National Technical University of Athens
Mechanical Engng Department
Thermal Engng Section
NATIONAL TECHNICAL UNIVERSITY OF ATHENS
MECHANICAL ENGNG DEPARTMENT
THERMAL ENGNG SECTION
INTERNAL COMBUSTION ENGINES LABORATORY
STUDY CONCERNING THE EFFECT OF FUEL ADDITIVES ON PERFORMANCE
AND SOOT EMISSIONS OF DIESEL ENGINES
Professor Assoc.: D.T. Hountalas
FINANCING: MSTP HELLAS Ltd.
ATHENS March 2001
National Technical University of Athens
Mechanical Engng Department
Thermal Engng Section
1. SCOPE OF THE RESEARCH
It was assigned to the NTUA Internal Combustion Engines Lab (ICEL), on behalf of company
MSTP HELLAS to conduct an investigation to validate the effect of a fuel and lube oil additive on the
specific fuel consumption and pollutant emissions of Marine Diesel Engines. It is a product that the
previous company supplies to the market and its composition is unknown to ICEL Lab.
To investigate the effect of the additive a detailed experimental investigation was conducted on the
engine of Vessel Aptera which belongs to ANEK LINES.. The additive was supplied by MSTP Hellas
which also managed to install a device in the engine room through which the additive was fed directly
on the fuel circuit towards the fuel pumps of the engine. The supply was made using a dose pump, at a
percentage defined by MSTP Hellas. The same procedure was followed for the lube oil additive as
well..
Two sets of measurements were conducted, a first without additive and a second with the use of
the fuel and lube oil additives in various proportions and combinations. During all tests the same fuel
quality was used from the vessels tanks that had the same preheat temperature and density.
Finally it is mentioned that all measurements were taken with the cooperation of the crew and their
assistance was direct and essential. As far as soot measurements are concerned the sampling was made
from the crew and the measured data (Bosch paper) where presented immediately to the Technical
Staff of the vessel for observation. The measured data (Bosch Paper Filters) are available in our
laboratory.
2. EXPERIMENTAL SETUP
2.1 Description of the Experimental Installation
The experimental investigation was made at sea in the engine room on engine No.2. The basic data
of these engine are:
- Engine Type MAN4054 18V
- Maximum Power Output 10000 KW at 500 rpm
- 4 Stroke
- One Turbocharger
- One Air Cooler
For the measurements it was used a portable sampling system which consist of :
a) Laptop Computer
b) Analog to Digital Converter
National Technical University of Athens
Mechanical Engng Department
Thermal Engng Section
c) Kistler Piezotron Pressure Transducer
d) Amplifier for pressure signal.
e) Special software for sampling and measurement processing.
For the analysis of gas emissions it was used a portable gas analyzer of company Kayne-May using
electrochemical sensors which was previously calibrated. For soot measurement it was used a portable
Bosch device. In the next Table 1a are given the data of all instrumentation and their characteristics,
that were used and the measured pollutants as well.
Table 1a
Pollutant
Manufacturer
Type
Soot
CO
NO
CO2
O2
Bosch
Kane-May
EFAW-65
KM 9106
Operating
Principal
Optical
Electro Chemical
Sensors
2.2 Engine Description
The engine used for the test is a marine engine MITSUBISHI 4054 18V, 4 stroke, 18 cylinder
supercharged. Its main characteristics are :
Max. Power Output
:10000 PS
Oper. Speed Range
:230-500 rpm
Bore
:400 mm
Stroke
:540 mm
Compression Ratio
:12.5:1
No. of Cylinders
:18
Layout
:V
No. of T/C
:1
The engine is mounted to a pitch propeller through reducing gear device.
National Technical University of Athens
Mechanical Engng Department
Thermal Engng Section
3. EXPERIMENTAL PROCEDURE
Before the measurements the engine was prepared so that it can operate with and without the use
of the fuel additive. This work was made by MSTP Hellas Ltd through the installation of a special
dose pump on the fuel supply line of the engines high pressure pumps.
The experimental investigation consisted of the following:
Measurements at 410 rpm engine speed using load indicator of 38 at a full scale of 42 without using
additive in the fuel oil. Following this cylinder pressure diagrams were taken together with pollutant
emission measurements at steady conditions. Pollutant emissions were measured in the exhaust duct
at the Turbine inlet.
Following this and using the fuel oil additive measurements was taken of cylinder pressure and
pollutants at a load index of 38 where all peripheral measurements were similar to the previous case
were no additive was used. to the case using the additive. These measurements were taken during
the return trip of the vessel to Piraeus and after using the additive for 8 hours. Thus the time the
additive was used was rather limited.
Afterwards and after 1 and 1/2 month measurements were repeated from the same engine at the
same conditions using only fuel additive in percentages of 0.3% and 0.1%. Then measurements
were conducted using the lube oil additive and fuel oil additive also in percentages 0.3% and 0.1%
. Thus a total of 4 additional measurements.
All measurements conducted together with their numbering are provided in Table 1b which follows:
Table 1b
.
Load
Indicator
Lube Oil
Additive
38
Percentage
of Fuel Oil
Additive
%
0
1
2
38
0,3
NO
38
0,3
NO
38
0,1
NO
38
0,3
YES
38
0,1
YES
NO
National Technical University of Athens
Mechanical Engng Department
Thermal Engng Section
3.2 Description of Symbols-Parameters
To comprehend the results of the measurements in the following diagrams, it is given below the
nomenclature of the main symbols and parameters used:
bsfc =brake specific fuel consumption (g/PS h)
BSN =Bosch smoke index
ppm =parts per million
4. ANALYSIS OF EXPERIMENTAL RESULTS
4.1 Specific Fuel Consumption
In Table .2 are given the specific fuel consumption for the various conditions examined. The
values refer to tests with and without the additive at operating conditions corresponding to a load
index (constant) of 38.
In the same Table is given the total power output of the engine and the total fuel consumption and
specific consumption as well. The values have been derived after processing the measured data using
the diagnosis EDS-2 with which the measurements were taken. The values of brake specific fuel
consumption for cases No.5,6. will be revised after the completion of the experimental investigation
for the estimation of the effect of the lube oil additive on engine mechanical losses (mainly frictional
losses). Thus at the current stage it can not be revealed the effect of the lube oil additive on specific
fuel consumption.
Observing the values we observe that the power with the use of the additive is higher while the load
indicator is constant and the peripheral measurements are practically the same, this is positive for the
operation of the engine (as far as exhaust gas temperature levels are concerned).
Table 2
.
Total
Total
Specific Fuel
Difference
Power
Consumption
kg
PS
g/PSh
Load
Percentage of
Lube Oil
Indicator
Fuel Additive
Additive Consumption
%
1
38
NO
821.4
5065
162.17
0.0
38
0,3
NO
820.7
5192
158.08
-2.6
38
0,3
NO
854.7
5444
157.00
-3.1
38
0,1
NO
846.1
5383
157.18
-3.0
38
0,3
YES
845.0
5352
157.89
-2.6
38
0,1
YES
848.9
5383
157.70
-2.8
National Technical University of Athens
Mechanical Engng Department
Thermal Engng Section
From Table 2 it is revealed that for load index of 38 we have a reduction of specific fuel consumption
in the order of 2.6% with the first use of 0.3% additive which reaches almost 3% after use of 40 days.
Also we observe that this reduction can be achieved with a percentage of fuel additive of 0.1% as well.
This values taking into consideration the size of the engine and its total consumption is very
significant. The reduction of specific fuel consumption is straightforward in all cylinders
In Fig. 1 is given the total consumption for all cylinders as estimated from the diagnosis system for
cases No. 1, 2 and 4 of the previous Table. Similar results are observed for cases 3, 5-6. The total
consumption appears higher when using the additive but the power is also higher as revealed from Figs
1 and 2. For this reason it is used for comparison the specific fuel consumption which as revealed from
Fig.3 is lower for all cylinders when using the additive.
National Technical University of Athens
Mechanical Engng Department
Thermal Engng Section
70
Engine Speed 410 rpm
Total Consumption:821.4 kg/h
Total Consumption:820.7 kg.h
Total Consumption:846.1 kg/h
0% Fuel Additive
0.3% Fuel Additive
Fuel Consumption (kg/h)
60
Specific Consumption:162.17
Specific Consumption:158.08
Specific Consumption:157.18
0.1% Fuel Additive
50
40
30
20
0
9 10 11 12 13 14 15 16 17 18
Engine Cylinder (-)
Fig.1
450
Total Power:5065 PS
Total Power:5192 PS
Total Power:5383 PS
425
Cylinder Brake Power (PS)
400
410 rpm Engine Speed
0% Fuel Additive
0.3% Fuel Additive
375
0.1% Fuel Additive
350
325
300
275
250
225
200
175
150
125
100
0
9 10 11 12 13 14 15 16 17 18
Engine Cylinder (-)
Fig.2
National Technical University of Athens
Mechanical Engng Department
Thermal Engng Section
Brake Specific Fuel Consumption (g/PSh)
180
Speed : 410 rpm
175
0% Fuel Additive
170
0.3% Fuel Additive
Reduction With 0.3%=-2.6%
Reduction With 0.1%=-3.0% After 40 days
Mean Value:162.17
Mean Value:158.08
Mean Value:157.18
0.1% Fuel Additive
165
160
155
150
145
140
135
130
0
9 10 11 12 13 14 15 16 17 18
Engine Cylinder (-)
Fig.3
4.2 Soot Emissions
A significant parameter for Diesel engines is soot emission. Its value indicates the quality of
combustion which contributes to the maintenance of the exhaust and the injection system in a
good condition. In Table 3 are displayed the values of soot emission as they were measured
with and without the use of fuel additive at both load indexes were mentioned. Additional
measurements were taken with the use of fuel additive at load index 38. For this reason, the
exact time of each measurement is displayed on Table 3.
National Technical University of Athens
Mechanical Engng Department
Thermal Engng Section
Table 3
.
Date
Time
05/02/01 21:15
Fuel
Additive
%
0
06/02/01 21:45
Lube Oil
1st
2nd
2nd
2nd
2nd Mean
Additive Reading Reading Reading Reading Reading Value
NO
0.90
1.0
1.0
0.90
1.30
1.02
0.3
NO
0.60
0.70
0.50
0.50
0.40
0.54
17/03/01 21:25
0.3
NO
0.60
0.60
0.40
0.50
0.40
0.50
17/03/01 23:50
0.1
NO
0.50
0.40
0.40
0.40
0.43
18/03/01 21:12
0.3
YES
0.30
0.40
0.40
0.38
0.37
18/03/01 23:46
0.1
YES
0.20
0.30
0.30
0.20
0.25
It can be observed from Table 3 that there is a straightforward reduction of soot emission values when
using the fuel additive. The mean value before using the additive is 1,02 but with the first application
of the additive is becomes 0.54. The reduction is in the order of 50% and is considered to be very
significant.
Using the additive for 40 days the value has been reduced to 0.50 for a percentage of 0.3% and to
0.43 for a percentage of 0.1%. Significant reduction is also observed when using the lube oil additive
where the final value is close to 0.25 , i.e. reduction of soot about 85%. The value is considered to be
very significant. At this point we must mention that results are satisfying with the use of 0.1% after a
percentage of 0.3% has been used for a serious period.
4.3 Composition of Exhaust Gas-Gaseous Pollutant Emissions
In Table 4 are given the gases concentrations within the exhaustgas, at the point where soot
emissions were measured. No significant differences are observed for O2 and CO2.
For CO and NO there are no noticeable differences values being practically the same with and
without the use of the fuel oil additive. The differences are mainly due to differences in load between
the tests.. Therefore, the use of fuel oil additive does not cause a noticeable positive or negative effect
on gaseous pollutants values emitted from the engine exhaust system.
National Technical University of Athens
Mechanical Engng Department
Thermal Engng Section
Table 4
. Percentage of Lube Oil
Fuel Additive Additive
Date
Time
CO
ppm
CO2
%
O2
%
NO
ppm
4.4
NO
05/02/01
21:15
198
6.3
12.5
1800
0.3
NO
06/02/01
21:45
169
6.2
12.6
1803
0.3
NO
17/03/01
21:25
195
6.3
12.5
1924
0.1
NO
17/03/01
23:50
224
6.4
12.4
2076
0.3
YES
18/03/01
21:12
217
6.4
12.4
1977
0.1
YES
18/03/01
23:46
219
6.1
12.8
1916
Cylinder Combustion Pressures
4.4.1 Cylinder Pressure Diagrams
In Figs. 4.1 and 4.2 are given the cylinder pressure diagrams as measured from the diagnosis system
before using the additive (No.1) and after its first use (No.2). Similar results where observed and for
cases No.3-6 which for the sake of space are not mentioned here. As far as the pressure diagrams are
concerned when the lube oil additive was used these have shown no differences as expected.
To derive a conclusion it is given in Fig.5 the peak cylinder pressure for all cylinders for cases 1,2
and 4 from where it results that the use of the additive causes in general a small increase of peak
pressure in most cylinders for both percentages of fuel oil additive used, 0.3 %( No.3) and 0.1%
(No.4). At this point it is required to be very careful since during the measurement procedure it is
possible to have deviations that can effect the final conclusion. For this reason we always compare
individual cylinder pressure diagrams before and after the use of the additive before a safe conclusion
is derived.
10
National Technical University of Athens
Mechanical Engng Department
Thermal Engng Section
110
Cylinder Pressure (bar)
100
Reference Measurement
NO Additive
1 2
Cylinder Number:
6
9 10 11 12 13 14 15 16 17 18
90
80
70
60
50
40
30
20
10
0
100
200
300
400
Measured Point (-)
500
600
Fig.4.1
110
With Additive:8h After Use
1 2
Cylinder Pressure (bar)
100
Cylinder Number:
6
9 10 11 12 13 14 15 16 17 18
90
80
70
60
50
40
30
20
10
0
100
200
300
400
Measured Point (-)
Fig.4.2
11
500
600
National Technical University of Athens
Mechanical Engng Department
Thermal Engng Section
110
Engine Speed:410 rpm
Max. Combustion Pressure (bar)
0% Fuel Additive
105
0.3% Fuel Additive
0.1% Fuel Additive
100
95
90
85
80
0
9 10 11 12 13 14 15 16 17 18
Engine Cylinder (-)
Fig.5
Then, through Figs. 6.1-6.4 it is given the comparison of cylinder pressure diagrams of some
cylinders that where measured with 0%, 0.3% and 0.1% fuel oil additive to detect its effect on the
combustion mechanism. Observing these figures it is obvious that we have an acceleration of
combustion around cylinder TDC which reveals the effect of the additive. This effect can explain the
observed reduction of soot values and the slight improvement of specific fuel consumption.
12
National Technical University of Athens
Mechanical Engng Department
Thermal Engng Section
120
410 rm Engine Speed
Cyl.No.1
110
0% Fuel Additive
Cylinder Pressure (bar)
100
0.3% Fuel Additive
90
0.1% Fuel Additive
80
70
60
50
40
30
20
10
0
60
90
120
150
180
210
240
270
300
270
300
Engine Crank Angle (deg)
Fig.6.1
120
Engine Speed:410 rpm
Cyl. No:3
110
0% Fuel Additive
Cylinder Pressure (bar)
100
0.3% Fuel Additive
90
0.1% Fuel Additive
80
70
60
50
40
30
20
10
0
60
90
120
150
180
210
240
Engine Crank Angle (deg)
Fig.6.2
13
National Technical University of Athens
Mechanical Engng Department
Thermal Engng Section
120
Engine Speed:410 rpm
Cyl. No:7
110
0% Fuel Additive
Cylinder Pressure (bar)
100
0.3% Fuel Additive
90
0.1% Fuel Additive
80
70
60
50
40
30
20
10
0
60
90
120
150
180
210
240
270
300
Engine Crank Angle (deg)
Fig. 6.3
120
Engine Speed:410 rpm
Cyl. No:14
110
0% Fuel Additive
Cylinder Pressure (bar)
100
0.3% Fuel Additive
90
0.1% Fuel Additive
80
70
60
50
40
30
20
10
0
60
90
120
150
180
210
240
Engine Crank Angle (deg)
Fig. 6.4
14
270
300
National Technical University of Athens
Mechanical Engng Department
Thermal Engng Section
4.4.2 Mean Cylinder Peak Pressure Values
To evaluate the general effect of the fuel oil additive on the peak cylinder pressure level it is given
in Table 5 the mean peak cylinder pressure value as estimated from the processing of measured data.
As observed the use of the additive results to no significant increase of maximum pressure except for
cylinders having lower injection timing where combustion is accelerated due to the improvement of the
fuel cetane number that the additive causes..
Table 5
.
4.5
Lube Oil
Additive
Mean Cylinder Peak
Pressure
Percentage of Fuel Oil
Additive
%
0
NO
91.0
0.3
NO
92.6
0.3
NO
92.7
0.1
NO
92.1
0.3
YES
91.2
0.1
YES
91.0
Heat Release Rate
In Figs. 7.1 and 7.2 is given the heat release rate inside the combustion chamber for load index of 38
without and with the use of the additive in percentages 0.3% and 0.1% for Cylinders No. 1 and 3. As
observed with the use of the additive the heat release rate is more intense at the first stages of
combustion. This has a positive effect on the produced work and the specific fuel consumption as
already mentions. Similar results are observed when using 0.1% of additive as well. The use of the
lube oil additive does not effect combustion as expected..
15
National Technical University of Athens
Mechanical Engng Department
Thermal Engng Section
8000
Engine Speed:410 rpm
Cyl.No.1
7000
Heat Release Rate (J/deg)
0% Fuel Additive
6000
0.3% Fuel Additive
5000
0.1% Fuel Additive
4000
3000
2000
1000
0
-1000
160
170
180
190
200
210
Engine Crank Angle (deg)
Fig.7.1
8000
Engine Speed:410 rpm
Cyl.No.3
7000
Heat Release Rate (J/deg)
0% Fuel Additive
6000
0.3% Fuel Additive
5000
0.1% Fuel Additive
4000
3000
2000
1000
0
-1000
160
170
180
190
Engine Crank Angle (deg)
Fig.7.2
16
200
210
National Technical University of Athens
Mechanical Engng Department
Thermal Engng Section
4.5
Effect of the Additive On the Injector Quality of Injection
Processing the measured data with the diagnosis system the injection quality of injectors is derived.
In Table 6 it is given the mean condition of all injectors for all measurements conducted. Form this it is
revealed that the use of the additive results to an improvement of injection quality which as shown by
Table 7 is straightforward eliminating the possibility of computational error.
Table 6
.
Percentage of Fuel Oil
Additive
%
Lube Oil Mean Quality of
Additive Injection for All
Injectors
%
NO
80
0.3
NO
87.1
0.3
NO
87.6
0.1
NO
86.7
0.3
YES
85.9
0.1
YES
86.1
5. STUDY TO DEFINE THE EFFECT OF THE LUBE OIL ADDITIVE
The evaluation of the fuel and lube oil additive was based on the processing of measured cylinder
pressure data. From the cylinder pressure diagrams it is derived the indicated power output of each
cylinder and following this using a mechanical efficiency is concerted to brake power on the engine
shaft. The relations which are used for this process are quite accurate and the resulting error for a
specific engine is very small. This is certified from the analysis of engine shop and sea trials.
Thus when the fuel oil additive is evaluated there is no significant error since any such error exists
in both measurements i.e. without and with the additive. But in the case of the lube oil additive this
affects the frictional losses of the kinematics mechanism and not the combustion. This has as a results
that any effect on the specific fuel consumption cannot be detected through the cylinder pressure
diagram. For this reason it was conducted in the NTUA ICEL s specific measurement to determine the
effect of the lube oil additive on engine frictional losses.
For this reason a special engine was used , a Ricardo E-6 single cylinder test engine which is
mounted to an electric brake which permits the estimation of mechanical losses through motoring. For
this reason two sets of measurements where conducted under motored conditions to estimate the
effect of the lube oil additive on frictional losses. Since in this engine mechanical losses are relatively
high due to its robust nature and its small size it was made an effort to estimate the absolute value of
17
National Technical University of Athens
Mechanical Engng Department
Thermal Engng Section
friction reduction and to express this as percentage of total power output of the engine at full load. In
this manner it is possible to determine the effect of the additive on specific fuel consumption.
It must
be noted that the specific engine is in excellent mechanical condition.
Before the measurement the lube oil was replaced with another of the type SAE15-40 and then
motoring tests were conducted at rotational speeds of 500-2000 rpm at intervals of 250 rpm and the
electric brake torque reading was stored. Measurements initiated when the temperature of the lube oil
was steady at 60 C and the cooling water temperature was at 55 C. This temperature was maintained
steady with the use of coolers. The process was repeated when using the lube oil additive which was
used at a percentage of 6%. It must be noted that the amount of engine lube oil is 9 lt.
In the following Table 7 are given the readings of both tests.
Table 7
No.
Additive
1
2
3
4
5
6
7
Mean
Power
Increase %
Engine
Speed
500
750
1000
1250
1500
1750
2000
Torque Reading
lb
NO
4.65
4.65
5.30
5.60
6.15
6.70
7.05
YES
4.15
4.45
4.80
5.10
5.65
6.05
6.50
Water Temp.
C
NO
55
55
55
55
54
54
54
YES
55
55
54
54
54
54
53
Lube Oil
Temp.
C
NO YES
61
61
61
61
62
63
61
61
61
61
61
61
61
61
Diff. In
HP
Difference as
% of Full
Load
0.07
0.05
0.15
0.18
0.21
0.33
0.29
+3.5%
+3.5%
+3.4%
+4.2%
+3.3%
+3.6%
For the specific engine and based on its operation at 1000 rpm up to 2000 rpm it is observed using the
lube oil additive an increase of power output in the order of 3.% which is straightforward and
observed in all measurements. This results reveals that for the specific engine the use of the lube oil
additive results to a decrease of specific fuel consumption in the order of 3.6% due to the reduction of
frictional losses. At this point we must note that the specific engine is in very good condition so that
not to have increased mechanical losses due to malfunction etc. The motoring method used is very
accurate and reveals immediately the trend for reduction or not of mechanical losses since its a direct
method for measuring their value.
18
National Technical University of Athens
Mechanical Engng Department
Thermal Engng Section
6. FINAL RESULTS OF THE STUDY
From the analysis of the previous results the following conclusions, that reveal the effect of the Fuel
and Lube oil additives of MSTP HELLAS on the performance and pollutant emissions of the engine
MAN4054 18V of vessel APTERA which belongs to ANEK LINES.
a) It is observed with the use of the fuel additive a reduction of specific fuel consumption in all cylinders
that where measured which is straightforward. The reduction is for fuel oil additive percentage of 0.3%
is in the order of 2.6% immediately after its first use. After using the product for 40 days the reduction
reached a value of 3% and remained the same when the percentage of the additive was reduced to
0.1%. As far as the effect of the lube oil additive is concerned and due to the difficulties that have
already been mentioned we use the conclusions derived from the experimental investigation in the ICEL
lab of NTUA on the prototype engine. Based on the test data on this engine it is expected a reduction of
specific fuel consumption about 3.5% due to the effect of the additive. This reduction is solely due to
the reduction of frictional losses.
b) No considerable variation of gaseous pollutants was observed with the used of the fuel oil additive as
expected. On the contrary it is important that no increase was observed especially in CO and NO as
observed with many additives.
c) There is a significant reduction of soot emissions when using the fuel oil additive observed in all
measurements. With the first use after 8 hours there was observed a reduction of soot in the order of
50% and after 40 days using a percentage of 0.3% it was about 55%. The reduction of soot after this
point with percentages of 0.3% and 0.1% of fuel additive was similar after using of course a higher
percentage (0.3%) for a long period. With the use of the lube oil additive it was observed a further
reduction of soot that reached a value of 75% after using the product for 12 hours. The reduction was
the same for both percentages of fuel oil additive 0.1 and 0.3%. This the combination of the two
additives can results to a decrease of soot in the order of 75% on the specific engine.
d) There is due to the use of fuel oil additive an acceleration of combustion which is revealed by a slight
increase of peak pressure (1-2 bar) in some cylinders and especially from the heat release rate. In
general the use of the additive does not result to a serious change of cylinder combustion pressure
especially in cylinders that have a proper injection timing. From the present analysis it results that the
mean cylinder pressure in all cases was practically the same without and with the use of the additive.
19
