Embedded Files
TESTING AND DESIGN ITERATIONS
To come up with our final intake manifold design, we went going through an iterative process that consisted in different categories/stages:
Design and Modeling - Different configuration were modeled for the intake and mounting system.
Simulations - Models were tested using Ansys Fluent to evaluate the mass flow of air and the pressure/velocity of profile. In addition, a GT Power model of the engine was set up.
Manufacturing - The CAD models for the intake and mounting systems were manufactured for testing as design iterations were perform, until the final design was obtained.
PhysicalTesting - Physical testing that was used to test the intakes and obtained data directly from engine.
Each category was subdivided into different iterations, to show our walk-through as modifications are made to improve each stage, which lead us to the optimization of our final design.
1 - DESIGN AND MODELING
CAD Modeling was used to design and make modification to the overall intake geometry, as well as to designed the mounting bracket that will be place at the cylinder head. The stage is divided into two parts the iterations for the intake system and the respective ones for the mounting bracket. This stage was necessary and important, as the models were later used in other stages, such as in the simulations and manufacturing. SolidWorks was used to model all the design considered.
Relation to project objectives: The Design and Modeling stage related to our functionality and maximization of performance objectives. This stage was where the geometries of the intake system were determined, to ensure it fit in the Panther Motorsport's car and follow the FSAE rules, while delivering the maximum amount of power.
ITERATION #1:
As mentioned in the Design Ideation, based on the constraints of the project and the researched conducted, the team came up with three initial concept designs for the geometry of the intake manifold. From there, it was determined that the "Initial Design Concept 3", which takes into account plenum volume while having a streamlined-like geometry, would used as the first design iteration to run simulations and testing.
To further understand how the plenum volume affects the performance of the engine, the design was model 4 times each with a different plenum volume thus to compare a small (3.7L) , medium (4.7L) and large (5.3L) size. In addition, to understand the effects of runner length three pairs of runners were model to be tested with infinite plenum volume. The runner lengths varied from 2, 6, 9, and 13 inches. These were determined based on Chrysler Wave Tuning, Chrysler Ramcharger Research, Jim McFarland Torque Peak Equations, as well as GT Power simulations.
In addition, other parameters that we taking into account when designing and modeling, such as 3D printing in two separate half's. This was done for ease of manufacturing, as the printer available to us would not of been able to print the entire intake manifold. Also, allowed to interchange different plenum volumes with runner lengths at ease. Lastly, on the bottom half were the runner lengths are, Bellmouth velocity stacks were modeled to allow smooth and even entry of air at high velocities into the intake tract. To easily manufacture the bottom half, it was also modeled into two separate halves.
The models will be tested using ANSYS and experimentally in the dyno. The results obtained from the simulations and testing, will allow us to modify and improve the geometry of the design.
ITERATION #2:
Changes from Previous Iteration:
An intake with the a smaller volume of 1.7 liters was designed.
Reasoning for changes:
After the dyno runs, a choked flow was not determined with currents intakes. In order to have an intake with the least amount of resonance and pressure losses. A smaller design was done, while keeping the current geometry and fitment in the car. Finding the chocked flow point let us know the least amount of volume to use while making sure that the intake was still functional, with the amount of air going into the engine.
2 - SIMULATIONS
Ansys Fluent is used to conduct flow studies on various intake designs at steady state to characterize and compare the flow characteristics of different intake designs. Different iterations are performed during this study, with different flow parameters being measured on the intake manifold such as mass flow rate going to the cylinder, and pressure drop. The flow simulation models the intake similarly to how it would be tested on a flow bench. Similar set up is used for both iterations with the only difference being the final parameters calculated as results. It is important to mention that for both iterations outlet boundary conditions are applied on only a single runner to better simulate the opening and closing of the valves.
Relation to project objectives: By selecting the model providing the highest mass flow rate, the project goal for optimal performance of reaching at least 80% of the volumetric efficiency is being supported by the fact that a higher amount of air supplied to the engine automatically provides a higher volumetric efficiency.
ITERATION #1
This iteration was performed with the main objective of measuring the drop in pressure between the inlet and the outlet. Boundary conditions, along with other set up parameter, are reported below. A velocity at the inlet.
Simulation Conditions and Parameters:
Inlet Velocity : 15 m/s
Outlet Pressure Runner 1: 70 kPa
Outlet Pressure Runner 2: Wall
Model: K-Omega SST (Turbulence Model)
-- Combines K-Omega and K-Epsilon turbulence models
-- Solves using K-Epsilon for free stream flow which involves (turbulence dissipation and turbulence kinetic energy) and
K-Omega near walls utilizing ( turbulence dissipation rate and turbulence kinetic energy)
Constant Density
-- Since flow is under Mach number 0.3
Coupled
-- Couples equations used to solve for velocity and pressure on each for higher accuracy and convergence.
Tetrahedral Mesh
-- Used for more complex geometries where hex meshes are difficult to utilize.
-- Elements: 2,728,497
-- Nodes: 480,962
Results Obtained:
Results for velocity magnitude and pressure contour are shown for the 3.7 L model. Desired results were not displayed by this iteration. Pressure did not show show significant differences between the inlet and the outlet as it was more realistically expected. A different approach was found to be necessary to evaluate the various intake performances.
ITERATION #2
Changes from Previous Iteration:
Different boundary conditions were applied, and different evaluation parameters were calculated as final results. Inlet and outlet pressure were applied as boundary conditions. Ambient pressure was applied at the inlet instead of velocity, and a more accurate outlet pressure was physically obtained by using a MAP sensor as the engine was running. Mass flow rate was used as the evaluation parameter for the intake by selecting the one that would provide the highest amount of air to the engine. A velocity profile was also analyzed to determine the correct behavior of the flow.
Simulation Conditions and Parameters:
Inlet Pressure : 101,325 kPa.
Outlet Pressure Runner 1: 93 kPa
Reasoning for changes:
Iteration #1 was not able to provide realistic results in terms of the evaluating parameter of pressure drop. The chosen boundary conditions were not ideal for the simulating software to give accurate results at steady state conditions. Therefore, the chosen boundary conditions and evaluating parameter needed to be revised.
Results Obtained:
The following results above finally showed a more realistic result in terms of velocity magnitude and mass flow rate. Velocity is showing accurate values as it is reported by literature [2], the values for the results are all around 70 m/s at the outlet and above 180 m/s at the inlet. Mass flow rate values was used as the comparison parameter to select the model with the highest value and smallest plenum volume for convenience of manufacturing,. Although the 4.3L reported the highest values, the 3.7 L was selected in this iteration for its mass flowrate is really close to the 4.3L but it has a smaller volume.
ITERATION #3
Changes from Previous Iteration:
An additional model was used to performed simulation, using the same set up. This model had a smaller volume of 1.7 liters. An it was expected to give a similar but slightly lower mass flow rate as the 3.7 and 4.3 liters intakes.
Reasoning for changes:
The simulation was performed after a new design based on iteration #2 for Design and Modeling was created with the intent of determining the smallest plenum that would prevent choke flow. The main reason for this simulation was to verify if a further smaller plenum volume compared to the 3.7 L was still able to provide similar results in terms of mass flow rate .
Results Obtained:
As expected, the 1.7 liters intake design gave a mass flow rate similar to the other intakes simulated. It had
By comparing the different mass flow rate value we can see that not a significant difference is shown in the models from 1.7L to 4.3L. However a much cleaner velocity profile can be observed on the plenum wall adjacent to the outlet side. Where the path of the flow can be more easily determined when compared to the other intakes. Therefore, based on the previously established criteria for the simulations, the 1.7L was recommended as the final geometry to be manufactured.
3 - MANUFACTURING
Prototypes were manufactured and tested in an iterative way as improvements were made. An additive manufacturing process, 3D printing, was used to physically construct the three-dimensional prototypes of the various CAD models as it offered minimum limitations for the geometry. Parts were first printed out of PLA material and once the final configuration was decided it was printed out of a different material.
Relation to project objectives: Indirectly the manufacturing stage contributes in the achievements of each of the objectives. The initial prototypes prints, ensured the proper fit of the design in the car and it allowed us to physically test, to compare and analyze the behavior of the different configuration. Later on, the print of the final design was done with the durability goal in mind.
ITERATION #1: 3D Printing
During Senior Design I, initial design concepts where modeled in SolidWorks. From this point, the team conceptualized 8 different intake variations, ranging from four intake manifold systems with the same runner lengths and varying plenum volumes, to 4 sets of runners with different runner lengths and infinite plenum volume, as mentioned in the Design and Modeling Stage "Iteration #1".
For the first part of manufacturing, the team decided to 3D print all the designs, so the prototypes could be used later on to physical testing and collect performance data, such as maximum power and torque. The 3D printing was accomplished by using the software Ultimaker Cura to prepare the print, then a 12in x 12in x 15.5in Creality CR10 3D printer was utilized to print the parts. The material used was PLA filament, which was chosen based on the cost-effectiveness of it, as this material was provided by on the the Panther Motorsports sponsors.
As mention in the Design and Modeling Stage, the main reasoning behind the prototypes being designed in two pieces, was to ease manufacturing for the selected geometry, as the size of the 3D printer available to us limited the size of the part to be printed. In addition, manufacturing in this form allowed us to vary the parameters (plenum volume and runner lengths) and be able to come up with all the configurations to be tested.
Plenum volume: 3.7L
Plenum volume: 4.3L
Plenum volume: 5.3L
Runner length: 6 in
Runner length: 9 in
Runner length: 13.5 in
ITERATION #2: Sealing
Changes from Previous Iteration:
Adding an Epoxy resin coat to the PLA 3D printed prototypes.
Reasoning for changes:
After 3D printing, the models were not perfectly sealed as a natural result of the process. In order to avoid flow escaping the intake system, Epoxy resin was used to fill the small voids between the fibers of the 3D printed models material. In this first sealing process, all models parts were sand and resin was applied after.
ITERATION 3: Vacuum sealing
Changes from Previous Iteration:
The parts were tested for air leakage, using a vacuum pump. The pump was connected and
Reasoning for changes:
After 3D printing, the models were not perfectly sealed as a natural result of the process. In order to avoid flow escaping the intake system, Epoxy resin was used to fill the small voids between the fibers of the 3D printed models material. In this first sealing process, all models parts were sand and resin was applied after.
In the second process of sealing the intake systems, the various 3D printed models, with the resin previously applied and dried, were again sand, bigger holes were sealed using duct tape, and each single intake system was attached to a vacuum pump to be sealed using Epoxy resin again and to allow the resin to penetrate more deep into the material fibers.
ITERATION 4: Final Print
For the final 3D print, it was done in Fused deposition modeling in Ultem 1010 due to its high mechanical strength, impact resistance, and high thermal resistance. Following an ASTM D638 testing method ultimate tensile strength of 80 MPa and 30 MPa on xz, zx orientation respectively. Impact strength was measured using a ASTM D4812 un-notched testing method that provided 260 J/m and 70 J/m with respect to an xz, zx orientation. A glass transition temperature of 410 ℉was also reported following an ASTM D4726 inflection point testing, and it is way above the engine working temperature of 210 ℉. Material properties and final 3D print were outsourced through our sponsor [8]as it will allow for the use of stronger engineering materials, in reference to "Design Ideation" where a matrix was done in order to chose the best overall best material.
4 - PHYSICAL TESTING
Prototypes manufactured were physically tested using an inertial dynamometer, DynoJet 224, to obtain data directly from the Kawasaki Ninja 650cc engine. This allowed us to compare and validate the data obtained through simulations.
Relation to project objectives: A main objective of the project was to produce 80% of stock power through the restrictor. Dynamometer testing allows a comparison from the stock power to restricted power with the prototype intakes. The goal is to test as many iterations during the same atmospheric conditions to reduce variability. Yet, the competition is not on a dynamometer, so testing at the racetrack will be performed to confirm our suggestion as transient throttle response is also important, as the race car does not stay at Wide Open Throttle (WOT) for the duration of competition.
DYNO TESTING
ITERATION #1:
All models were physically tested in the dyno, where we were able to get the maximum power and torque obtained when using the different prototypes:
Restricted intake: 3 varying plenum volume w/ same runner lengths
Unrestricted: 4 infinity plenum volume w/ varying runner lengths
As explained in the in the CAD Modeling portion, the main reason for testing the varying plenum volume intakes was to understand and see the behavior it had on the maximum power and torque, as well as the rpm band at which it occurred. Similarly, we wanted to analyzed the infinite plenum volume and varying runner length prototypes
Each prototype was tested twice and all of the runs were made on the same day, thus to reduce variability factors that could fluctuate the data. It was performed from around 10am to 1pm, with the following conditions:
Ambient temperature = 70.64 ℉
Humidity = 75.25%
Absolute pressure = 30.11 in-Hg
IMG_5756.movBelow, there is a simple sheet that was used to record the maximum torque and power for each design.
ITERATION #2:
Changes to Previous Iteration:
The previous iteration was done with the factory Kawasaki Electronic Control Unit (ECU) and the factory calibration. This iteration was performed with the Haltech Elite 1500 ECU which is what will be used hereupon to competition.
Reasoning for changes: The Haltech ECU allows for changes in calibration and for use of the integrated restrictor throttle body in compliance to the Formula SAE rules. This iteration of testing will be to calibrate the intake prototypes to optimally take advantage of the competition fuel of 100 Octane (R+M)/2 gasoline. We will specifically be using Sunoco 260 GT race fuel for testing.
Observations: From the results, trends in physical properties of the intake manifested themselves to make a more conclusive recommendation. Plenum volume did not reach a choke flow point during the ramp runs on the inertia dynamometer. However, track testing with a driver will be able to give feedback on transient throttle response on tip-in throttle delay with each intake volume iteration.
Overall results of both iteration:
All models were physically tested in the dyno, where we were able to get the maximum power and torque obtained when using the 6 inch 1.7 Liters. For a maximum of 60.05 Hp which is a 94.26% of the unrestricted engines power. For an average torque of 37.80 ft-lbs in comparison to the 39.73 ft-lbs made unrestricted. This shows the engines ability to not drop of its torque value at a certain rpm, but to stay consistent throughout the rpm range.
Driver/On-Track Testing
Since the competition is not on a dynamometer, on-track testing was performed to confirm our suggestion as transient throttle response is also important, as the race car does not stay at Wide Open Throttle (WOT) for the duration of competition. Based on the driver’s feedback, the intake was shown to have a linear throttle response.
WhatsApp Video 2022-04-25 at 12.04.34 PM.mp4ENGINE INTAKE MOUNT
In order to be in compliance with the FSAE rules, the intake manifold must be securely attached to the engine block or cylinder head with brackets and mechanical fasteners. Hose clamps, plastic ties, or safety wires do not meet this requirement [7]. The stock engine uses rubber bushings to attach the injector housing, which are only allowed for sealing air passages but not as a structural attachment. Thus an injector housing mounting bracket was designed and manufactured, which later was TIG welded onto the injector housing so it could be bolted on top of the engine block. This helped us satisfy the our functionality goal, as it is one first things that judges look when inspecting the vehicle at competition.
Unfortunately, either structural nor modal analysis were performed on the piece as it was out of the scope of the project, but it should be considered for further projects.
Page updated
Report abuse