Académique Documents
Professionnel Documents
Culture Documents
Available at www.sciencedirect.com
Argonne National Laboratory, 9700 South Cass Avenue, Argonne, IL 60439, USA
BMW Group, 80788 Munich, Germany
c
BMW Group, 5900 Arcturus Avenue, Oxnard, CA 93033, USA
b
article info
abstract
Article history:
This article summarizes the testing of two BMW Hydrogen 7 Mono-Fuel demonstration
The BMW Hydrogen 7 Mono-Fuel demonstration vehicles are derived from the BMW
27 August 2008
Hydrogen 7 bi-fuel vehicles and based on a BMW 760iL. The mono-fuel as well as the bi-fuel
vehicle(s) is equipped with cryogenic hydrogen on-board storage and a gaseous hydrogen
Keywords:
well as emissions on the Federal Test Procedure FTP-75 cold-start test as well as the
BMW Hydrogen 7
highway test. The results show that these vehicles achieve emissions levels that are only
Efficiency
a fraction of the Super Ultra Low Emissions Vehicle (SULEV) standard for nitric oxide (NOx)
Emissions
and carbon monoxide (CO) emissions. For non-methane hydrocarbon (NMHC) emissions
the cycle-averaged emissions are actually 0 g/mile, which require the car to actively reduce
emissions compared to the ambient concentration. The fuel economy numbers on the
FTP-75 test were 3.7 kg of hydrogen per 100 km, which, on an energy basis, is equivalent to
a gasoline fuel consumption of 17 miles per gallon (mpg). Fuel economy numbers for the
highway cycle were determined to be 2.1 kg of hydrogen per 100 km or 30 miles per gallon
of gasoline equivalent (GGE).
In addition to cycle-averaged emissions and fuel economy numbers, time-resolved (modal)
emissions as well as air/fuel ratio data is analyzed to further investigate the root causes of
the remaining emissions traces. The BMW Hydrogen 7 vehicles employ a switching
strategy with lean engine operation at low engine loads and stoichiometric operation at
high engine loads that avoids the NOx emissions critical operating regime with relative air/
fuel ratios between 1 < l < 2. The switching between these operating modes was found to
be a major source of the remaining NOx emissions.
The emissions results collected during this period lead to the conclusion that the BMW
Hydrogen 7 Mono-Fuel demonstration vehicles are likely the cleanest combustion engine
vehicles ever tested at Argonnes APRF.
2008 International Association for Hydrogen Energy. Published by Elsevier Ltd. All rights
reserved.
* Corresponding author. Tel.: 1 630 252 3003; fax: 1 630 252 3443.
E-mail address: twallner@anl.gov (T. Wallner).
0360-3199/$ see front matter 2008 International Association for Hydrogen Energy. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijhydene.2008.08.067
7608
Nomenclature
APRF
BMS
CH4
CO
CO2
CVS
DC
DOE
ECU
EURO
NCAP
FID
FTP
GGE
H2
H2O
1.
Introduction
2.
2.1.
Background
HC
HP
kg
km
kW
l
MJ
mpg
N2
NDIR
Nm
NMHC
NOx
ppm
RPM
SULEV
THC
hydrocarbon
horsepower
kilogram
kilometer
kilowatt
relative air/fuel ratio (lambda)
megajoule
miles per gallon
nitrogen
Non-Dispersive Infrared
Newton meters
non-methane hydrocarbons
nitric oxides
parts per million
revolutions per minute
Super Ultra Low Emissions Vehicle
total hydrocarbons
2.2.
2.3.
BMW 760Li
w100
4
6.0 L V12
6-speed automatic
2-wheel
2460 kg (5420 lb)
3124 mm (123 in)
Gaseous port injection
Series development
hydrogen vehicle
Naturally aspirated
Lean burn/stoichiometric
3-way catalyst
2.4.
Differences between mono-fuel demonstration
vehicle and bi-fuel vehicle
Several changes were made to the BMW Hydrogen 7 bi-fuel
vehicle for the mono-fuel demonstration application. These
changes include hardware adaptations as well as software
and calibration adjustments. On the hardware side the most
significant changes are the removal of gasoline fuel system
including fuel injectors, fuel lines, charcoal filters for tank
ventilation and fuel rail. The two high-pressure fuel pumps
were also removed, which reduces the parasitic losses on the
engine. For stability reasons the gasoline fuel tank remains in
the vehicle because it is a structural element. The vehicles are
equipped with improved catalysts. A schematic of the catalyst
setup is shown in Fig. 3 [8]. The catalyst setup consists of two
monoliths; the first one covers the stoichiometric operating
regime whereas the second one reduces NOx peaks that occur
when switching from lean to stoichiometric operation. Also,
one of the two mono-fuel vehicles tested at Argonne used
non-petroleum based engine oil (however the results do not
show a measureable difference). Changes to the software
include an adapted calibration of the engine control unit (ECU)
as well as the shift characteristics.
2.5.
7609
7610
LH2-Tank
1) Side view
12 V + Cable
Feed Line (GH2)
Gasoline Tank
Batteries
Boil-Off-Management-System
(exhaust)
2) Top view
Engine Cover /
Hydrogen Rail
3) Isometric view
Liquid hydrogen
Vacuum insulated cryogenic tank
1
170 L
5.1 bar (74 psi)
20 K
8 kg (8 GGE)
Generation II cryo nozzle
1.53 W
16 g/h
33 g/h
Monolith 1
Monolith 2
600 cells
120 150 g/ft
400 cells
150 g/ft
Fraction (Pt)=max
Fraction (Pd)=max
Fraction (Rh)=60
Fraction (Pt)=12
Fraction (Pd)=2
Fraction (Rh)=1
3-way-concept
for general
emissions reduction
A schematic of the engine operating strategy implemented in the BMW Hydrogen 7 vehicles is shown in Fig. 4
[11]. For low engine loads the engine is operated in the lean
burn region. Load adjustment is accomplished by solely
adjusting the relative air/fuel ratio, which also minimizes
losses usually caused by use of a throttle. For higher load
demands the engine switches from lean operating to stoichiometric operation and thereby avoids operation at the NOx
critical air/fuel ratio regime. Stoichiometric operation allows
using a 3-way catalyst effectively in order to reduce nitric
oxide emissions using unburned hydrogen as a reducing
agent.
3.
3.1.
Vehicle dynamometer
3.2.
Facility instrumentation
3.3.
3.4.
7611
7612
H2
metering
and safety
panel
H2 safety and
emergency
system
H2 exhaust
content
sensor
H2 delivery
hose 250
psi
Outside safety
and conditioning
system
Hydrogen line
Air
handling
unit
H2 12 pack
cylinders at
2000 psi /
~7.3 kg
Test cell
Calibration
gas room
Conditioning room
Control room
Outside
Emissions bench
Data
acquisition
system
60
40
20
4.5
1.5
500
1000
1500
2000
0
2500
4.
Fuel consumption measurement and
validation
The BMW Hydrogen 7 vehicles have a fairly complex fuel
system that does not allow tapping into it for direct fuel
consumption measurement. For that reason the preliminary
tests in fall of 2007 were used to implement an advanced
approach for fuel consumption measurement developed for
hydrogen vehicles. This approach requires measurement of
H2O [Vol-%]
3
2
5.
Test program for BMW Hydrogen 7
vehicles
The test program for the BMW Hydrogen 7 mono-fuel
demonstration vehicles focused mainly on the Federal Test
Procedure FTP-75 cycle. With the vehicle soak area as well as
the dyno cell equipped with deer antlers each of the two BMW
Hydrogen 7 Mono-Fuel demonstration vehicles could be
tested on a cold-start test once per day. This was intended to
generate robust and statistically proven results. In addition to
using two vehicles the results of two drivers operating the
mFuel
1.5
1
0.5
100
200
300
CO, HC [ppm]
H2O
CO
HC
7613
Intake air
Humidity
Dilution air
Humidity
Sample
point
[H2O]
[H2]
0
400
Time [sec]
Fig. 8 Influence of increased water content on CO and HC
analyzer readings.
0.8
0.6
0.4
6.2.
0.2
0
25
15
5
-5
-15
0
150
300
450
600
Difference [%]
7614
-25
750
Time [sec]
Fig. 10 Correlation between fuel consumption
measurement methods for steady-state operating points.
6.
Results
6.1.
The fuel economy numbers for the BMW Hydrogen 7 MonoFuel demonstration vehicle are summarized in Table 3. The
average fuel consumption during the FTP-75 cold-start test is
3.7 kg of hydrogen per 100 km. When assuming a lower
heating value of gasoline of 41.5 MJ/kg and a density of
0.765 kg/l, 1 kg of hydrogen has the same energy content than
1 gallon of gasoline. Converted to the widely used miles per
gallon (mpg) scale results in approximately 17 miles per gallon
of gasoline equivalent (GGE) on the FTP-75 cycle. The equivalent numbers for the highway cycle are 2.1 kg/100 km and
30 miles/GGE. These numbers are fairly good considering the
weight of the vehicle and the large size of the engine.
A sample plot of a post-processing sheet used for determining the fuel economy and emissions numbers is shown in
3.7
17
2.1
30
6.3.
Modal results
7615
Fig. 11 Sample result sheet for FTP-75 test on a BMW Hydrogen 7 vehicle.
The impact of the changes in relative air/fuel ratio on the
nitric oxide emissions is shown in Fig. 15. The trace shows the
modal NOx readings of the diluted exhaust stream during
a FTP-75 cold-start test. During most of the test cycle the
emissions trace actually matches the zero-line. Four phases
during the entire test are relevant for the NOx emissions
measurement. These are the two engine starts as well as the
two fast accelerations during the second hill of the 505 cycles.
The first engine start shows slightly higher spikes than the
second one, which might be due to the cold catalyst during the
7616
10% SULEV
3.9%
0%
0.3 %
NOx
NMHC
CO
0.0008 g/mi
0 g/mi
0.003 g/mi
0% SULEV
60
40
20
-20
-40
-60
0
500
1000
1500
2000
1
2500
Time [sec]
10% SULEV
first start. The second critical phase during the fast acceleration shows a very distinct spike during the second 505 cycle
and only a minimal blip during the first 505 cycle. The reason
for the spikes during the acceleration is the switching from
lean operation to stoichiometric operation (see Fig. 14). The
engine switches every cylinder individually (possible with
individual port fuel injection) from lean burn to stoichiometric
operation. The average exhaust air fuel ratio during this
transition is lean resulting in a low catalyst conversion ratio
while some cylinders run stoichiometric and consequently
exhaust high levels of NOx. These two peaks during the fastest
accelerations occurred throughout the testing performed; the
height of the spikes did vary significantly.
The modal diluted total hydrocarbon (THC) emissions are
plotted in Fig. 16. The horizontal section during the 10 min
soak shows the background level of THC emissions in the
dilution air. The BMW Hydrogen 7 vehicle is capable of
actively reducing the THC concentration as can be seen by the
fact that the average curve during engine operation is lower
than the one during the soak phase. The most interesting
phases during vehicle operation are again the engine starts as
well as the switch-over from lean to stoichiometric operation.
During the first engine start the THC emissions decrease from
an increased level. This is likely due to the catalyst warm-up.
The second engine start only shows a minor peak during that
phase. Both switching phases from lean to stoichiometric
operation show a distinct THC emissions peak. Unlike NOx
emissions that can only be converted at stoichiometric operation, the oxidation of hydrocarbons in the catalyst occurs
60
40
20
4.5
1.5
2%
0% SULEV
0.5%
0.12 %
NOx
NMHC
CO
0.0001 g/mi
0.0002 g/mi
0.0012 g/mi
500
1000
1500
2000
0
2500
Time [sec]
Fig. 15 Modal nitric oxide (NOx) emissions during FTP-75
test.
7617
60
40
20
1.6
1000
1500
2000
20
1000
100
10
0.4
2500
500
60
40
20
2000
60
40
20
4.5
1
2500
Time [sec]
Fig. 17 Modal methane (CH4) emissions during FTP-75
test.
1.5
500
1000
1500
2000
0
2500
Time [sec]
Fig. 19 Modal carbon monoxide (CO) emissions during
FTP-75 test.
Modal CO [ppm]
1500
1
2500
1000
2000
engine operation are lower than the background concentration indicating that the vehicle actively reduces emissions.
However, the CH4 trace also shows two significant spikes
when the vehicle switches from lean operation to stoichiometric operation during the hard acceleration of the second
hill in each 505 cycle.
Both, the peaks in total hydrocarbons as well as methane
are significantly higher than the ambient concentration of the
respective concentration. Therefore the peaks cannot just be
attributed to a lack of emissions conversion efficiency of the
catalyst. The flame ionization detectors used for measuring
hydrocarbon emissions are sensitive to changes in humidity.
The sudden increase in humidity caused by switching from
lean to stoichiometric operation could increase the hydrocarbon readings (see [17]). Flame ionization detectors use
hydrogen as a burner gas. A sudden increase in hydrogen
concentration as seen during the transition from lean to
stoichiometric operation (see Fig. 18) could also cause
increased hydrocarbon readings. There are several other
possible causes for the HC peaks including for example actual
500
1500
Time [sec]
Time [sec]
1000
Modal H2 [ppm]
0.8
500
40
1.2
60
7618
references
7.
Conclusions
Acknowledgements
The submitted manuscript has been created by UChicago
Argonne, LLC, Operator of Argonne National Laboratory
(Argonne). Argonne, a U.S. Department of Energy Office of
Science laboratory, is operated under Contract No. DE-AC0206CH11357. The U.S. Government retains for itself, and others
acting on its behalf, a paid-up non-exclusive, irrevocable
worldwide license in said article to reproduce, prepare derivative works, distribute copies to the public, and perform
publicly and display publicly, by or on behalf of the
Government.
This research was funded by DOEs FreedomCAR and
Vehicle Technologies Program, Office of Energy Efficiency and
Renewable Energy. The authors wish to thank Lee Slezak and
Gurpreet Singh, program managers at DOE, for their
continuing support of hydrogen activities.
Testing of the BMW Hydrogen 7 vehicles was only possible
with the extensive support through BMW of North America,