Journal of KONES Powertrain and Transport, Vol. 19, No.

2 2012

VESSEL MAIN PROPULSION ENGINE PERFORMANCE EVALUATION

Tadeusz Borkowski, Przemysaw Kowalak, Jarosaw Mykw
Maritime University of Szczecin
Engineering Department
Waly Chrobrego Street 1-2, 70-500 Szczecin, Poland
tel.: +48 91 4809400, fax: 48 91 4809575
e-mail: [email protected], [email protected], [email protected]
Abstract
In general, the performance of a ship in service is different from that obtained on shipyard sea trial. Apart from
any differences due to loading conditions, and for which due correction should be made, these differences arise
principally from the weather, fouling and surface deterioration of the hull and propeller. The influence of the weather,
both in terms of wind and sea conditions, is an extremely important factor in ship performance analysis.
Consequently, the weather effects needs to be taken into account if a realistic evaluation is to be made. The primary
role of the ship service analysis is a standard of performance data, under varying operational and environmental
conditions. The resulting information, derived from this data, becomes the basis for operational and chartering
decision. In addition, the part for the data records is to enable the analysis of trends of either the hull or machinery,
from which the identification of potential failure scenarios and maintenance decisions can be derived. The traditional
method of data collection is the deck and engine room log records, and this is the most commonly used method today.
In terms of data processing and capabilities, this method of data collection is far from ultimate, since involves
significant data distortion risk. Instrumentation errors are always a potential source of concern in performance
analysis methods. Such errors are generally in the form of instrument drift or gross distortion of the reading.
However, these can generally be detected by the use of trend analysis techniques. The procedure for the evaluation of
the ships service performance, that relies on proven methods of main propulsion engine service data analysis used
and applied for container vessel  small feeder. The vessel is equipped with indirect main propulsion, driven by means
of modern medium speed engine. The different approach demonstrated to achieve the reliable and accurate main
engine performance. The difference in developed engine power has been found, that corresponds well to registered
sea trial results and engine retrofitting reports done, in order to limit the effective power.
Keywords: ship main propulsion, marine diesel engine, energy efficiency design index,

1. Introduction
In general, the performance of a ship in service is different from that obtained during the
shipyard sea trial. Apart from any differences due to loading conditions, and for which due
correction should be made, these differences arise principally from the weather condition, fouling
and surface deterioration of the hull and propeller [1]. The influence of the weather, both in terms
of wind and sea conditions, is an extremely important factor in ship performance analysis.
Consequently, the weather effects needs to be taken into account if a realistic evaluation is to be
made. The primary role of the ship service analysis is a standard of performance data, under
varying operational and environmental conditions. The resulting information, derived from this
data, becomes the basis for operational and chartering decision. In addition, the part for the data
records is to enable the analysis of trends of either the hull or machinery, from which the
identification of potential failure scenarios and maintenance decisions can be derived [2]. The
traditional method of data collection is the deck and engine room log records, and this is the most
commonly used technique today. In terms of data processing and capabilities, this method of data
collection is far from ultimate, since involves significant data distortion risk. Instrumentation
errors are always a potential source of concern in performance analysis methods. Such errors are
generally in the form of instrument drift or gross distortion of the reading. However, these can

T. Borkowski, P.Kowalak, J. Myskow

generally be detected by the use of trend analysis techniques [3, 4]. The procedure for the
evaluation of the ships service performance, that relies on proven methods and contains service
data analysis of main propulsion engine, used and applied for container vessel  small feeder, that
basic details are shown in Tab. 1. The vessel is equipped with indirect main propulsion, driven by
means of modern medium speed engine. The different approach demonstrated to achieve the
reliable and accurate main engine performance. The main engine performance survey was carried
out as a part research project and appropriate measurements conducted, during the ships steaming
at the Baltic sea. Throughout the sea trials, vessel was in loaded condition and all relevant details
are given in Tab. 2.
Tab. 1. Ship main details

No
1
2
3
4
5
6
7

Description
Class
Length over all
Breadth moulded
Draught design, loaded
Deadweight
Container intake
Main engine

Open top container vessel

134.44 m
22.50 m
8.50 m/ 8.71 m
11000 TDW
872 TEU
MAK 8M43

Tab. 2. Nautical and ambient condition during the sea trials

No
1
2
3
4
5
6

Description
Draught (forward/afterward)
Cargo on board
Sea state  wind force and direction
Barometric pressure
Ambient air temperature
Shaft generator load

8.2 m / 8.6 m
6790.9 mt
3 4oB, SW
1015 hPa
16oC
270kW

2. The main propulsion performance evaluation  scope and procedure

The ship speed and propulsion power analysis recommended by ITTC [5], follows
a methodology ISO [6] and was utilized in the recent work. Preferably, the wind resistance
coefficients of the ship should be obtained from model tests. In most cases, model tests are not
available and environmental influences on the performance of sea trials, speed runs should only be
performed against and with the waves. The correction methods existing so far account for
the influences of waves only for these two conditions; in the case when waves do not come from
the bow or the stern, the correction methods are not sufficiently reliable and the effects of steering
and drift on the ships performance might be underestimated. Hence, no corrections were utilized
in analysis. Main engine performance estimation was performed under ships sea service
condition, in good weather and under loaded condition. The main engine performance, speed and
torque were measured to obtain the brake effective power. The engine power estimation was
attained through three alternative methods and is shown in Fig. 1.
1. engine rotational speed and torque measurement on engine flywheel,
2. rotational speed and torque measurement main propulsion shaft
3. engine mean indicated pressure measurement each engine cylinder.
The electronic MIP (Mean Indicated Pressure) calculator with TDC sensor enables cylinder
direct pressure measurements. With the TDC (Top Dead Centre) sensor connected, the course of
pressure with relation to the crankshaft angle is measured thus enabling calculation of indicated
power and extensive combustion analysis. Powerful analysis software allows the computation of
54

Vessel Main Propulsion Engine Performance Evaluation

power for each engine cylinder. For the purpose of main engine performance, evaluation the ship
engine standard monitoring system was utilized. Shaft torque measurements are generally
conducted with a full-bridge strain gauge rosette excited with a battery box and amplified with
a purpose built amplifier-decoder and transmitted to stationary receiver through antenna as shown
in Fig. 2. Calibration is performed by placing a shunt resistor into one of the arms of strain gauge
bridge, which simulates corresponding strain. The uncertainty of this measurement system consists
of elemental error sources based on: a strain gauge, calibration of measurement system, installation
on a ship and torque calculation. Shaft speed measurements are made with magnetic pulse sensor
and a frequency counter.

Peff

ind

K m

i 1

Fig. 1. Main engine and propulsion effective power evaluation

The number of pulses are counted for a predetermined time and divided into number pulses per
revolution to find shaft rate of revolution. Bias error in pulse count is 1 pulse, there is no
uncertainty in revolution estimation. As time window gets larger, the bias error associated with
shaft rate of revolution drops. However, then the transient changes in the shaft power are not
acquired. For the current work, time window is taken as 1s, as power is calculated once every
second. The propulsion shaft torque measurement telemetry system (presented in Fig. 2, with
specification in Tab. 3) offers wireless transmission of strain gauge signals from rotating shafts.

Fig. 2. The main engine and propulsion effective power measurement setup

The recorded torque data were recalculated into power with aid of a rotational speed signal
collected from the pickup sensors, which are built in torque meters acquisition system. The main
engine performance has been recorded for further analysis that includes:
continuous engine output shaft torque  speed profile,
engine cylinder pressure data,
engine fuel oil consumption,
55

T. Borkowski, P.Kowalak, J. Myskow

engine operational condition parameters.

Tab. 3. Experimental instrumentation details

Specification
Engine torque and rotational speed
Propulsion shaft torque and rotational speed
Indicated cylinder pressure and power

Instrument
Strain gauge type, wireless torque meter, Krauss, type:
TEL1-PCM-STG
Strain gauge type, wireless torque meter, Philips, type:
PR9910
Digital pressure indicator, Leutert, type: DPI-2

3. Results and discussion

The experimental data sets were collected for several different engine load levels, ranging from
approximately 25% of nominal load and up to the maximum acceptable limit for engine operation.
Main propulsion load profile was created as an effect of ships accelerating from dead slow to full
ahead that is presented in Fig. 3. The main propulsion overload can be achieved only by means of
controllable pitch propeller setting. The maximum engine effective power was reached when limits
position of the injection pumps fuel racks and controllable pitch propeller overload system
activated.
Pe

Vs

18

Pe, Ps [kW]

Ps

15

6000

12
9

3000

Vs [kts]

9000

6
3

0
4:48

7:12

9:36

12:00

14:24

16:48

19:12

21:36

0
2:24

0:00

Time [hr:min]

Fig. 3. Main propulsion power demanded by ships acceleration

The recorded data were processed, and extracted results are presented in form of the graphs.
The Fig. 4, is meant to compare the actual engine performance with the test bed results, listed in
manufacturer engine shop trial report. The engine specific fuel oil consumption, cylinder
maximum combustion pressure and turbocharger rotational speed was utilized to validate the
experimental torque  speed data set and measurement methodology. Possible discrepancies
between test bed and sea trials results may indicate improper acquisition installation or other signal
disturbances.
220

220

200
Sea trials - ship

210

180
Pressure [bar]

SFOC ISO [g/kWh]

215

Bed test - engine

205
200
195

Bed test - engine

140
120
100
80

190
185
1000

160

Sea trial - ship

60
2000

3000

4000

5000

6000

7000

8000

40
1000

9000

Engine Effective Power [kW]

2000

3000

4000

5000

6000

7000

8000

Engine Effective Power [kW]

Fig. 4. Engine effective power versus SFOC (left) and maximum combustion pressure (right)

56

9000

Vessel Main Propulsion Engine Performance Evaluation

Based on the engine performance comparison presented in Fig. 4, the measurement

methodology and experimental installation were appraised as free from significant methodology
and experimental installation inaccuracies. It can be noticed that the engine maintains the test bed
performance and operational factors, even after 18000 running hours and over 5 years of
exploitation, that confirms engine excellent condition. The diagram in Fig. 5 presents the engine
effective power comparison derived from the four different sources, caused by controllable pitch
propeller setting increase and relevant ship acceleration.
The engine calculated mechanical efficiency varies from 78% at low load (approximately 25%
of rated power) to 91% at high load (approximately 78% of rated power). Such set of values
reflects expected result for modern large bore marine four-stroke and medium speed diesel engine.
It can be noticed that the engines control system  CAT Dicare, presents values reflecting
indicated power rather than the engines effective power as it can be expected.
8000
Engine brake power

Engine Power [kW]

7000

Propeller shaft power

6000

Engine Indicated power

5000

CAT Dicare power

4000
3000
2000
30

40

50

60

70

80

90

100

Propeller Pitch [%]

Fig. 5. The engine and shaft power comparison

45
40

Injection Pumps
Local Indicator
CAT Dicare
Bed Test

Sea trial max. power

Fuel Rack position [mm]

50

35
30
25
20
2000

3000

4000

5000

6000

7000

Bed test max. power

The diagram on Fig. 6 shows three engines fuel rack indicators comparison with different
alignment, which is adequate to engine developed power. Adjustment of engine control local
indicator and fuel injection pumps are almost parallel to the readings taken during test bed trial.
However, indication given by CAT Dicare system has different span (inclination). The most
important aspect is related to the maximum available engine effective power, which was reached
with local indicator position of 52 mm. This position reflects engine nominal load and is
confirmed by the results observed during test bed operation
However, the maximum developed engine effective power, during the sea trial (reflecting to
the local rack position of 52 mm) was only 6950 kW, despite the test bed result where the engine
developed rated power was 8400 kW, with the same fuel rack setting 52 mm.

8000

9000

Engine Power [kW]

Fig. 6. The engines fuel rack indicators comparison in useful engine load range

57

T. Borkowski, P.Kowalak, J. Myskow

The fuel oil grade has to be taken into account if relation engine effective power and fuel rack
settings is analysed. Due to the obvious differences of the fuel grades used during sea trials and
test bed operation which presents Tab. 4, expected fuel rack position might slightly vary from the
one reached during the test bed trial. This distinction could be calculated, following the fuel oil
property.
The fuel injection pump represents volumetric positive displacement type and the volumetric
calorific value of the fuel rate should be taken into account based on the expression:
WQ

kj
Ws x Uinjection 3
dm

kJ kg
x

kg dm3

(1)

where:
Wv fuel oil volumetric calorific value,
Ws fuel oil specific calorific value,
Pinjection delivered fuel oil density.
Tab. 4. Fuel oil property comparison

Determination

Test bed

Sea trial

830 kg/m3

991 kg/m3

42232 kJ/kg

40380 kJ/kg

50oC (estimated)

132oC

807.3 kg/m3

917.9 kg/m3

Fuel oil density @15oC

Fuel oil lower heating calorific value
Delivered fuel oil temperature
Delivered fuel oil density

The corrected volumetric calorific rates for the fuel oil in service  sea trials and test bed are
presented in Tab. 5. Subsequently, based on the LHV data for both fuel grades, the volumetric
correction factor can be derived. This factor can be utilized for further estimation of the expected
fuel rack position for engine heavy fuel oil operation.
Cv

WvMGO ,
WvHFO

(2)

where:
Cv
volumetric correction factor,
WvMGO volumetric calorific value for MGO,
WvHFO volumetric calorific value for HFO.
Tab. 5. Fuel oil volumetric calorific values and volumetric correction factor

Temperature at injection

Test bed Wv

Sea trials Wv

Cv

20oC

34917 kJ/dm3

0.9421

30oC

34643 kJ/dm3

0.9347

40oC

34368 kJ/dm3

0.9273

50oC

34094 kJ/dm3

0.9199

60oC

33819 kJ/dm3

0.9125

132oC

37064 kJ/dm3

58

Vessel Main Propulsion Engine Performance Evaluation

Finally, the estimated engine fuel rack settings set against realistic sea trials and test bed
operation data set is presented in Fig. 7. The highest line (local indicator), injection pumps, and
Cat Dicare sets denotes engine sea trial results and substantial discrepancy needs to be
emphasized. Thus, it can be pointed that the sea trials engine rack positions are far from the
expectation.
The estimated injection pump rack position tendency, caused by fuel oil grade was confirmed
by the engine maker: When using heavier fuel grades independent on the density of the fuel, the
rated power will be reached at a fuel rack position that is accordingly lower, (engine manual).

Fuel Rack position [mm]

50
45
40
35

Injection Pumps
Local Indicator
CAT Dicare
Bed Test
Expected Rack

30
25
20
2000

3000

4000

5000

6000

7000

8000

9000

Engine Power [kW]

Fig. 7. The engines fuel rack indicators comparison

4. Conclusions
The comparative process of realistic main engine performance related to the effective power
(independently from the fuel rack position) with the test bed shop trial report results provides clear
and consistent view of engine actual condition. Basic set of engine performance factors expressed
by effective power and fuel consumption shows reasonable agreement with factory performance
report. It can be noticed that all recorded engine operational parameters are very close to the new
state. Especially, cylinder maximum combustion pressure follows the values taken during the
engine shop trial test. It indicates that the cylinder combustion process is not altered significantly.
However, due to fuel rack position shift engine is not capable to develop the service and rated
power. Currently, engine fuel rack settings at 52 mm caused 6950 kW of engine developed
effective power, whilst estimation is 8400 kW, with fuel rack position of 48mm. Such enginereduced output may lead to difficulties in achieving the ships service or contract speed. During
sea trials, the maximum available propellers pitch was 93% with shaft generator disconnected.
When steaming with higher amount of reefer containers the maximum available pitch might be
much lower and therefore the vessels required speed cannot be maintained.
The engine effective power evaluated by the engines control system  CAT Dicare
arrangement is far from correctness. Almost parallel course of the sea trial engine rack position
with the test bed results specifies fuel injection pumps assembly incorrectness. Hypothetically,
similar situation may appear when in all fuel injection pumps the control sleeve is misaligned with
control rod by the same value. Another hypothetical reason might be installation of the plungers
with different angle of the control edge than that installed during bed test. The above-mentioned
possible reasons and solutions should be understand as possible and hypothetical only and should
be consulted with the engine maker before the further action.
References
[1] Min, K., Kang, S., Study on the form factor and full-scale ship resistance prediction method,
Journal of Marine Science Technology, 15:108118, DOI 10.1007/s00773-009-0077-y, 2010.
59

T. Borkowski, P.Kowalak, J. Myskow

[2] Andresen, P., Borrod, A. Blanchot, H., Evaluation of the Service Performance of Ships,
Marine Technology and SNAME News, 42,4 ProQuest Central pg. 177, 2005.
[3] Hountalas, D. T., Prediction of marine diesel engine performance under fault conditions,
Applied Thermal Engineering 20, 1753-1783, www.elsevier.com/locate/apthermeng, 2000.
[4] Insel, M., Uncertainty in the analysis of speed and powering trials, Ocean Engineering 35,
11831193, Elsvier, 2008.
[5] International Towing Tank Conference, Recommended Procedures and Guidelines, Testing and
Extrapolation Methods Loads and Responses, Stability Model Tests on Intact Stability, UK 2005.
[6] ISO 15016:2002, Ships and marine technology - Guidelines for the assessment of speed and
power performance by analysis of speed trial data.

60