SCR Performance Optimization Through Advancements in Aftertreatment PackagingJared Blaisdell, Andrew Gilb, Phebe Preethi, Joe Sweeney, Karthik Viswanathan, Paul Way, Nathan Zambon
DEER 2008
Outline Optimization Through Aftertreatment Packaging
•
Urea Doser Integration–
Urea deposit formation & chemistry
–
Eliminating deposit formation
•
Urea Preparation–
Mixer design & simulation
–
Urea solution vaporization
–
Flow distribution
–
Urea distribution & mixing
•
Single & Dual Wall Packaging–
NOX
reduction impact
–
Skin temperature
•
Summary
Control Strategy
Urea Dosing System
Optimized Packaging
Catalysts
Doser Integration What Are Deposits?
232
322 )(CONHOHHNCOHNCONHNHCO
+→++⎯→⎯Δ
Step 1 (Vaporization): Water evaporates from spray dropletStep 2 (Decomposition): Urea thermolysis & hydrolysis reaction
Incomplete decomposition results in deposit formation
High Performance Liquid Chromatography
Confirms composition of collected deposits:• Urea• Melamine• Cyanuric acid• Biuret
AU
-0.10
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
1.10
1.20
1.30
1.40
1.50
1.60
1.70
Minutes2.00 2.20 2.40 2.60 2.80 3.00 3.20 3.40 3.60 3.80 4.00 4.20 4.40 4.60 4.80 5.00 5.20 5.40 5.60 5.80 6.00 6.20 6.40 6.60 6.80 7.00 7.20 7.40 7.60 7.80 8.00 8.20 8.40
Cyanuric Acid
UreaBiuret
Melamine
Thermogravimetric Analysis
•
Deposit decomposition requires significant energy & time
•Prevention through proper doser integration & urea preparation
Doser Integration Understanding Deposit Chemistry
0
20
40
60
80
100
0 100 200 300 400 500 600 700Temperature [C]
Wei
ght C
hang
e [%
]
UreaBiuretMelamineCyanuric AcidDeposits with DPF
Urea Doser Integration Deposit Formation & Root Cause
•
Deposits form in three areas:
–
Injector location (spray tip/boss)
•
Cool tip/boss temperatures, direct spray impingement, spray recirculation
–
Interior wetted walls
•
Direct spray impingement on cool wall surface
–
Catalyst surface
•
“Wet”
urea contacting catalyst surface (poor urea mixing/preparation)
Injector Tip/Boss
Wetted Body/Pipe Walls
Urea Doser Integration Eliminating Deposits
•
Eliminated wall wetting at injection location and pipe walls (CFD)
•
Eliminated flow recirculation zone (CFD)
•
Incorporate Wire-mesh “accumulator”–
Direct spray impingement–
Eliminates wall wetting–
Re-directs urea spray
Recirculation
Straight-line Flow
Urea Preparation Mixer Design and Simulation
•
What is the “best”
mixer type?–
In-pipe mixers
–
In-body mixers
–
Multiple injection locations
–
Multiple mixer variations
•
Water injection used to simulate urea
•
Model Outputs:
–
% Vaporization
–
H2
O vapor uniformity (γH2O
)
–
Velocity uniformity (γVel
)
–
Backpressure penalty
• Simulation/Test Modes:
–
332g/s @ 500˚C exh
(high flow)–
151g/s @ 290˚C exh
(low flow)
A
B
C
D
E
Urea Preparation Mixer Design and Simulation
•
Ideal mixer determined to be in-body design–
Low backpressure penalty
–
Maximum mixing/vaporization
–
Maximum injection location flexibility
BP (kPa)
γH2OHigh Flow
γH2OLow Flow
γVelHigh Flow
γVelLow Flow
A –
Continuous Body & On-Mixer
Doser 2.2 0.77 0.90 0.95 0.98B –
DPF Outlet Mixer & On-Mixer
Doser 2.0 0.97 0.98 0.96 0.99C –
SCR Inlet Mixer & On-Pipe
Doser 2.6 0.87 0.92 0.94 0.97D –
SCR Inlet Mixer & On-Mixer
Doser 1.6 0.82 0.93 0.94 0.98
E –
In Pipe Vortex & On-Pipe Doser 1.9 0.96 0.87 0.97 0.99
Incomplete vaporization at SCR face
In-Body
In-Pipe
Urea Preparation Simulation Validation
•
In-body mixer refined via simulation
•
Simulation results compared against engine test:–
15 liter, Cummins ISX 500
–
342 g/s
@ ~477 C 1,616 ml/hr injection
–
190 g/s
@ ~429 C 638 ml/hr injection
–
224 g/s
@ ~204 C 1,741 ml/hr injection
•
Baseline (no mixer) compared to with-mixer case
Gas Sensing ArrayMixer
Empty Inlet(normally DOC + DPF)
Single 10.5”
x 6”
SCRpromotes ammonia slip for distribution measurement
Flow Distributor
Urea Injection/Mixer
Model SCR InletγH2O
= 0.66Model SCR Inlet
γH2O
= 0.69Model SCR Inlet
γH2O
= 0.67
•
Linear flow with some turbulence (limited vaporization time)
•
> 75% of urea impacts core face as liquid
•
Poor reactant distribution at the SCR core face
Mode 1 Mode 2 Mode 3
Baseline Simulation Results
Urea Preparation Simulation Validation
Model SCR InletγH2O
= 0.97Model SCR Inlet
γH2O
= 0.97Model SCR Inlet
γH2O
= 0.96
Mode 1 Mode 2 Mode 3
•
Strong vortex = increase vaporization time
•
No liquid urea escapes
•
Uniform distribution
Mixer Simulation Results
Urea Preparation Simulation Validation
Urea Preparation Measured Ammonia Slip Distribution
γNH3
= 0.81 γNH3
= 0.90 γNH3
= 0.97
γNH3
= 0.55 γNH3
= 0.55 γNH3
= 0.56
Mode 1 Mode 2 Mode 3
Bas
elin
eW
ith M
ixer
Significant Urea Pooling
No Pooling = Increased SlipCompared to Baseline
Urea Preparation On-Engine Performance Evaluation
•
2004 CAT C7
•
600 HP eddy-current dynamometer
•
Pseudo FTP cycle–
Cold start transient (duration: 20 min)
–
20 min hot soak
–
Hot start transient (duration: 20 min)
•
3 steady-state SET modes •
1901 RPM, 175 ft-lb (B25)
•
1901 RPM, 525 ft-lb (B75)
•
2240 RPM, 611 ft-lb (C100)
•
Alpha = NH3
:NOX
= 1.0 while exhaust > 200˚C
Empty
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
1 2 3Mode
% N
ox R
educ
tion
With Mesh No MixerNo Mesh With MixerWith Mesh With Mixer
SCR Inlet Temp: 300 C
SCR Inlet Temp: 450 C
SCR Inlet Temp: 483 C
Urea Preparation On-Engine Performance
Evaluation
FTP NOx Red No Mixer With Mixer
No Accumulator
47.2 % 442ppm NH3
74.9 % 15ppm NH3
With Accumulator
62.5 % 132ppm NH3
81.9 % 15ppm NH3
Cycle Peak Slip
With Accumulator, No Mixer
No Accumulator, With Mixer
With Accumulator, With Mixer
Packaging Style Dual or Single Wall?
•
Single or dual wall drivers:
-
Skin temperature: dual wall has lower skin temperature
-
Heat retention: improved cold start NOx reduction
-
Cost: dual wall is higher cost
•
Performance tested
-
Cold start + hot start FTP
-
Steady state, ~ 480˚C exhaust gas temperature
Test System
050
100150200250300350400450
Skin
Tem
pera
ture
(C)
Dual WallSingle Wall
Test Cell Evaluation Impact of Dual Walled Packaging –
NOX
ReductionFTP %NOx Reduction
Dual Wall (System) 77.2%
Single Wall (SCR only) 75.7%
Post-D
PF
Post-M
ixer
SCR
Post-SCR
Exhaust Temp ~ 480˚C
Summary•
Doser integration priorities–
Zero wall wetting–
Zero recirculation–
In-pipe accumulator reduces deposits
•
Urea preparation significantly impacts system performance–
In-pipe mixers require long pipe lengths for full vaporization–
In-body swirl mixer provides the best performance:•
100% vaporization •
excellent flow distribution •
6”
package space, doser flexibility–
Combination of in-body mixer with in-pipe accumulator offers performance and deposit advantages
•
Single or dual wall packaging? –
Little to no impact on NOx reduction over cold + hot FTP–
Skin temperature, cost -
primary drivers–
Impact on deposit formation needs to be evaluated
Control Strategy
Urea Dosing
Optimized Packaging
Catalysts
