Use Use of Euler of Eulerian ian multi multi-ph phas ase e - - PowerPoint PPT Presentation

1

Ford Motor Co. CONFIDENTIAL

Use Use of Euler

• f Eulerian

ian multi multi-ph phas ase e (E2P) (E2P) algo algorith rithm m to to mode model l leak leakag age e of

• f

hydrocarbon hydrocarbon vapors vapors duri during ng vehi vehicl cle e re refueling fueling

Star Global Forum 2009 December 8 Syed Ali & Marek Lockhart Ford Motor Company

2

Ford Motor Co. CONFIDENTIAL

Agenda

• Background

Refueling systems Fill Nozzle Capless filler pipe

• Star E2P model and BCs
• Results
• Conclusion
• Q & A

3

Ford Motor Co. CONFIDENTIAL

Standard & ORVR Systems

Fuel tank Fill pipe Evaporative system with Carbon canister Recirculation line (RL) To engine

4

Ford Motor Co. CONFIDENTIAL

Fuel Storage and Delivery System

Fuel System

• Fill pipe
• Storage Tank
• Carbon Canister

System

5

Ford Motor Co. CONFIDENTIAL

Nozzle Geometry & Air flow

Air flow rate into nozzle shut-off mechanism Contents of this slide have been derived from the SAE Paper # 970311

Fuel, GPM Air, LPM Air/Fuel Ratio 4.0 3.3 0.218 7.0 3.9 0.147 10.0 4.0 0.106

6

Ford Motor Co. CONFIDENTIAL

Fill Pipe Geometry

FP entrance Drain hose Recirc Line Fill pipe Flexible Hose

7

Ford Motor Co. CONFIDENTIAL

Problem Definition

• Early ORVR SHED test of the first vehicle with

capless FP failed at 4 GPM fill rate.

• CAE was tasked to investigate!
• In general, FP flows are both multi-phase

(Liq./Gas) and multi-constituent (Air/HC vapors).

• Since the flow in the capless part is gas only,

liquid flow in the FP can be omitted.

8

Ford Motor Co. CONFIDENTIAL

About E2P – from Star notes

• Both phases coexist everywhere in the flow domain:

concept of “Interpenetrating continua”.

• Conservation equations for mass, momentum and

energy are solved for each phase.

• Cover the full range of volume fractions, from 0 to 1.
• Mean quantities: velocity, pressure, volume fraction are
• btained directly.
• Turbulence can be included at little extra cost.

9

Ford Motor Co. CONFIDENTIAL

E2P Model for Fill Pipe

• A simplified, gas-only model has been solved

using the CFD code StarCD.

• Air and hydrocarbon (HC) flows are solved

using the Eulerian-2-Phase model.

• Estimates of both air and HC flow rates are

used as input at various ports.

• The discharge of HC vapor from the drain port

is used as the metric of performance.

10

Ford Motor Co. CONFIDENTIAL

E2P Model for Fill Pipe

• The CFD models have about 1.7 million cells.
• Boundary conditions are derived from test data.
• Steady state solution is run with double

precision.

• Need over 15,000 iterations for good

convergence.

11

Ford Motor Co. CONFIDENTIAL

BCs for Multiphase Flow

Air from ambient Flow of HC/air from fuel tank Flow of HC/air into refueling nozzle Pressure boundary condition

12

Ford Motor Co. CONFIDENTIAL

Results of Design 1

Run # 4 GPM 10 GPM 4 GPM 10 GPM 1 0.20966 0.066942 0.31944 0.103775 2 0.39084 SHED Test CAE

Design 1- BC CAE driven design - D2 The tables present mass of HC/Gallon. Acceptable emission: 0.2 g HC/gallon

13

Ford Motor Co. CONFIDENTIAL

ORVR Test & CAE Results

The tables present mass of HC/Gallon. Acceptable emission: 0.2 g HC/gallon

Run # 4 GPM 10 GPM 4 GPM 10 GPM 1 0.20966 0.066942 0.04868 0.02440 2 0.39084 0.04870 0.02440 3 0.00830 0.01700 Design 1 Design 2

Rig test CFD

4 GPM 10 GPM 4 GPM 10 GPM 0.31944 0.103775 2.140E-07 3.116E-07 Design 1 Design 2

14

Ford Motor Co. CONFIDENTIAL

D1.

4 GPM 10 GPM Q(HC)/Q(Air) 0.759 0.173 Q(HC)/Total 0.431 0.148 Total Q, lpm 1.610 3.816

Drain port flow ratios from CFD simulations

D2.

4 GPM 10 GPM Q(HC)/Q(Air)

• 5.7E-07

6.3E-07 Q(HC)/Total

• 5.7E-07

6.3E-07 Total Q, lpm 0.818 2.688

Results D1, D2 & D1-modified

D1 Modified – RL rotated 30 degrees

4 GPM 10 GPM Q(HC)/Q(Air)

• 6.667E-06
• 1.716E-06

Q(HC)/Total

• 6.667E-06
• 1.716E-06

Total Q, LPM 1.027 2.850

15

Ford Motor Co. CONFIDENTIAL

Results D1 @ 4 GPM

HC vapors leaving the drain port – plane near the top. Figure 1. Contour plots of concentration of HC (red) in the capless insert

16

Ford Motor Co. CONFIDENTIAL

Results D1 @ 10 GPM

Flow near bottom of drain port. No flow was observed near the top of the drain port. Figure 2. Contour plots of concentration of HC (red) in the capless insert

17

Ford Motor Co. CONFIDENTIAL

Results D2 @ 4 GPM

No HC flow observed in the vicinity of the drain port. Figure 3. Contour plots of concentration of HC (red) in the capless insert

18

Ford Motor Co. CONFIDENTIAL

Results D2 @ 10 GPM

Figure 4. Contour plots of concentration of HC (red) in the capless insert

19

Ford Motor Co. CONFIDENTIAL

D1-Modified; Results @ 4GPM

Contour plots of concentration of HC (red) in the capless insert Velocity vectors in the capless insert.

20

Ford Motor Co. CONFIDENTIAL

Conclusions

• The simple gas-based multi-phase

model works quite satisfactorily to model flow of hydrocarbon vapors in the filler cup – both standard and capless filler pipes.

• The methodology has been instrumental

in several programs to guide the design

• process. Both time and money have

been saved.

21

Ford Motor Co. CONFIDENTIAL

Conclusions

• The investigation exhibits the important

role CAE tools play to better reveal the physics related to failures.

• The model will be extended to include

the liquid fuel/air flow into the fill pipe. This will allow the model to naturally ingest the ambient air.

22

Ford Motor Co. CONFIDENTIAL

Q & A