +0.1038 •t) − 0.254565 •
ρ
15 +23.708 •log[log(
ν
+0.7) ].(5)
In the last stage of the experiment, bearing in mind that the CII and CCAI indices do not always describe the properties of the so-
called problematic fuels [32], which were signalled in earlier publications [33], the fuel ignitability was analysed using the Fuel
Ignition Analyzer FIA 100/3 apparatus. A detailed description of this apparatus can be nd in [33].
The FIA 100/3 apparatus works efciently in assessing the ignition properties of marine fuels and allows to determine CN
equivalent called FIA CN [34]; it acts as one of the alternatives in this type of research [35,36]. The value of the cetane number, or its
equivalent, should be within the range that ensures reliable and safe operation of the engine, depending on the type of engine and the
anticipated loads [17,37,38] at which the engine will be used. Failure to follow these guidelines may result in severe engine damage
[39,40]. After 2004 fuel ignition analyser FIA analyser were substituted by fuel combustion analyser FCA and instead of FIA CN a new
parameter called estimated cetane number ECN were introduced [41]. The recommended estimated ECN values and the potential
impact of the fuel on engine operation are shown in Fig. 2.
The apparatus enables determination of the ignition delay and maximum heat release rate under test conditions and, on this basis,
L. Chybowski et al.

Engineering Failure Analysis 150 (2023) 107362
5
establishes the equivalent cetane number (CN), marked as FIA CN (FIA cetane number). Interpretation of the indicators describing the
self-ignition properties of the fuel is presented in Table 2. The FIA device and its subsequent versions (i.e., fuel combustion analyser
(FCA)) are used to assess the ignition quality for four-stroke engines at low loads [42].
The apparatus, in controlled conditions in a chamber with a constant volume, performs the combustion process while recording the
course of pressure p (bar) as a function of time
τ
(ms):
p=f(
τ
).(6)
Ignition delay ID (ms) in this apparatus is dened as a time delay from the start of injection until an increase in pressure of 0.2 bar
above the initial chamber pressure has been detected. From the end of the ignition delay to the end of the combustion process, the
combustion period CP (ms) occurs. In turn, the start of the main combustion SMC (ms) phase is determined as the time when an
increase in pressure of 3 bars above the initial chamber pressure has been detected. The SMC indicator is used to establish the ignition
quality of a fuel tested in the form of FIA CN (cetane number). The f function used to determine the FIA CN is the reference curve for the
FIA 100/3 instrument, which shows the ignition properties for mixtures between the reference fuels U15 and T22 from Phillips Pe-
troleum International. The calibration procedure is analogous as in previous version of apparatus [32]. This reference curve establishes
the relationship between the ignition quality (recorded in milliseconds) and the cetane number for the different mixtures of the
reference fuels [44]. For HFOs, FIA CN values typically range from 18.7 to over 40, and the functional dependency can be represented
in general as:
FIACN =g(
τ
=SMC).(7)
In the current version of IP 541/06 standard the FIA CN is substituted by ECN, which in new FCA apparatus is calculated according
to the formula [41]:
ECN =153.15 •e−0.2861•SMC.(8)
Pre-2004 versions of the FIA equipment were individually calibrated for conversion from SMC to ECN. A such apparatus was used
for analysis of the problematic fuel in the presented case study. In this case FIA CN can be converted to ECN with use of the equation
[41]:
ECN =1.2175 •FIACN −3.5713 (9)
To determine the parameters describing the ignitability of a given fuel sample, the apparatus performs a minimum of 12 injections/
ignitions in each test. Based on this data, the mean pressure trace and mean value for ID, SMC, FIA CN and CP are dened. In addition,
the device calculates and determines the course of the rate of heat release (ROHR), which is determined according to the following:
ROHR =f′(
τ
) = dp
d
τ
.(10)
In addition to the listed indicators, the maximum ROHR
max
value (bar/ms) and the moment when this maximum occurs
τ
ROHRmax
Fig. 2. Recommended ECN operational reference ranges (modied from [16]).
L. Chybowski et al.

Engineering Failure Analysis 150 (2023) 107362
6
(ms) are used to describe the ignition and ammability properties of the tested fuel sample, so that:
ROHRmax =ROHR(
τ
=
τ
ROHRmax) = max
τ
→∞ [ROHR(
τ
) ].(11)
Exemplary results of the analysis of ignitability of residual fuels with extreme ignition properties is presented in [43]. All the
mentioned indicators are determined for the tested fuel sample, and they are used to draw conclusions summarising the tests.
3. Results and discussion
All the quantitative results of the analysed fuel are summarised in Table 3. The results of standard tests of physical and chemical
properties, computational auxiliary indicators and the results of fuel ignitability tests performed using the FIA 100/3 apparatus are
indicated here. The results were compared with the standard ISO 8217:1996 for the required quality of residual fuel of this grade (RMG
35). We agree that this standard is applied for bunkered fuel not for the fuel at the inlet of the engine, nevertheless the RMG 35 grade
fuel were approved by the engine manufactures for the analysed engines. Taking it into account we used this standard as a reference in
the performed comparative analysis of used fuel properties against requested properties [45].
The course of the pressure during the combustion process and the rate of heat release (ROHR) of the tested fuel are shown in Figs. 3
and 4, respectively.
The viscosity, density, micro-carbon residue and ash point that are determined as a result of the tests are within the correct range,
indicated by the fuel quality standards displayed in Table 3. The content of the elements, such as phosphorus, sodium, sulphur, va-
nadium and zinc in the fuel samples, is also within the norm. On the other hand, fuel components whose content exceeds the
permissible standards are ash, aluminium, total aluminium and silicon, water and calcium.
Although the total sediment potential is normal, a high value close to the limit and other fuel components indicates signicant fuel
contamination. Ash-contaminating fuel increases the abrasive wear of elements in contact with these substances. Ash is non-ammable
foreign fuel components such as sand, which is a product of corrosion processes of elements and catalyst nes (cat nes). The latter
group of pollutants in HFOs stems from catalytic cracking processes. Cat nes have a diameter of 100
μ
m and smaller ones are as small
as 1
μ
m [46]. These are usually aluminium oxides and silicon oxides, which cause high abrasive wear in the injection system (seizing
pairs of precision injectors and fuel pumps) and in the engine (wear of piston rings and cylinder liners) and formation of deposits [47].
These are very hard particles with a hardness of up to 8.2 on the Mohs scale. For this reason, it is necessary to monitor the aluminium
and silicon content in the fuel [48]. In the analysed case, aluminium and silicon were detected in excessive amounts. In the presented
case, all three engines suffered fuel injector nozzle leaks, which is one of the common symptoms of abrasive wear of the nozzle and
needle sealing surface. It is very likely that elevated ash content, including hard cat nes, signicantly contributed to this. Additional
failure of the fuel pump also indicates a similar nature of the malfunction.
On the other hand, water in the fuel causes chemical and electrochemical corrosion, accumulation of deposits and emulsication,
and also favours the emergence and development of living organisms in the fuel, especially sulphur bacteria, which cause the clogging
of lters, corrosion and worsens the quality of fuel atomisation and fuel pumpability. Water also lowers the specic energy of fuel: 1%
H2O m/m causes energy loss of more than 1% compared to reference conditions [42]. Water can enter the fuel during storage,
transportation and preparation due to leaks in heat exchangers, heating coils in tanks, malfunctioning fuel centrifuges, damaged or
improperly maintained debris drainage systems, etc. High water content could cause fuel emulsication and, consequently, insufcient
atomization. Together with the increased level of hard particles, conditions have been created to transfer pollutants behind the cyl-
inder, into the exhaust ducts and even into the turbocharger. Over time, the hard particles accumulated in the turbocharger could
hinder the free movement of the impeller and, consequently, its seizing and, eventually, its ignition and explosion. Another possibility
observed in the case of engines operating only on residual fuels is incomplete combustion during the starting up and low load
operation. Both conditions may result in the transfer of some amount of incomplete combusted fuel droplets towards the exhaust duct,
where they stick and build up. Higher water content in the fuel may intensify this negative process.
The calcium content is high, and although phosphorus and zinc are normal, calcium should not usually be present in the fuel.
However, in small amounts, calcium, along with vanadium, sodium and nickel, may originate from crude oil and may arise as a result
of fuel contamination during transport. It is a component of lubricating oils. Fuel should be free from used lubricating oil (ULO),
however it was not specied in the standards valid when the event occurred. According to the later standards e.g. ISO 8217:2017, it
should be considered as containing ULO when either one of the following conditions is met:
•calcium above 30 mg/kg and zinc above 15 mg/kg,
•calcium above 30 mg/kg and phosphorus above 15 mg/kg.
Table 2
Ranges of CCAI and CII values for residual fuels of different ignition quality (based on [43]).
Qualitative description CII / FIA CN CCAI
Very bad ignition properties <25 900 – 950
Bad ignition properties 25 – 28 870 – 900
Acceptable to good ignition properties 28 – 35 850 – 870
Good ignition properties 35 – 40 830 – 850
Very good ignition properties >40 790 – 830
L. Chybowski et al.

Engineering Failure Analysis 150 (2023) 107362
7
In this case, these conditions are not met. However, the calcium content indicates contamination of the fuel with external sub-
stances that should not typically be included in the composition of marine fuels. The fuel also contained metals such as iron, lead,
magnesium, nickel and silicon, which presumably wear parts of engines, machines and devices that come into contact with fuel before
it is delivered to the engine injection system. The caloric value, calculated on the basis of the chemical composition, is close to the
average for residual fuels. However, reducing the water content in the fuel, due to a properly implemented centrifugation and sedi-
mentation process, would increase the net and gross caloric value.
The calcium and ash content are slightly affected by separation (ltering) processes. On the other hand, aluminium and silicon can
Table 3
Measured and calculated parameters of the analysed fuel and their comparison with various specications describing the same fuel class.
Limit Parameter CIMAC 1990 / ISO 8217:1996 RM-35G /
IFO-F RMG 35
Analysed fuel RM-35G Normal or alarm level
Max. Viscosity at 50 ◦C (mm
2
/s) 380 285 OK
Max. Viscosity at 100 ◦C (mm
2
/
s)
35 29.4 OK
Max. Density at 15 ◦C (kg/m
3
) 991 986.4 OK
Max. Conradson carbon residue /
micro carbon residue (%
m/m)
18 11 OK
Max. Total sediment potential (%
m/m)
To be developed 0.07 OK
Max. Ash (% m/m) 0.15 0.11 OK but higher required in later standards
(>0.1)
Max. Water (% V/V) 1.0 2.1 HIGH
Max. Aluminium (mg/kg) 30 44 HIGH
Max. Aluminium +silicon (mg/
kg)
N/A 103 OK (not specied in this standard) but
higher required in later standards (>60)
Max. Calcium (mg/kg) N/A 124 OK (not specied in this standard) but
higher required in later standards (>30)
Max. Phosphorus (mg/kg) N/A 3 OK (not specied in this standard but in
limit according to later standards) (<15)
Max. Sodium (mg/kg) N/A 43 OK (not specied in this standard but in
limit according to later standards) (〈100)
Max. Sulphur (% m/m) 5.0 or statutory requirements 1.7 OK
Max. Vanadium (mg/kg) 300 112 OK
Max. Zinc (mg/kg) N/A 5 OK (not specied in this standard but in
limit according to later standards) (<15)
Max. Used lubricating oils
(ULO): Calcium and Zinc;
or Calcium and Phosphorus
(mg/kg)
N/A Calcium >30 and zinc
<15; or Calcium >30
and phosphorus <15.
OK (not specied in this standard but
according to the later standards the fuel
shall be free from ULO, and shall be
considered to contain ULO when either one
of the following conditions is met: Calcium
>30 and zinc >15; or Calcium >30 and
phosphorus >15).
Min. Flashpoint (◦C) 60 70 OK
Net specic energy, N
a
(MJ/kg)
There is no direct reference to these
parameters in the standards, while the
increased water content reduces the’engine’s
energy efciency.
41.21 –
Gross specic energy G
(MJ/kg)
43.62 –
Max. Calculated carbon
aromaticity index CCAI (-)
To be developed 850 OK (not specied in this standard but in line
with later requirement 〈870)
Min. Cetane number CN (or
equivalental parameter)
There is no direct reference to this parameter
in the standards. The interpretation of the
CN/CII value for HFO is given In Table 2.
However, exemplary FIA CN values in
relation to average intermediate fuel oil are
shown In [43].
CII =34 OK
FIA CN <18.7 LOW
ECN <19.2 LOW
Ignition delay ID (ms) There is no direct reference to these
parameters in the standards. The
9.95 HIGH
Start of main combustion
SMC (ms)
interpretation of the values in relation to
average intermediate fuel oil is given In [43].
14.8 HIGH
Maximum ROHR level
ROHR
max
(bar/ms)
1.4 LOW
Maximum ROHR position
τ
ROHRmax
(ms)
15.0 LOW
Combustion period CP (ms) 15.1 OK
L. Chybowski et al.

Engineering Failure Analysis 150 (2023) 107362
8
be strongly reduced by separation [42]. Thus, a high content of aluminium and a high total content of silicon and aluminium imply
improperly conducted service processes in the ship’s fuel purication system and/or disturbances in the operation of fuel centrifuges.
High water content in the analysed fuel also indicates improper fuel cleaning.
The calculated CCAI and CII indexes indicate proper ignition properties. However, it should be remembered that these values are
only supplementary to the detailed laboratory analysis of the fuel. The fuel centrifugation purication process itself does not affect
signicantly parameters such as density and viscosity and, therefore, does not signicantly alter the calculated values of indicators
such as CCAI and CII [47]. In the analysed case, both of these indicators corresponded to the correct fuel ignition properties. To avoid
similar situations, engine manufacturers e.g. MAN B&W [47] recommend additional tests (e.g., FIA) when CCAI exceeds 840. On the
other hand, when FIA CN is below 20, engine operation problems can be expected, especially at low loads [49–51].
Although the CCAI and CII values are normal, fuel ignitability tests performed with the FIA 100/3 apparatus clearly show that the
tested fuel has very poor self-ignition properties. Thus, unburned fuel accumulates in the engine, which causes engine damage. When
the accumulated fuel suddenly ignites, it causes res in the exhaust manifold and air tank, which may lead to the explosion of the
turbocharger. All these factors were at play in the present case of the severe accident in the ship’s power plant.
The tested fuel had a much longer ignition delay ID and the period from fuel injection to the start of main combustion SMC was
signicantly longer than the average for residual fuels. However, the maximum value of the ROHR level, and its position on the time
axis, are lower than the average. The summary of the laboratory fuel ignitability analysis is the FIA CN index, which in the present case
is below the value of 18.7, that corresponds to very poor self-ignition parameters of the fuel and can cause serious problems when
operating four-stroke engines at low loads, e.g., when a ship is manoeuvring into or out of a harbour.
In the presented case, it was found that the fuel sampled from the engine’s fuel supply system was of poor quality. It is uncertain
whether the bunkered fuel was of poor quality or was contaminated and deteriorated on board. There is insufcient evidence to judge
it with the available data and methods. However, the contamination likely occurred on board because the fuel samples were taken
during the bunkering process for laboratory analysis. One of the possible explanations is the insufcient draining of the sludge and
water from the settling and service tank and, eventually, improper operation of the fuel purication system. Interestingly, the addi-
tional analysis of the signs of ULO detected such contamination. Again, there is no evidence regarding the origin of this contamination
– was it bunkered with it or contaminated on board? However, regardless of the origin of the contamination, damage to the engines
shows the validity of the later introduction of restrictions for markers of admixtures of used oils in marine fuels in the standard.
Fig. 3. Averaged pressure changes in the chamber of the FIA 100/3 apparatus during the ignitability tests of the tested fuel (modied from [14]).
Fig. 4. Averaged change in the rate of heat release in the chamber of the FIA 100/3 apparatus during the ignitability tests of the tested fuel
(modied from [14]).
L. Chybowski et al.

Engineering Failure Analysis 150 (2023) 107362
9
After the failure of each of the engines, the auxiliary engines were overhauled [52,53]. The remnants of fuel mix A and B contained
in storage tanks 7 and 8 were handed over to the port, proper drainage of all tanks was ensured, fuel lters were cleaned, and the
operation of fuel centrifuges was checked and adjusted. Strict procedures have also been introduced to increase the frequency of the
draining fuel tanks and to test fuel in service tanks and fuel supply manifolds of the auxiliary and main engines more frequently on
board for water content.
Due to the lack of injured people, external bodies performed no investigation, such as marine accident investigation committees.
However, given the stressful situation during the incident, it can be assumed that such information could contain subjective opinions of
the crew members. It is also difcult for us to refer to the specic actions of the crew members during the event. There is no doubt,
however, that several acts of negligence (as indicated in the article) have contributed to the accident. Negligence could result from the
crew’s lack of proper training, sluggishness in operation and ignoring operational procedures, crew fatigue (e.g., in connection with
the performance of other duties), lack of adequate control of subordinates by the management, and falling into the so-called routine.
However, considering the chronology of events, if the crew had drawn the correct conclusions after the failure of engine 3 and taken
appropriate corrective actions as soon as possible, the failures of engines 1 and 2 could have been prevented. However, this did not
occur. One of the reasons in relation to this may be the low level of training of the crew.
4. Conclusions
The performed fuel tests showed that the fuel might the leading and common cause of failure of all three auxiliary engines in the
present case. Based on the detailed observations, arising from the discussion of the results of the tests, several general conclusions
regarding the failure in question can be drawn:
1. It cannot be ruled out that the combination of fuel parameters that contributed to the failure of the engines could have resulted from
contamination of the fuel taken from the bunker with fuel residues left in the bunker tanks, and the synergistic effect of the
composition and properties of these fuels on the physical and chemical properties of the fuel formed after their mixing.
2. High content of water, aluminium and silicon in fuel is associated with improperly conducted cleaning processes in lters, cen-
trifuges and tanks. However, it cannot be ruled out that the composition of the fuel was a factor that deteriorated the quality of
cleaning with the use of standard equipment used in installations for transporting, cleaning and supplying marine engines with fuel.
3. The present case shows that the calculated indexes of residual fuels’ self-ignition capacity, such as CCAI or CII, can be used as
information about the poor fuel quality in the case of extremely unfavourable combinations of fuel rheological properties. On the
other hand, the indication of proper quality by these indicators does not necessarily reect the actual correct ignition delay of fuel.
The case in question shows a very high ignition delay described by a low FIA CN value, despite the normal CCAI and CII values.
4. Particularly unfavourable factors related to the fuel properties could rapidly intensify their synergistic effect, as a result of the
gradual deterioration of the engine’s technical condition and operating conditions. Namely, an increase in specic fuel con-
sumption as a result of a decrease in engine power and the related increase in the thermal and mechanical load of the engine.
5. Poor ignition properties could additionally contribute to high-pressure peaks and increased heat load of the combustion chamber.
All these factors contribute to a noisy, hard and knocking running engine, which was not observed by the crew. Of course, the
reason could be due to the rapid development of the sequence of events leading to the failure.
6. Additional accumulation of individual negative factors contributing to the failure takes place when the engine is operated at low
load, which may intensify the accumulation of impurities in the combustion chamber, burnout the piston crown, damage the piston
rings, create a loss of lubrication of the cylinder liners and cause possible fatigue damage, especially under metal-to-metal
conditions.
7. Incomplete combustion at jammed exchange valves causes unburnt fuel to ow out and accumulate outside the combustion
chambers, in the exhaust manifold and scavenging air accumulator. This, in turn, can cause res and explosions in the engine’s
exchange system.
Based on the conclusions presented above, some recommendations can be made to avoid similar situations in future. For this
purpose, we propose the following improvements:
1. Conducting proper supervision over fuel, which should be checked each time in laboratory conditions before their initial use as fuel
for powering engines. If possible, it is advisable to perform complete tests (as much as possible) of the fuel taken into the storage
tanks during bunkering, including an assessment of self-ignition properties (especially for fuels with a CCIA value above 840).
2. Do not mix different types of fuels (or worse, add used oil or crude oil waste to fuel for engines).
3. Fuel bunkering should be performed only in the case of tanks with the minimum possible content of the so-called dead fuel (fuel
remaining in empty tanks). If this is not possible, additional tests of fuel compatibility and self-ignition properties of the mixture of
two fuels should be carried out.
4. It is also necessary to strictly monitor the quality of the fuel on board and to take care of the fuel, including monitoring the fuel’s
water content in settling and outgoing service tanks and constantly supervising the proper operation of fuel purication devices,
such as centrifuges and lters.
5. The technical condition of the engine components (i.e., combustion chambers, fuel apparatus, and valve timing system) should also
be monitored on an ongoing basis, strictly adhering to the schedule of maintenance works and the guidelines of the engine
manufacturer contained in the manuals and service lists.
L. Chybowski et al.

Engineering Failure Analysis 150 (2023) 107362
10
Data availability statement
All data are presented in the paper.
Funding
Costs associated with the preparation and publication of the article were funded by the Ministry of Science and Higher Education
(MEiN) of Poland, grant number 1/S/KPBMiM/23.
CRediT authorship contribution statement
Leszek Chybowski: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data
curation, Writing – original draft, Writing – review & editing, Supervision, Project administration, Funding acquisition. Jarosław
My´
sk´
ow: Validation, Formal analysis, Investigation, Writing – original draft, Writing – review & editing. Przemysław Kowalak:
Validation, Formal analysis, Investigation, Writing – original draft, Writing – review & editing.
Declaration of Competing Interest
The authors declare that they have no known competing nancial interests or personal relationships that could have appeared to
inuence the work reported in this paper.
References
[1] A. Bejger, L. Chybowski, K. Gawdzi´
nska, Utilising elastic waves of acoustic emission to assess the condition of spray nozzles in a marine diesel engine, J. Mar.
Eng. Technol. (2018), https://doi.org/10.1080/20464177.2018.1492361.
[2] J. Herdzik, Problems of propulsion systems and main engines choice for offshore support vessels, Zesz. Nauk. Akad. Morskiej w Szczecinie, Sci. Journals Marit.
Univ. Szczecin 36 (2013) 45–50.
[3] M. Szczepanek, Biofuels as an alternative fuel for West Pomeranian shing eet, J. Phys. Conf. Ser. 1172 (2019) 012074, https://doi.org/10.1088/1742-6596/
1172/1/012074.
[4] A. Krystosik-Gromadzi´
nska, Engine room re safety, Sci. Journals Marit. Univ. Szczecin, Zesz. Nauk. Akad. Morskiej w Szczecinie 47 (2016) 29–35, https://doi.
org/10.17402/145.
[5] L. Chybowski, S. Strojecki, W. Markiewicz, Simulation-Based Training in Fire Prevention and Fire-Fighting of Scavenge Air Receivers Fires, Syst. Saf. Hum. -
Tech. Facil. - Environ. 2 (2019) 100–111.
[6] D. Wiaterek, L. Chybowski, Comparison of selected models useful in ranking the root causes of explosions in marine engine crankcases, Sci. Journals Marit. Univ.
Szczecin, Zesz. Nauk. Akad. Morskiej w Szczecinie 72 (2022) 77–85, https://doi.org/10.17402/536.
[7] L. Chybowski, D. Wiaterek, A. Jakubowski, The Impact of Marine Engine Component Failures upon an Explosion in the Starting Air Manifold, J. Mar. Sci. Eng.
2022 (1850) 10, https://doi.org/10.3390/jmse10121850.
[8] D. Wiaterek, L. Chybowski, Assessing the topicality of the problem related to the explosion of crankcases in marine main propulsion engines (1972–2018), Sci.
Journals Marit. Univ. Szczecin, Zesz. Nauk. Akad. Morskiej w Szczecinie 71 (2022) 33–40, https://doi.org/10.17402/515.
[9] Machinery Spaces Reason of Turbocharger damages and recommendations to avoid turbo failure Available online: http://www.machineryspaces.com/turbo-
failure-prevention.html (accessed on Nov 20, 2022).
[10] M. Jensen, S. Jakobsen, Increased Protection of Turbochargers against Overspeed. Service Letter SL2015-599/PEKI, Augsburg, MAN Diesel and Turbo, 2015.
[11] Melett Overspeeding; Wabtec Corporation: Barnsley, 2018.
[12] C.N. Hughes, Lessons to learn from turbocharger explosions, MER - Mar. Eng. Rev. (2000) 32.
[13] Insight Hull and machinery incident - Turbocharger damage Available online: https://www.gard.no/web/updates/content/51669/hull-and-machinery-
incident-turbocharger-damage (accessed on Dec 4, 2022).
[14] L. Chybowski, Z. Matuszak, Marine Auxuliary Diesel Engine Turbocharger Damage (Explosion) Cause Analysis, J. Polish CIMAC 2 (2007) 33–40.
[15] S. Grzywacz, A. Stępniak, J. Listewnik, Analiza awarii silnik´
ow gł´
ownych i pomocniczych; ITESO, Szczecin, Wy˙
zsza Szkoła Morska w Szczecinie, 1988.
[16] CIMAC Fuel Quality Guide - Ignition and Combustion; CIMAC: Frankfurt, 2011.
[17] P. Kowalak, J. My´
sk´
ow, T. Tu´
nski, D. Bykowski, T. Borkowski, A method for assessing of ship fuel system failures resulting from fuel changeover imposed by
environmental requirements, Eksploat. i Niezawodn. - Maint. Reliab. 23 (2021) 619–626, https://doi.org/10.17531/ein.2021.4.4.
[18] J. Herdzik, Remarks of the FMEA procedures for ships with dynamic positioning systems, Zesz. Nauk. Akad. Morskiej w Gdyni 91 (2015) 47–53.
[19] Z. Łosiewicz, Bezpiecze´
nstwo pracy na morzu - werykacja kompetencji zał´
og w realnych warunkach zagro˙
zenia po˙
zarowego statku, Autobusy 6 (2016)
260–263.
[20] B. Ünver, S. Gürgen, B. Sahin, ˙
I. Altın, Crankcase explosion for two-stroke marine diesel engine by using fault tree analysis method in fuzzy environment, Eng.
Fail. Anal. 97 (2019) 288–299, https://doi.org/10.1016/j.engfailanal.2019.01.007.
[21] Marine Diesels Operational Information. The Medium Speed 4 Stroke Trunk Piston Engine Available online: http://www.marinediesels.info/4_stroke_trunk_
piston_engine_access.htm (accessed on Mar 17, 2021).
[22] L. Chybowski, M. Szczepanek, K. Gawdzi´
nska, O. Klyus, Particles Morphology of Mechanically Generated Oil Mist Mixtures of SAE 40 Grade Lubricating Oil with
Diesel Oil in the Context of Explosion Risk in the Crankcase of a Marine Engine, Energies 16 (2023) 3915, https://doi.org/10.3390/en16093915.
[23] N. Hou, N. Ding, S. Qu, W. Guo, L. Liu, N. Xu, L. Tian, H. Xu, X. Chen, F. Zaïri, et al., Failure modes, mechanisms and causes of shafts in mechanical equipment,
Eng. Fail. Anal. 136 (2022) 106216, https://doi.org/10.1016/j.engfailanal.2022.106216.
[24] X.-J. Yi, C.-H. Xu, S.-L. Liu, M.-X. Xing, H. Mu, Failure analysis of repairable systems with k-out-of-m conguration considering with common cause failure and
maintenance correlation based on goal oriented method, Eng. Fail. Anal. 140 (2022) 106615, https://doi.org/10.1016/j.engfailanal.2022.106615.
[25] M. Yucesan, V. Bas
¸han, H. Demirel, M. Gul, An interval type-2 fuzzy enhanced best–worst method for the evaluation of ship diesel generator failures, Eng. Fail.
Anal. 138 (2022) 106428, https://doi.org/10.1016/j.engfailanal.2022.106428.
[26] I. Piotrowski, K. Witkowski, Okrętowe silniki spalinowe, 3. Ed.;, Gdynia, Trademar, 2013. ISBN 978-83-62227-48-8.
[27] International Organization for Standardization ISO 8217:2017. Petroleum products — Fuels (class F) — Specications of marine fuels; 6th ed.; ISO: Geneva, 2017.
[28] B. Monique, Vermeire Everything you need to know about marine fuels, Ghent, Chevron marine products, 2021.
[29] O. Ramadan, L. Menard, D. Gardiner, A. Wilcox, G. Webster, Performance Evaluation of the Ignition Quality Testers Equipped with TALM Precision Package
(TALM-IQT
TM
) Participating in the ASTM NEG Cetane Number Fuel Exchange Program. SAE Tech. Pap. 2017, 2017-01–07, doi:10.4271/2017-01-0720.
[30] CIMAC HFO Working Group Recomendations regarding fuel quality for diesel engines; CIMAC: Frankfurt, 2003; Vol. 21.
L. Chybowski et al.

Engineering Failure Analysis 150 (2023) 107362
11
[31] CIMAC, Recomendations regarding fuel requirements for diesel engines, in: Number 11, CIMAC, London, 1990.
[32] K. Shiode, T. Kijima, G. Fiskaa, Ignition properties for marine fuels established on the fuel ignition analyser FIA 100/2, in: CIMAC Congress 1998, CIMAC,
Copenhagen, 1998, pp. 369–380.
[33] C. Burnete, The study of auto-ignition and combustion qualities of rapeseed oil and diesel fuel mixtures, Agric. – S
¸tiint¸˘
a s¸i Pract. 1–2 (2011) 163–172.
[34] N. Cordos
¸, P. Bere, O. Nemes
¸, Effects of 2-ethylhexyl nitrate on auto-ignition and combustion qualities of rapeseed oil, Stud. Ubb Chem. LVII (2012) 175–184.
[35] T. Ovaska, S. Niemi, T. Katila, O. Nilsson, Combustion property analyses with variable liquid marine fuels in combustion research unit, Agron. Res. 16 (2018)
1032–1045, https://doi.org/10.15159/ar.18.089.
[36] ASTM International D6890–13b -, Standard Test Method for Determination of Ignition Delay and Derived Cetane Number (DCN). of Diesel Fuel Oils by
Combustion in a Constant Volume Chamber, West Conshohocken, ASTM, 2014.
[37] M. Bonisławski, M. Hołub, T. Borkowski, P. Kowalak, A Novel Telemetry System for Real Time, Ship Main Propulsion Power Measurement, Sensors 19 (2019)
4771, https://doi.org/10.3390/s19214771.
[38] T. Borkowski, P. Kowalak, J. My´
sk´
ow, Vessel main propulsion engine performance evaluation, J. KONES 19 (2012) 53–60.
[39] J. Downes, Y. Pu, Reliability-based sensitivity analysis of ships, Proc. Inst. Mech. Eng. Part M J. Eng. Marit. Environ. 219 (2005) 11–23, https://doi.org/
10.1243/147509005X10468.
[40] L. Chybowski, K. Gawdzi´
nska, R. Laskowski, Assessing the unreliability of systems during the early operation period of a ship-A case study, J. Mar. Sci. Eng. 7
(2019), https://doi.org/10.3390/jmse7070213.
[41] Energy Institute IP 541/06: Determination of ignition and combustion characteristics of residual fuels - Constant volume combustion chamber method; Energy
Institute Publications: London, 2006.
[42] C. Røjgaard, Fuel and Lube Oil; MAN Diesel A.S, 2008.
[43] Ship Operations Cooperative Program Introduction to FIA - 100/3 Available online: http://www.socp/projects/completedproj/BunkerFuelOil/Introduction_to_
FIA.pdf (accessed on Oct 25, 2006).
[44] H. Kondoh, T. Kawano, K. Masuda, Evaluation of Ignition Characteristics on Blended Residual Oil relaimed from Waste Plastics, J. Japan Inst. Mar. Eng. 37
(2002) 806–810, https://doi.org/10.5988/jime.37.806.
[45] M. Sawicka, M. Szczepanek, Legal conditions regarding the energy efciency of shing vessels, Zesz. Nauk. Akad. Morskiej w Szczecinie, Sci. Journals Marit.
Univ. SzczecinScientic Journals Marit. Univ. Szczecin 36 (2013) 170–174.
[46] Hill, P.; David, J.L. Fuel cat nes - problems and mitigation Available online: https://iumi.com/images/gillian/London2013/Wednesday1809/Paul Hill John L
David_FINAL.pdf (accessed on Dec 22, 2022).
[47] MAN B&W Diesel Fuel oil specication. In Project Guides L21/31; 2003; p. B 11.
[48] Joint Hull committee JHC Report - Marine Engine Damage due to Catalytic Fines in Fuel (Executive Summary) Available online: https://iumi.com/images/
documents/JHC_Catnes_Pack.pdf (accessed on Dec 27, 2022).
[49] L. Chybowski, Study of the Relationship between the Level of Lubricating Oil Contamination with Distillation Fuel and the Risk of Explosion in the Crankcase of
a Marine Trunk Type Engine, Energies 16 (2023) 683, https://doi.org/10.3390/en16020683.
[50] K. Nozdrzykowski, Z. Grządziel, R. Grzejda, M. Warzecha, M. Stępie´
n, An Analysis of Reaction Forces in Crankshaft Support Systems, Lubricants 10 (2022) 151.
[51] J. Włodarski, Stany eksploatacyjne okrętowych silnik´
ow spalinowych, Gdynia, Wydawnictwo Uczelniane WSM w Gdyni, 1998.
[52] L. Piaseczny, Technologia napraw okrętowych silnik´
ow spalinowych, Gda´
nsk, Wydawnictwo Morskie, 1992.
[53] D. Kazienko, The analysis of class survey methods and their impact on the reliability of marine power plants. 55 Sci. J. Marit. Univ. Szczecin, no. 55 / 2018
2019, 55, 34–43, http://doi.org/10.17402/299.
L. Chybowski et al.