SAE TECHNICAL PAPER SERIES

2006-01-2633

Three-dimensional Transient Simulation of an Intake Manifold using CFD Techniques

Cesareo de La Rosa Siqueira CFD Servios e Sistemas de Computao LTDA (SMARTtech Fluidos) Martin Poulsen Kessler CFD Servios e Sistemas de Computao LTDA (SMARTtech Fluidos) Luiz Augusto Rachid de Araujo FPT Powertrain Technologies Erlon Castro Rodrigues FPT Powertrain Technologies

FILIADA

XV Congresso e Exposio Internacionais da Tecnologia da Mobilidade So Paulo, Brasil 21 a 23 de novembro de 2006

AV. PAULISTA, 2073 - HORSA II - CJ. 1003 - CEP 01311-940 - SO PAULO SP

2006-01-2633

Three-dimensional Transient Simulation of an Intake Manifold using CFD Techniques

Cesareo de La Rosa Siqueira and Martin Poulsen Kessler
CFD Servios e Sistemas de Computao Ltda. (SMARTtech Fluidos)

Luiz Augusto Rachid de Arajo and Erlon Castro Rodrigues

FPT - Powertrain Technologies
Copyright 2006 Society of Automotive Engineers, Inc

ABSTRACT The proposed paper aims to present a threedimensional transient CFD (Computational Fluid Dynamics) analysis of the flow inside an automotive intake manifold conducted by SMARTtech Fluidos for FPT Powertrain Technologies (site at Betim, Brazil). The more important goals of the study are the determination of pressure distribution on each of the four runners during four cycles of operation and the analysis of flow uniformity (characterized by velocity distribution) at the outlet surface of the intake runners. Three different plenum geometries were tested and a comparison among them was conducted in order to verify which one presents the best performance. Transient simulations have been performed in order to take into account the sequential opening/closing of intake engine valves during the engine firing sequence. INTRODUCTION Following [1], usually the try-and-error approach to develop intake-manifolds, consisting of designing a product that would meet the requirements, building a prototype, running it on a test bench, and finding out that it has not met the targets providing the desired flow, is usually timeconsuming. Coming back and define a new design to fix the problems, would have taken so much time, resulting in problems to meet the project deadline. This process can easily take months to be completed for a new intake manifold design. On the other hand, with the use of CFD, a significant number of plenum design changes, including new designs of intake runners, can be fast performed if the developing staff (product engineers, CAD engineers, etc.) takes part in the simulation process. The relevance of this work is to evaluate numerically the pressure distribution and mass flow rate on each of the four runners during four cycles of operation, considering three different plenum geometries and visualize details of

the flow in areas that are extremely difficult (or may be impossible) to adopt measuring techniques. Eventually the boundary conditions can not be properly controlled during an experimental analysis, and the measuring procedure may influence on the results due to the disturbance caused on the fluid flow. Such studies are usually carried out at the concept phase of a vehicle or equipment where design changes are less costly to implement. Another practical application of this investigation is the observation of the flow path lines inside the intake manifold, which can identify relevant recirculation areas that will affect performance and increase pressure drop over the system. The uniformity of the flow at the outlet of each runner , in order to obtain a balanced air flow load into the engine cylinders, and the mass flow rate at the inlet of the intake manifold are also items of careful consideration. The transient simulation provides more realistic results when compared to steady state analysis and gives better results when compared to testing. However, in the latter case it usually takes longer time and requires reasonable boundary conditions. In this situation, the coupling between a general purpose 3D CFD code and 1D system simulation may be necessary. The frequent occurrence of fluid flow phenomena, including heat transfer, mass transfer and chemical reaction in industry has ensured that Computational Fluid Dynamics (CFD) is now part and parcel of Computer Aided Engineering (CAE) repertoire.

METHODOLOGY The baseline geometries for the intake manifolds were provided by FPT- Powertrain Technologies (Betim site, Brazil). Only the volume defined by the surfaces occupied by the fluid (air) is relevant to the CFD analyses, and has been provided by FPT in a given CAD format. In this work, three different plenum geometries have been analyzed for the operation conditions for a given engine. For the considered cases, a depression in the outlet of each runner has been applied during an interval of time equivalent to the opening/closing engine valves. This span of time has been calculated based on the frequency (cycles per minute) during engine operation. According to FPT Powertrain Technologies experience, it takes around 5 complete engine cycles to get pressure boundary conditions established at the inlet of the intake manifold. The flow was considered incompressible, turbulent, isothermal and transient, where the boundary conditions at the outlet surface of the intake runners were changed automatically during the process, in order to account for the opening and closing of the valves. The three geometries were then imported in GAMBIT, the mesh generator from FLUENT Inc. After some geometry clean-up, where ambiguous or missing lines and gap filling are corrected, a 2D triangular surface mesh was generated over the surfaces of the model. Due to the relative complexity of the volume to be discretized, this surface mesh was then exported to T-GRID, another volume mesh generator from FLUENT Inc. developed for complex domains. This mesh generator does not import CAD formats, and in this sense an initial surface discretization is required prior to the volume meshing. In TGRID, the mesh on the surfaces is re-arranged (smooth process) and 3D tetrahedral cells have been generated. Once this generation process is finished, the 3D mesh is exported to FLUENT. The meshes were generated in Gambit pre-processor and have around 600.000 tetrahedral elements, each. The size of the elements is very suitable for this type of application and the value adopted was 2 mm. This represents a small size, even less than the standard recommended value for this type of application (3 mm). The physical model was set, analyzed and post-processed in Fluent CFD code. At this stage all the set up of the problem is made, which includes applying the boundary conditions, setting up the flow regime, choosing the turbulence model and applying initial conditions. The conservation equations are then solved in the FLUENT code by means of an iterative process. This process runs until a convergence criterion is reached. Finally, post-processing is done in FLUENT code. The results are then analyzed using visualization tools and text reports according projects needs.

At the present work, some additional computational tools have been added to the numerical models developed. The purpose of using such tools is to automate a series of commands instead of entering them repeatedly on the command line. In this sense, the gain in terms of time schedule and project deadline was considerable, since each geometry represents a different plenum using the same boundary conditions for comparison purposes. Generally speaking, generating a mesh from a clean intake manifold geometry, setting-up the problem and solving the transient flow during 5 cycles of engine operation, took less than one week for each plenum design. This represents an important step during the design process, where engineers can obtain information about product performance very fast compared to the traditional try-anderror method. COMPUTATIONAL MODEL A three-dimensional CFD model has been built in order to accomplish the targets involved. All the relevant aspects of the flow have been taken into account and fully isothermal incompressible viscous flow simulation with turbulence modeling for steady state condition has been performed. Incidentally, the standard high (turbulent) Reynolds number k model in conjunction with algebraic law of the wall has been employed to model the turbulence phenomenon. A typical mesh for one of the three intake manifolds is given in figure 1. The surface represented in blue denotes the inlet boundary condition and the surfaces in red the outlet boundary conditions. In this work, a negative pressure is applied during an interval of time in each outlet condition respecting the engine fire sequence. After 4 or 5 cycles of transient simulation, the boundary conditions at the inlet can be considered established and the flow reaches a pattern during the transient analysis.

Figure 1: Fully 3D tetrahedral intake manifold mesh for FPT Powertrain Technologies.

NUMERICAL RESULTS Three different plenum geometries were tested and a comparison among them was conducted in order to verify which one presents the best performance. Transient simulations have been performed in order to take into account the sequential opening/closing of intake engine valves during the engine firing sequence. Table 1 shows operation conditions adopted for all simulations performed in this work. Pressure defined at -50 mBar runners outlet Frequency (to define 1250 cycles per minute time interval of pressure applied at outlet runners) Firing sequence 1-3-4-2 Table 1. Operation conditions for the simulations. This paper presents results for one of the three intake manifolds analyzed in this work. The reason for this is that the behavior of the fluid flow does not change considerably from varying the plenum geometries, since the intake runners were kept with the same dimensions for the three plenum geometries. Figures 2 to 5 show path lines colored by velocity magnitude for one of these intake manifolds geometry. Each of these figures represents a snapshot considering one runner opened at a time, for the 4th cycle of operation. From the analysis of these figures, one can see recirculation areas characterized by small velocities in some regions of the plenum. As regards intake runners, the flow does not present recirculation areas and the lines are somewhat well behaved. Due to expansion in the plenum area, the flow experiments small velocities in the chamber and, as it moves downstream, faces an area reduction in the intake runners region, with consequent velocity increase, as expected from the physical point of view. For the sake of project confidentially, scale values on FLUENTs post processing results have been omitted in this paper.

Figure 3: Path lines colored by velocity magnitude. Snapshot considering runner 3 open. (t = 0.168 s, 4th cycle).

Figure 4: Path lines colored by velocity magnitude. Snapshot considering runner 4 open. (t = 0.18 s, 4th cycle).

Figure 5: Path lines colored by velocity magnitude. Snapshot considering runner 2 open. (t = 0.192 s, 4th cycle). As has been pointed out earlier, the uniformity of the flow at the outlet of each runner (in order to obtain a balanced air flow load into the engine cylinder) and the mass flow rate at the inlet of the intake manifold are relevant items to be considered during the analysis. In this sense, figures 6 to 9 show snapshots of velocity vectors

Figure 2: Path lines colored by velocity magnitude. Snapshot considering runner 1 open. (t = 0.156 s, 4th cycle).

distribution at the outlet surface of each intake runner. In general, the flow is fairly well distributed across the section, where some small recirculation areas of the flow appear at the bottom of these sections.

Figure 6: Velocity vectors at the outlet surface. Snapshot considering runner 1 open. (t = 0.156 s, 4th cycle).

Figure 9: Velocity vectors at the outlet surface. Snapshot considering runner 2 open. (t = 0.192 s, 4th cycle). Figure 10 presents the instantaneous mass flow rate for one of the three intake manifolds over time, during 4 cycles of operations. This result considers the opening/closing of engine valves during operation conditions and presents the mass flow rate at the inlet of the intake manifold.
coletor_3

Figure 7: Velocity vectors at the outlet surface. Snapshot considering runner 3 open. (t = 0.168 s, 4th cycle).

Mass Flow Rate (kg/s)

0

0,05

0,1

0,15 tempo(s) Time (s)

0,2

0,25

Figure 10: Instantaneous mass flow rate (kg/s) for one of the three intake manifolds over time (seconds) during 4 cycles of operation. At the beginning of the simulation, the air comes into the intake manifold when a port is opened. The flow continues to enter until the port is closed, which is represented by a peak in figure 10. Each interval (0.05 s) contains four peaks, which represents the closing of each intake runner port during firing sequence. It can be seen from this figure that there is an equal balance between the mass flow rates that enters in the manifold when each individual port is opened. Figure 8: Velocity vectors at the outlet surface. Snapshot considering runner 4 open. (t = 0.18 s, 4th cycle). Table 2 shows an average of the mass flow rate for the inlet of the intake manifold during the last 4th cycle.

This equilibrium in the uniformity of flow distribution in each intake manifold port becomes quite clear through the analysis of this table. For the sake of confidentiality, the values are considered dimensionless with respect to runner 3 (maximum flow rate). Runner 1 -0.25% Runner 2 -0.219% Runner 3 Runner 4 -1.64%

Mass Flow Rate (kg/s)

1 (reference value ) Table 2. Average mass flow rate for each runner.

CONCLUDING REMARKS A 3D CFD model has been successfully employed to study the internal flow inside an automotive intake manifold for FPT Powertrain Technologies (site at Betim, Brazil). Fully transient isothermal incompressible viscous flow simulations with turbulence modeling have been performed. Important relevant recirculation areas that will affect performance and increase pressure drop over the system have been identified. The uniformity of the flow at the outlet of each runner, in order to obtain a balanced air flow load into the engine cylinders, and the mass flow rate at the inlet of the intake manifold have been computed in this work. Generally speaking, the changes in the plenum geometry without modifying intake runners had not provided significant changes in the flow pattern, and there have not been considerable changes in the mass flow rate among the three alternatives of geometry. The transient simulation provides more realistic results when compared to steady state analysis and gives better results when compared to testing. However, in the latter case it usually takes longer time and requires reasonable boundary conditions. In this situation, the coupling between a general purpose 3D CFD code and 1D system simulation may be necessary in future analysis. ACKNOWLEDGMENTS The main authors (C. R. Siqueira and M. P. Kessler) are grateful to FPT- Powertrain Technologies for allowing them to publish the results of this study. REFERENCES [1] FLUENT Company Distributors and Web Site. [2] FLUENT User Manual (2005). FLUENT Inc., Methodology Manual. [3] Laz Foley, FLUENT Auto communication, May 2005. Team, personal

The appearance of the ISSN code at the bottom of this page indicates SAEs consent that copies of the paper may be made for personal or internal use of specific clients. This consent is given on the condition however, that the copier pay a $ 7.00 per article copy fee through the Copyright Clearance Center, Inc. Operations Center, 222 Rosewood Drive, Danvers, MA 01923 for copying beyond that permitted by Sections 107 or 108 of U.S. Copyright Law. This consent does not extend to other kinds of copying such as copying for general distribution, for advertising or promotional purposes, for creating new collective works, or for resale. SAE routinely stocks printed papers for a period of three years following date of publication. Direct your orders to SAE Customer Sales and Satisfaction Department. Quantity reprint rates can be obtained from the Customer Sales and Satisfaction Department. To request permission to reprint a technical paper or permission to use copyrighted SAE publications in other works, contact the SAE Publications Group.

All SAE papers, standards, and selected books are abstracted and indexed in the Global Mobility Database.

No part of this publication may be reproduced in any form, in an electronic retrieval system or otherwise, without the prior written permission of the publisher. ISSN 0148-7191 Copyright 2006 Society of Automotive Engineers, Inc. Positions and opinions advanced in this paper are those of the author(s) and not necessarily those of SAE. The author is solely responsible for the content of the paper. A process is available by which discussions will be printed with the paper if it is published in SAE Transactions. For permission to publish this paper in full or in part, contact the SAE Publications Group.
Persons wishing to submit papers to be considered for presentation or publication through SAE should send the manuscript or a 300 word abstract of a proposed manuscript to: Secretary, Engineering Meetings Board, SAE.