Overview of plastic injection molding technology Injection molding technology is the process of injecting molten plastic to fill the cavity of the mold.. Plastic injection machine struct
INTRODUCTION
Rationale
CTM (Constant-Torque Mechanism) is an important aspect in the field of mechanics and engineering, widely applied in many different fields including technology, healthcare and daily consumer products CTM is defined as the ability to provide stable output torque without being affected by variations in the input rotation angle
Currently, there are many articles and studies on CTM mechanism such as Chia-Wen Hou, Chao-Chieh Lan researched “Functional joint mechanism with constant-torque outputs” [1], Phan Thanh Vu, Pham Huy Tuan researched “Design and Analysis of a Compliant Constant-Torque Mechanism for Rehabilitation Devices” [2], Piyu Wang, Sijie Yang and Qingsong Xu researched “Design and Optimization of a New Compliant Rotary Positioning stage with Constant Output Torque” [3], Hari Nair Prakashah, Hong Zhou researched
“Synthesis of Constant Torque Compliant Mechanisms” [4], many of these papers and studies primarily concentrate on optimizing design and developing CTM mechanisms with a broad constant torque region, without delving into the creation of prototypes with varied geometric designs to assess the degree of influence of geometric designs on torque
Today, to meet practical demands, models designed according to the CTM must exhibit large and consistent torque with adequate flatness Plastic materials are frequently utilized for designing structures due to their simplicity, lightweight nature, and cost- effectiveness Thanks to their excellent mechanical properties, wear resistance, dimensional stability, and lack of lubrication requirement, as well as elasticity To address this issue, our team has undertaken a study involving different models varying in style and size, all designed according to the CTM mechanism Prototypes constructed will undergo evaluation through experimentation to compare the impacts of model variations and size on torsional strength
Continuing from there, the group narrowed down our choice of thesis: “ Study on the effect of geometric shape on the torsional strength of injection molded ”
Scientific and practical significance of the thesis
- Scientific significance: applying knowledge about plastic injection molds, delving into study to design molds effectively
- Practical significance: the project contributes to the creation of structures that help people in restoring joint function, and devices that support human mobility.
Scopes
- The thesis limits the scope of knowledge to the design and fabrication of prototypes with different geometric parameters by plastic injection molding method, samples with constant torque mechanism
- The product has PP plastic material After injection molding, the product will be tested for torsional strength by torque testing machine.
Object of study
- Studying the CTM structure and manufacturing plastic injection molds contributes to creating products that can be applied in dynamic and static balancing of machinery and equipment to restore child joint function people and moving equipment
- Applying the materials and knowledge learned to the thesis helps students gain experience and feel more confident when graduating.
Methods of study
To carry out this thesis, we used several methods:
- Refer to articles, magazines, books about CTM From there, the idea of designing the shape of the CTM structural product was formed After that we use ANSYS software to check mechanical simulation with torque output
- Refer to the documentation about the template Knowledge over time has been accumulated Reference materials are collected through books, textbooks and the Internet
- Use NX 12.0 software to design the product, then proceed to the next steps such as mold separation, analysis
- Use Moldex3D software to analyze the injection molding process
- Conduct pressing and torque testing at the mechanical department workshop.
Structure of thesis
Chapter 3: Products and mold design
Chapter 4: Results of simulation and experiment
3 Chapter 5: Conclusions and future develop
LITERATURE REVIEWS
Overview of plastic injection molding technology
Injection molding technology is the process of injecting molten plastic to fill the cavity of the mold Once the plastic is cooled and solidified in the mold, the mold is opened and the product is ejector out of the mold thanks to the ejector system, during this process there is no chemical reaction (Fig.1)
Fig 1: Plastic injection machine HAITIAN MA1200III
The injection molding machine is composed of the following parts (Fig.2):
Clamping system Injection Hydraulic system system
Fig 2: Plastic injection machine structure [5]
- Hydraulic system: frame, hydraulic system, electrical, cooling system
- Injection system: hopper, barrel, heater band, screw, non-return assembly, nozzle
- Clamping system: machine ejector, clamp cylindero, moverable platen, stationary platen, tie bars
- Mold system: two fixed plate and one movable plate, 4 tie bars, locking mold cylinder, hydraulic cylinder, adjustment of mold, front and back safety doors
- Control system: computer screen, control panel
Raw materials are fed into the injection molding machine periodically After being plasticized, the raw material is injected into the mold (which has been clamped tightly) The shape of the mold will create the shape of the product After being shaped and cooled in the mold, the mold opening process is performed to get the product
The characteristic of injection molding technology is that the production process takes place in a cycle
Cycle time depends on the weight of the product, the temperature of the mold cooling water and the efficiency of the mold cooling system
The quality and productivity of the product depends on the quality of the injection molding machine and the quality of the mold.
Overview of mold
A mold is a tool (equipment) used to shape a product using the shaping method The mold is designed and manufactured to be used for a certain number of cycles, which can be one time or many times time The structure and size of the mold designed and manufactured depends on the shape, size, quality and quantity of the product to be created In addition, there are many other issues that need to be considered such as the technological parameters of the product (tilt angle, mold temperature, pressure, machining, etc.), and the properties of the processed materials (shrinkage, elasticity, hardness,), economic indicators of the mold set A plastic product production mold is a cluster of many parts assembled, divided into two main mold parts:
- Cavity part (main mold part, fixed mold part): mounted on the fixed plate of the plastic injection machine
- Core part (male mold part, movable mold part): mounted on the movable plate of the plastic injection machine
In addition, the space between the cavity and core (product forming part) is filled with molten plastic Then, the plastic is cooled, solidified and then removed from the mold using a product removal system or by hand The resulting product has the shape of the mold cavity
In a set of molds, the concave part will determine the external shape of the product called the mold cavity, while the protruding part will determine the internal shape of the mold The product is called a core (also known as positive mold, male mold, punch, core) A set of molds can have one or more mold cavities and cores The contact area between the mold cavity and the core is called the mold parting surface (Fig.3) [5]
Fig 3: Cavity plate and core plate in the closed state [5]
According to the number of mold cavity levels:
According to the type of the runner:
According to some color to make product:
In addition to the core and cavity, there are many other parts in the mold These parts assemble together to form the basic systems of the mold set, including (Fig.4) [5]:
- Guidance and positioning system: includes all guide pins, guide rings, locating rings, locators, return pins, responsible for keeping the correct working position of the two mold parts when joining together to create precise mold cavities
- Plastic delivery system into the mold cavity: includes injection stem, plastic channel and nozzle, which is responsible for supplying plastic from the injection molding head into the mold cavity
- Side core system: includes side core, core cheek, guide bar, oblique pin cam, hydraulic cylinder, responsible for removing parts that cannot be removed (undercut) immediately in the opening direction of the mold
- Vent system: includes vents, responsible for removing air remaining in the mold cavity, allowing the plastic to easily fill the mold cavity and preventing the product from bubbling or burning
- Cooling system: includes water lines, grooves, heat pipes, connectors, responsible for stabilizing mold temperature and cooling the product quickly
1 Locating ring 6 Guide bushing 11 Ejector retainer plate
2 Sprue bushing 7 Guide pin 12 Bottom plate
3 Sprue 8 Core plate 13 Ejector plate
5 Top plate 10 Spacer block 15 Return pin
The runner is the connection between the sprue bushing and the sprue Perform the task of putting plastic into the mold of cavity [5]
Therefore, when designing, it is necessary to comply with some technical principles to ensure the quality of most products Here are some guidelines to follow:
- Minimize changes in runner cross-section
- The plastic in the runner must drain the mold easily
- The entire runner length should be as short as possible, so that it can be quickly filled without pressure and heat loss during the filling process
- The size of the runner depends on the different type of material One side of the runner must be small enough to reduce scrap, shorten the cooling time, reduce clamping force
On the other hand, it must be large enough to transfer a significant amount of material, fill the mold bed quickly, resist and suffer little pressure loss
For this product, the channel should be designed so that the flow of plastic is smooth and always stable so that it is easy to fill the plastic into the mold
There are many options for choosing channel cross section such as: circular section, square section, trapezoidal section, to meet the flow requirements as well as ease of processing, we choose a runner with a circular cross-section
Sprue is the part located between runner and the cavity of mold
When designing the sprue, the following points should be paid to:
- The sprue needs to be positioned so that the flow of plastic into the place with the largest to smallest wall thickness so that the material can fill the product
- The optimal sprue position will create a smooth flow of plastic
- Place the sprue in a non-critical position of the product because the place where the sprue is located tends to exist residual stress during machining
- The sprue should be positioned so that all air can be expelled from the vent without creating air trap in the product
- Place the sprue so that no weld line is left, especially when using multiple sprues
- For circular sides, the cylinder needs to place the sprue at the plate to maintain concentricity
- The sprue is usually held to the smallest size and expanded as needed However, consideration should be given to limit the time to perform additional cutting and avoid creating marks on the product
The inside of the mold always contains air that needs to be ejector out when the plastic fills the mold This air must be drained quickly throughout the filling process As such, the vent system is to provide multiple pathways for air trapped in the mold to be released quickly and easily The vent system needs to be designed so that air easily escapes but does not allow molten plastic to pass through
When there is no vent system or the vent system is not well designed, it will cause some serious defects on the product such as welding seams, burn marks, unfilled parts,
The most used vent system is the air outlet grooves on the face of the mold and the grinding surface around the product tire In addition, the gas in the mold can also escape through the coolant line, small gaps of the sliding system, the graft
In fact, there are many exhaust options that can be used, depending on the structure of the mold bed, the location of the injection port, machinability, injection pressure,
Some of the options that are widely used today include:
- Vent through the vent groove on the mold face
- Vent through the propulsion system on the mold
- Vent through the vacuum system
- Vent through the cooling system, insert, slide,
2.2.3.3.1 How to get products out of the mold?
Overview of plastic
2.3.1 Mechanical characteristics of plastic resins
Mechanical properties related to the displacement or breakdown of plastics due to some mechanical changes are the same as applied to some loads [6]
Mechanical properties depend on temperature, Force (load), and load time applied It can be affected by ultraviolet rays when used externally
2.3.2 Thermal characteristics of plastic resins
Thermal properties include heat resistance or flammability
Thermoplastics have a greater coefficient of thermal expansion and flammability and less thermal conductivity or heat than other materials such as metals
2.3.3 Chemical properties of plastic resins
Resistance to chemical corrosion, resistance to chipping rifts, resistance to changing environments as chemical properties
When a plastic encounters chemicals, there are several types to change After a plastic encounters the chemical under no cracking for about a week, changes in appearance, weight and size of the plastic are checked Varieties, changes are called chemical properties
2.3.4 Electromagnetic properties of plastic resins
Electromagnetic properties are called magnetic properties, Electromagnetic properties include electrical insulation, electrical conductivity and electrical insulation properties
Due to their good electrical insulation, plastics are heavily used in the plastics industry However, plastic also has its shortcomings; they are easily electrified
Specific gravity, refractive index and moisture absorption are called physical properties The density of plastic is small, and each type of plastic varies depending on the bonding of the polymer, or the heat and mechanical treatment of the plastic
2.3.6 Some common plastic in injection molding
- Extremely impact resistant and has chemical strength, low temperature resistance as well as good electrical insulation
- At high melting point, good heat resistance
- Due to its self-lubricating properties, it is often used as the gear of machine parts
- Often used moving parts in machines (gears, ball bearings, cams) or bolts
- Specific gravity is the lightest of all common plastics
- Applicable to various spray gates such as point spray gate, direct "direct gate", special spray gate, etc
- There is no need to dry the umbrella before molding because it absorbs very little moisture
- "Molding shrinkage" varies depending on the mold temperature
- Often used for very large details or extremely thin details
- Because it has very good fatigue strength, it is often used as cyclic load-bearing hinge couplings
- Food packaging, household electrical appliances, auto parts, artificial lawns, suitcase covers, medical equipment, plastic bags,
- There are two types of polyethylene: low-density polyethylene and high-density polyethylene
- Low-density polyethylene is softer than high-density polyethylene It is great for casting
- High-density polyethylene has good hardness and impact resistance
- There is no need for drying before molding because it is not hygroscopic
- Low-density polyethylene is used for products that require softness and plasticity It is often used for plastics with complex shapes or packaging materials
- Low-density polyethylene is often used to improve the dilution of molding materials
- High-density polyethylene is often used for cylindrical containers, or for large plastic products such as phi containers
- Beverage containers, bottles and jars, wrappers, plastic bags, food wrap,
- Elastic and difficult to break
- As "Amorphous plastics", it is capable of withstanding bad climatic temperatures
- Easy to achieve dimensional accuracy, material stability
- It is a material that is easy to perform further machining (electroplating mechanical machining, flow welding, etc.)
- Often used as indoor or interior electrical appliances
- Keys on the computer, cases in electric poles, Lego toys,
- High flow temperature, viscosity fusing is also high
- The molding shrinkage rate is quite small (0.5 - 0.8%), and is not affected by the position of the gates
- Used as parts with durability requirements or resistant to dynamic loads and large loads
- Make CDs, DVDs, lenses, headsets, helmets,
Operational principle of the CTM
Most ideal compliant mechanisms (CMs) will obey the Hooke’s law when they are operated in the elastic regime However, some special CMs such as the CTM will exhibit an irregular torque curve that differs from the purely elastic mechanism as shown in Fig 8 It includes two regions: the pre-stress stage and the working range During the initial loading process, any elevation of the input rotation angle would lead to the increment of the reaction output torque If a CTM is properly designed, after this stage, the torque will remain stable in a certain range despite of the increasing of the rotation angle It is the working range of the CTM [2]
Fig 8: Torque – rotation angle of a compliant [2]
Artificial neuron networks in MATLAB
ANN-Artificial neuron network is an information processing model that mimics the way biological neuron systems process information It is made up of many elements (called
19 neurons) connected to each other through links (called link weights) that work to solve a particular problem
ANN, particularly deep artificial neural networks, have become known for their proficiency at complex identification applications such as face recognition, text translation, and voice recognition These approaches are a key technology driving innovation in advanced driver assistance systems and tasks, including lane classification and traffic sign recognition
There are three common learning methods: supervised learning, unsupervised learning, and reinforcement learning Supervised learning is the most used method, typically the backpropagation technique
Fig 9: Typical neuron network architecture [7]
Deep learning refers to neural networks with many layers, whereas neural networks with only two or three layers of connected neurons are also known as shallow neural networks Deep learning has become popular because it eliminates the need to extract features from images, which previously challenged the application of machine learning to image and signal processing Although feature extraction can be omitted in image processing applications, some form of feature extraction is still commonly applied to signal processing tasks to improve model accuracy (Fig.9) [7]
There are three common types of neural networks used for engineering applications:
- Feedforward neural network: Consists of an input layer, one or a few hidden layers, and an output layer (a typical shallow neural network)
- Convolutional neural network (CNN): Deep neural network architecture widely applied to image processing and characterized by convolutional layers that shift windows across the input with nodes that share weights, abstracting the (typically image) input to feature maps You can use pretrained CNN networks, such as SqueezeNet or GoogleNet
- Recurrent neural network (RNN): Neural network architecture with feedback loops that model sequential dependencies in the input, as in time-series, sensor, and text data; the most popular type of RNN is a long short-term memory network (LSTM)
Artificial neuron network is configured for a specific application (pattern recognition, data classification, ) through a process learned from a set of training patterns In essence, it is the process of calibrating the bond weights between neurons so that the error function value is minimal
2.5.3.1 Assess the elements of the learning process
Weighting initialization: Since the nature of the error back-propagation algorithm is a method of reducing gradient deviation, initializing the initial values of random small value weights will cause the network to converge to different minimum values
Learning step α: Choosing the initial arithmetic constant is very important For each problem, we have a different arithmetic option When a back-propagation training process converges, it cannot be said that it has converged to the optimal solution We need to experiment with some initial conditions to ensure that we get the optimal solution
The MATLAB Toolbox are collections of m-files that allow extending the capabilities of MATLAB to several modern control techniques, data optimization processing and ANN, ANN Toolbox provides 12 high-performance training functions To use this toolbox, we must define a structure including creating input data and target data matrices, calling the ANN Toolbox in an m-file to set and select network parameters.
Torsional strength tester
Torsional strength tester is composed of motor, speed reducer, encoder, 3-jaw chuck, torque sensor, PLC control circuit, computer (Fig.10)
Fig 10: Structure of torsional strength tester
Initially turn on the power switch supplied to the motor to help the motor rotate and create torque This torque is transmitted to the axis attached to the steering wheel through a gear belt transmission with a gear ratio of 1:2 This shaft is made of steel with a high load capacity to withstand torque from the force of rotating the steering wheel [8]
The steering wheel shaft is connected in series to the input shaft of the gearbox through a rigid coupling This coupling helps to drive from the steering wheel shaft to the gearbox continuously and accurately The gear reducer is used to reduce the rotational speed and increase the torque on the output shaft
The output shaft of the gearbox connects to the chuck by means of a rigid coupling The rigid coupling ensures the torsional drive correctly
On the shaft attached to the chuck, a pair of 1:1 gear wheel is mounted to transmit rotation from the shaft attached to the chuck The tooth belt transmission helps synchronize movement to provide a torsion angle to the encoder The encoder is used to collect the torsion angle of the test specimen, allowing the twist data to be observed and recorded through a small screen
Movable chuck is connected to the twisted test specimen through an intermediate part to link the chuck and sample together Unlike movable chuck, the other chuck in this model is fixed and cannot rotate
Torque sensor is used to measure the torque of the test specimen It is attached to the chuck fixed through an intermediate detail designed in accordance with the input and output of the sensor Torque sensor captures torque from the test sample allowing observation and recording of torque data
As the motor rotates, the transmission of torque through parts in the model helps to create torque in the prototype, while measuring the torque angle through the encoder and torque through the torque sensor This process allows torsional strength testing and data collection for analysis and evaluation of the mechanical properties of the test specimen
PRODUCTS AND MOLD DESIGN
Products design
Fig 11: Product design constant-torque
Based on the Fig 11 groups changed the geometry of the arm while retaining the outer diameter dimensions: ∅100 mm, height: 3mm, width of the arm: 1mm
Design products using Inventor 2022 software
Click New to start creating the working program
Use module Part the process of design the product
Fig 12: Select New and module part to design the product
Step 1: Use command Extrude diameter 100mm, high 2.5mm (Fig.13)
Fig 13: Profile and size when using extrude command
Step 2: Use command Extrude design some small details (Fig.14, 15, 16,17)
Fig 14: Shape and dimensions part 1
Fig 15: Shape and dimensions part 2
Fig 16: Shape and dimensions part 3
Fig 17: Shape and dimensions part 4
To facilitate the production of product injection molds, the team decided to divide the product into many parts without changing the torque (Fig.18) Because:
- Convenient for designing and arranging layouts
- Reduce mold size (reduce processing cost)
- Easy to get the product from the mold
After the product is injection molded, the team will design a jig for assembly and torque testing
3.1.2 Analysis of a Compliant Constant-Torque Mechanism
In order to generate the relationship between the reaction torque and applied displacement, the static analysis is carried out with ANSYS software to simulate the torque change along with the variation of input stroke In the simulation, the material is assigned as
PP, which has the yield strength of 34.6 MPa, Young’s modulus of 1.461 GPa, and Poisson’s ratio of 0.4087 and initial conditions: number of steps is 75 (part 1), 40 (part 2, 3, 4), initial time is 0.5s, minimum time step is 0.5s, maximum time step is 1s, rotation Z The result Analysis of a Compliant Constant-Torque Mechanism using ANSYS Workbench software as shown in Fig 24, Fig 26, Fig 28, Fig 30
Material: PP Boundary mesh: 2mm
Fig 19: Model mesh & boundary conditions part 1
Material: Alumium Boundary mesh: 2mm
Fig 20: Model mesh & boundary conditions part 2
Material: Alumium Boundary mesh: 2mm
Fig 21: Model mesh & boundary conditions part 3
Material: Alumium Boundary mesh: 2mm
Fig 22: Model mesh & boundary conditions part 4
Fig 23: 3D model of the CTM part 1
The maximum stress of the mechanism stands at 29.342 MPa, shown in Fig 23, is much smaller than the yield strength of the PP material (σ y = 34.6 MPa) The part works normally with safety factor of 1.16
Fig 24: Torque-rotation angle, θ (degree) part 1
In Fig 24, we observe that the constant torque region has torque values of approximately 0.164 N.m to 0.167 N.m, with a rotation angle ranging from about 23° to 75°
Fig 25: 3D model of the CTM part 2
The maximum stress of the mechanism stands at 27.547 MPa, shown in Fig 25, is much smaller than the yield strength of the PP material (σ y = 34.6 MPa) The part works normally with safety factor of 1.25
Fig 26: Torque-rotation angle, θ (degree) part 2
In Fig 26, we observe that the constant torque region has torque values of approximately 0.137 N.m to 0.14 N.m, with a rotation angle ranging from about 16° to 43°
Fig 27: 3D model of the CTM part 3
The maximum stress of the mechanism stands at 17.542 MPa, shown in Fig 27, is much smaller than the yield strength of the PP material (σ y = 34.6 MPa) The part works normally with safety factor of 1.97
Fig 28: Torque-rotation angle, θ (degree) part 3
In Fig 28, we observe that the constant torque region has torque values of approximately 0.127 N.m to 0.13 N.m, with a rotation angle ranging from about 20° to 50°
Fig 29: 3D model of the CTM part 4
The maximum stress of the mechanism stands at 15.507 MPa, shown in Fig 29, is much smaller than the yield strength of the PP material (σ y = 34.6 MPa) The part works normally with safety factor of 2.23
Fig 30: Torque-rotation angle, θ (degree) part 4
In Fig 30, we observe that the constant torque region has torque values of approximately 0.142 N.m to 0.144 N.m, with a rotation angle ranging from about 17° to 45°
Mold design
3.2.1 Material, mass and volume of product
PP plastic stands for Polypropylene and is a type of thermoplastic produced from derivatives of propylene This type of plastic has the properties of plasticity, elasticity and high temperature resistance, along with high durability, chemical resistance and mechanical strength, PP plastic is widely used in applications such as the production of children's toys, manufacturing household appliances, shopping bags, bottles, food containers and many other applications This is also an easily recycled and environmentally friendly plastic
PP plastic (Polypropylene) is a synthetic polymer with a simple chemical structure consisting of propylene molecules The main component of plastic is propylene molecules (C3H6), produced from propylene, a hydrocarbon found in oil and gas
Propylene molecules are linked together through simple chemical bonds to form a polymer network PP plastic has a simple crystalline polymer structure, with tightly arranged molecules, allowing it to have good mechanical and physical properties
In addition, during the production and processing process, PP plastic is often mixed with other additives such as colorants, anti-corrosion protection agents, softeners, durability enhancers and texturizers, depending on the intended use of the final product
- Chemical resistance: PP plastic is able to withstand many chemicals, including acids, alkalis, organic solvents, ethanol, ethylene glycol, gasoline and engine oil
- Water absorption: This is a plastic material with low water absorption, only about 0.01% to 0.03%
- Corrosion resistance: It is itself a plastic material, so it also inherits the properties of plastic, including high corrosion resistance to water and organic solvents
3.2.1.4 Product mass and volume inspection
Use Inventor 2022 software to calculate the mass and volume of the product Open product → Properties tab → Physical → Select material: polypropylene as shown in Fig 31
Fig 31: Check mass and volume part 1
Similarly for the remaining samples, we get the mass and volume table of each sample as follows:
Table 3: Product mass and volume
3.2.2.1 The significance of the shrinkage coeficient
The shrinkage coefficient of PP (Polypropylene) plastic is an important characteristic to evaluate the shrinkage of the material when it is cooled from the molten state to the solid state This characteristic is often used in the design and production of plastic products,
35 especially products with precise dimensions and requiring precision in the manufacturing process
- Predict deformation: The shrinkage coefficient helps predict the degree of shrinkage of the PP material after it is cooled, thereby helping engineers and designers predict the deformation of the final product after it is produced
- Dimension control: Understanding the shrinkage coefficient helps ensure that the dimensions of the product after production are accurate and meet technical requirements
- Optimize the production process: By adjusting production parameters such as temperature and pressure, one can optimize the production process to minimize unwanted shrinkage of the material
- Application in product design: Designers can use information about shrinkage coefficient to design products to suit the final size requirements of the product
3.2.2.2 How to apply shrinkage coefficient to products
Look up Table 4, plastic shrinkage of different types of plastic:
Table 4: Plastic shrinkage of different types of plastic
Based on table 3, we see that PP plastic shrinkage is in the range of 1.0-3.0% → Choose 1% The shrinkage coefficient of PP plastic is 0.01 according to the formula 1 + S, the mold size will be larger than the reference size 1.01 times (Fig.32):
3.2.3.1 Calculating the number of mold cavities
Typically, the required number of mold lumen can be calculated on the mold in the following ways:
- Calculated by the number of product batches
- Calculated according to the injection output feature of the machine
- Calculated according to the plasticization capacity of the machine
- Calculated according to the mold clamping force of the machine
- Calculated according to the grip size of the press
3.2.3.1.1 Number of mold cavities calculated by the number of product batches n ≥L × K × t c t m (1)
1 − k (2) Where: n: Minimum number of mold cavity in the mold
L: Number of products in 1 batch
K: Coefficient due to waste products (%) k: Waste rate of each company (%) t c : Injection molding cycle time of 1 product (s) t m : Time required to complete 1 batch of products (day)
Let us say we need to make 1 shipment of 70000 products within 1 month The factory works 18/24 hours, the waste rate is 2%, the maximum allowed cycle time is 20 seconds We have:
3.2.3.1.2 The number of mold cavities calculated according to the injection capacity of the machine
The injection capacity of the machine is also a factor affecting the mold lumen n ≤ 0.8 × S
4.02 = 48.95 (3) Where: n: Maximum number of mold lumen in mold
S = 246: Spray capacity of the machine (g/1 spray)
3.2.3.1.3 The number of mold cavities calculated according to the plasticization capacity of the machine n ≤ P
3 × 4.02= 148.26 (4) Where: n: Maximum number of mold lumen in mold
P = 29.8 g/s = 1788 g/min: Plasticization capacity of the machine (g/min)
X = 3: Spray stage (estimated) per minute (1/min)
3.2.3.1.4 The number of mold cavities calculated according to the clamping force of the machine
5836.948 × 50 = 4.109 (5) Where: n: Maximum number of mold lumen in mold
F 𝑝 = 1200000: The maximum mold clamping force of the machine (N)
S = 5836.948: Average surface area of products in the direction of molding (mm 2 )
P = 50: Pressure in the mold (Mpa)
From the above calculations, we choose 1 or 2, 3, 4 cavities
The goal is to design a mold set for 4 different parts in one injection molding session
So, we choose 4 mold cavities because:
- Mold size is smaller than the machine's limited size
- Easy to arrange symmetrically inside the mold
- Ensure design goals are met
- Select the save folder, name the details, choose the plastic type and shrinkage coefficient for the product (Fig.33)
- Create Family Mold to manage parts (Fig.34)
• Definition of workpiece coordinates, workpiece creation
- Workpiece coordinate system: Click on the icon (Mold CSYS) in tab Mold Wizard (Fig.35)
- Create (Workpiece): Click on the icon (Workpiece) to create Select workpiece sized 90x70x65 for 4 parts (Fig.36, 37, 38, 39)
Click to select Cavity Layout, then arrange the cavity of the mold so that the part with the largest volume is closest to the runner (Fig.40)
- Select (Check Region) creating mold draft for the product (Fig.41)
Fig 41: Selecting mold draft direction
- Select (Define Regions) to select the Cavity and Core faces (Fig.42)
- Select (Edge Patch) repairing holes or voids on the product (Fig.43)
- Select (Design Parting Surface) to extend the mold surface around the workpiece dimension (Fig.44)
- Select (Define cavity and core) separating the core plate and cavity plate (Fig.45)
Fig 45: Defining cavity and core
- The result obtained is core plate and cavity plate (Fig.46)
Fig 46: Core plate and cavity plate part 1
- Do the same with part 2, part 3, part 4 Then we use the command (Merge Cavities) to Merge 4 core plates and 4 cavity mold plates, respectively (Fig.47, 48)
3.2.4.1 Calculate the dimension of the runner according to theory 3.2.4.1.1 Sprue
The products have a maximum thickness of 2.5 mm
The sprue has the following dimensions (Fig.49):
Fig 49: Selecting the sprue size [5]
Select a runner with a circular cross – section (Fig.50)
→ Due to the complex shape of the part, the team flexibly adjusted the gate to suit the part while still filling ability
3.2.4.2 Design the runner on the software according to the dimension
Select the icon Runner in tab Main→ Select curve→ select type→ diameter as shown in Fig 52
Fig 52: Choosing style and diameter of runner
- Select the type of plastic gating (Fan gate) and the layout results as shown in Fig 53, Fig 54
Fig 53: Injection gate specifications and layout
Fig 54: 3D design of cavity and runner
The results of the filling analysis are presented in Fig 55 The results showed that the filling time at the furthest locations on the product was 0.38s
The filling process is represented by colors that vary continuously from green to red corresponding to an increasing filling time It is noticed that, as soon as the injection molding process begins, the molten plastic is filled into the sprue, then the liquid resin gradually preferentially fills the part around the product sprue position and then proceeds to fill the part crush In other words, plastic is preferably filled in the area that is located closer to the adjective from sprue After this zone is filled, the resin will be injected into the rest The red areas are filled in last
Comment: In general, the whole mold is filled in a reasonable and uniform way from outside to inside symmetrically and filled with cavity mold
Air trap during injection is present in the product These air trap form holes inside the product, affecting the mechanical as well as aesthetic properties of the product
Air traps occur when the flow of the same plastic surrounds the air trap Air trap defects make it impossible for plastic to fill the mold completely and decorate the surface of the product In addition, air trap creates compressive stress on other areas of the product and heats causing burn marks on the surface of the product
Air trap defects on the product appear usually due to the following reasons:
- When products have concentrated currents, they often put gas in one spot causing air trap in that spot
- Often air trap is caused by unbalanced filling flows
- During the mold filling process, air is retained in the product in areas near the product surface
According to Fig 56, the analysis will help determine the centralized location of air trap on the product, thereby offering solutions to minimize the occurrence of air trap defects on the product to avoid affecting the mechanical as well as aesthetic properties of the product
Filling pressure, showing the change in pressure filling, the filling process will usually use the maximum pressure in the whole cycle, the injection pressure affects the entire analysis results, the pressure change will lead to a significant change in the results achieved For this products, the maximum pressure required according to the analysis (Fig.57) is 70.827 MPa, which is relatively suitable for injection molding machines
The pressure of the product at the end of the fixing point is 70.827 Mpa This pressure difference is intended to help the product shaping process be more efficient, the product is less prone to shrinkage and warping during the cooling process
Injection molding temperature is a very important parameter, the graph below shows the temperature change in the Filling process, each material can withstand a certain temperature, when exceeding that temperature will cause the material to be mechanically changed, it is necessary to change the parameters of mold temperature, machining temperature to limit the effects of material overheating
On the Fig 58, the highest temperature is 217.467 o C and the lowest is 63.43 o C, it can be observed that because only the exterior of the model is shown, the material that meets the surface of the mold will lose heat quickly and become a solidified outer shell The change in the temperature inside the product and the part in contact with the surface of the mold bed, due to heat loss, the temperature of the contact part will decrease rapidly and solidify, this is the cause of sliding stress and causes friction heat during injection molding
Fig 59: Warpage_Total displacement simulation
The warpage of the largest product is 0.303mm, and the lowest is 0.004mm (Fig.59) Warpage occurs more for parts located far from the sprue and the closer to the sprue the warpage almost disappears Warpage greatly affects product quality, so when designing we should pay attention to it
RESULTS OF SIMULATION AND EXPERIMENT
Sample’s manufacturing
Lifting, securing, and clamping tools for fixing molds on the injection molding machine Hexagon lock, wrench, pliers, caliper, hoist, overhead crane
Mounting the mold on the injection molding machine
Check machine operation Prepare the plastic
Fine-tune plastic injection parameters.
Sample manufacturing Clean the nozzle
The mold is being injected using the Haitian injection molding machine:
Fig 91: Mold mounted on Haitian injection molding machine
According to this thesis, our group uses a type of PP plastic in different cases of changing injection molding parameters For PP plastic, we changed the injection parameters of injection pressure, holding pressure, melting temperature, mold temperature, holding time (Fig.92, 93, 94, 95, 96) and kept the injection speed 85% of the machine’s maximum injection speed, injection time was 3s, cooling time was 15s
With PP plastic, the group weighs each case corresponding to the plastic weight of 6.5kg-7kg, depending on extruding the remaining old plastic and testing the model until it flows stable plastic
Using Tab Taguchi of Minitab software to optimize the injection molding case, we get the following set of parameters:
Table 7: Taguchi parameters for some types of situations
Fig 95: Change plastic melting temperature
Case 1 Case 2 Case 3 Case 4 Case 5
Case 6 Case 7 Case 8 Case 9 Case 10
Case 11 Case 12 Case 13 Case 14 Case 15
Case 16 Case 17 Case 18 Case 19 Case 20
Case 21 Case 22 Case 23 Case 24 Case 25
Testing torque
To secure the model and ensure the fixture stays in place during the torque test, we have designed specific details to facilitate the fixture placement process
Fig 98: Structure of fixture to testing torque part 1
Fig 99: Structure of fixture to testing torque part 2
Fig 100: Structure of fixture to testing torque part 3
Fig 101: Structure of fixture to testing torque part 4
Fig 102: 3D design of the fixture for the products
Fig 103: Secure the product using the fixture
Fig 104: Assembling the product into fixture
Fig 105: Completely assembling the product into fixture
Step 1: Check the sample (the test sample is PP plastic) to preserve the specimen:
- Before testing, specimens must be stored in a suitable environment, namely room temperature (25-34 o C), proper storage is very important because this ensures that the specimen does not deform when exposed to excessive heat, because this is plastic, so when exposed to high temperature, it will be deformed, leading to inaccurate test results
- Sample case numbering: Because we have a total of 25 cases tested, this numbering is quite important when it helps us distinguish the number of cases and the order of testing, along with that it will be easy to record the information obtained by each sample, distinguish and thereby synthesize data
Step 2: Connect the Ethernet cable to the computer and launch the Tia Portal software
- After opening the Tia Portal software, we open the Torque Broken file (programmed to control the torsion test machine through PLC control circuit) to create a new data recording file when conducting the test torque At the same time press Set 0 torque and Set 0 angle (due to the possible rotation of the chuck during the mounting process) Step 3: Inspect the electronics and install the jig into the torsion test machine
- Check Rotary Encoder: Check the operation status by pressing Jog left (right) and see if the rotation angle on the screen changes accordingly
- Install the Torque sensor on the side of the fixed chuck, align the recess to keep it straight for easy installation of the test torque jig
- Install the part into the jig suit and then install it on the chuck, pay attention to tighten the chuck so that the jig shaft is concentric with the test machine shaft twisted
Step 4: Set the recording angle on the screen and press Start
- The motor starts recording and the torque value on the screen gradually increases, along with that the angle value on the screen increases accordingly
- After the recording details are completed, we download the resulting file with the torque data column and rotation angle
- After having enough data of all 25 cases, we use Excel software and Origin software to process data, compare and draw graphs, from which we can know with which set of
82 pressed parameters, the details have the longest constant-torque segment to be able to make the conclusion of the thesis
Based on the results obtained from the torque testing machine, we have determined the optimal compression parameters for each sample:
Fig 107: Experimental torque graph part 1
Fig 108: Experimental torque graph part 2
Fig 109: Experimental torque graph part 3
Utilizing ANN in MATLAB
With the aim of predicting torque, the team utilized neuron networks to aid in predicting results with arbitrary parameters:
To perform network training and create a neuron network block for use in Simulink, follow these steps New available → With input as the compression parameters: injection pressure, molding pressure, plastic temperature, mold temperature, molding time, rotation angle, and the target being the torque
Fig 112: Prediction result graph ANN part 1
Fig 114: Prediction result graph ANN part 2
Fig 116: Prediction result graph ANN part 3
Fig 118: Prediction result graph ANN part 4
The results after training in Fig 111, 113, 115, 117 indicate that after 1000 iterations, the training error is only less than 0.0008 Typically, a Neuron Network is considered reliable when the training error is less than 0.01 After training, we can use the model to recognize
89 new datasets When applying the test file to compute output parameters, the error is less than 3%.
Comparing results
From the results of CAE simulation, experiment, and ANN prediction we can see:
Fig 119: Torque graph CAE, experiment, ANN part 1
Compare the CAE simulation results with the experiment and the ANN prediction results: looking at Fig 119, we see that the results of both the experiment and CAE simulation lines appear at a constant-torque section, but the CAE simulation results have a constant- torque section The constant-torque segment appears approximately from 23-75 degrees, while the experiment results and ANN prediction results show the torque segment longer, from 23-125 degrees The constant-torque segment value of CAE simulation prediction reaches about 0.165 N.m, experiment and ANN prediction around 0.124-0.127 N.m
Experiment run results with ANN prediction results: look at Fig 119 experiment run results and ANN prediction results are almost 98% similar because ANN prediction results are learned from experiments
Fig 120: Torque graph CAE, experiment, ANN part 2
Compare the CAE simulation results with the experimental and ANN prediction results: Upon examination of Fig 120, it is evident that both the experimental and CAE simulation lines exhibit a constant-torque section However, in the CAE simulation results, this constant-torque section appears to be shorter, spanning approximately from an angle of
16 to 43 degrees, with the torque value around 0.139 N.m In contrast, the experimental and ANN prediction results show a longer constant-torque segment, ranging from 15 to 45 degrees, with the torque value fluctuating around 0.172-0.174 N.m
Comparing the experimental run results with the ANN prediction results: As depicted in Fig 120, there is a remarkable similarity between the experimental run results and the ANN prediction results, with an approximate match of 97% This high level of similarity can be attributed to the fact that the ANN prediction results are learned from the experimental data
Fig 121: Torque graph CAE, experiment, ANN part 3
Comparing the CAE simulation results with the experimental and ANN prediction results: Observing Fig 121, it is evident that all three lines exhibit a constant-torque section However, in the CAE simulation results, this section spans approximately from 20 to 50 degrees, while in the experimental results and ANN prediction, the constant-torque segment appears longer, ranging from around 20 to 57 degrees The torque value for the constant- torque segment in the CAE simulation is approximately 0.128 N.m, whereas in the experimental and ANN prediction results, it is nearly equal, reaching about 0.136-0.138 N.m
Comparing the experimental run results with the ANN prediction results: In Fig 121, it's notable that the experimental run results closely match the ANN prediction results, exhibiting a similarity of approximately 98% This high degree of resemblance can be attributed to the fact that the ANN prediction results are derived from the experimental data
Fig 122: Torque graph CAE, experiment, ANN part 4
Comparing the CAE simulation results with the experimental and ANN prediction results: Upon inspection of Fig 122, it is evident that both the experimental and CAE simulation lines display a constant-torque section However, in the CAE simulation results, this section spans approximately from 17 to 45 degrees, whereas in the experimental results and ANN prediction, the constant-torque segment appears longer, ranging from around 20 to
45 degrees The torque value for the constant-torque segment in the CAE simulation is approximately 0.142 N.m, whereas in the experimental and ANN prediction results, it is nearly equal, reaching about 0.164-0.167 N.m
Comparing the experimental run results with the ANN prediction results: Upon examining Fig.122, it is evident that the experimental run results closely resemble the ANN prediction results, showing a similarity of approximately 97% This high level of similarity can be attributed to the fact that the ANN prediction results are trained based on the experimental data
Fig 123: Torque simulation CAE graph 4 parts
From Fig 123, it is apparent that part 1 exhibits the highest torque value and rotation angle, ranging from approximately 23° to 75°, with a torque value reaching 0.165 N.m In contrast, part 2 displays the smallest torque value and rotation angle, spanning from about 16° to 43°, with a torque value of approximately 0.139 N.m
Fig 124: Torque Experiment graph 4 parts
From Fig 124, it is evident that part 1 exhibits the highest torque value and rotation angle, spanning approximately from 23° to 125°, with a torque value of about 0.124 N.m Conversely, part 2 displays a lower torque value and rotation angle, ranging from approximately 15° to 46°, with the torque value reaching 0.174 N.m
Fig 125: Torque simulation ANN graph 4 parts
From Fig 125, it's evident that part 1 exhibits the highest rotation angle, ranging from approximately 25° to 125°, with the smallest torque value recorded at only 0.127 N.m On the other hand, part 2 displays the highest torque value, reaching about 0.172 N.m, with a smaller rotation angle ranging from approximately 17° to 45°
Table 8: Compare the torque and rotation angle of 4 parts
When both simulation and actual measurement exhibit a constant-torque section with nearly identical torque magnitudes, the disparity lies primarily in the rotation angle This difference can be attributed partly to variations in the mechanical properties of materials between CAE simulations and real-world conditions, as well as external factors such as temperature, humidity Additionally, different product designs can lead to variations in torque values
→ Conclusion: After conducting CAE simulations, experiments, and utilizing ANN for 4 different parts, it is evident that both the torque values and rotation angles vary among the parts This highlights the significant impact of geometric changes on the torsional strength of plastic injection-molded products When designing products, careful attention must be paid to their geometric shapes to achieve the desired torque values
CONCLUSIONS AND FUTURE DEVELOPMENT
Conclusions
The main objective of the project is to study the influence of geometric shape on the torsional strength of plastic injection molding products In fact, the group has solved some of the main and crucial issues of the report, including:
- Understanding the structure of 2-plates mold, 3-plates mold, molds with hot runner systems
- Utilizing Ansys software to simulate changes in torque
- Applying NX 12.0 and Inventor 2022 software for designing products and mold
- Implementing Moldex3D software for simulating plastic flow
- Knowing how to operate injection molding machines
- Measuring torque on torque testing machines
- Using ANN in MATLAB to predict torque results
- Comparing torque measurement results between CAE, experiment, and ANN
Through this graduation thesis, in addition to applying the knowledge learned in basic and specialized subjects, the group is also exposed to mold manufacturing technology
During the project process, the group tried to complete this thesis as best as possible with the knowledge they had learned and the guidance of Assoc Prof Dr Do Thanh Trung, Assoc Prof Dr Pham Son Minh and Dr Tran Minh The Uyen Due to the limited time to carry out the thesis and write the report, our thesis will inevitably have shortcomings We respectfully hope to receive your enthusiastic comments and guidance.
Study direction and future development
During the process of studying and implementing the project, in addition to the results achieved, the group did not test different types of plastic to compare the influence of different types of plastic on torque
Therefore, our group hopes that the following groups of students will continue to study the remaining issues to make the thesis more complete
[1] Chia-Wen Hou, Chao-Chieh Lan (2012) Functional joint mechanisms with constant- torque outputs vol 16, no 1
[2] Phan Thanh Vu, Pham Huy Tuan (2020) Design and Analysis of a Compliant Constant- Torque Mechanism for Rehabilitation Devices vol 10, chapter 44 10.1007/978-3-030- 45120-2_44
[3] Piyu Wang, Sijie Yang, and Qingsong Xu (2018) Design and Optimization of a New Compliant Rotary Positioning stage with Constant Output Torque.Vol 19, no.12, pp1843-
[4] Hari Nair Prakashah, Hong Zhou (2016) Synthesis of Constant Torque Compliant Mechanisms vol 8
[5] TS Phạm Sơn Minh, ThS Trần Minh Thế Uyên (2014) Giáo trình Thiết kế và chế tạo khuôn phun ép nhựa Việt Nam Nhà xuất bản Đại Học Quốc Gia TP.HCM
[6] Đặc tính của các loại nhựa Truy cập ngày 18/12/2023 từ: https://moldviet.net/chi-tiet- tin/dac-tinh-cua-cac-loai-nhua-197.html
[7] What is a neural network Truy cập ngày 16/12/2023 từ: https://au.mathworks.com/ discovery /neural-network.html
[8] Phạm Thế Sơn, Nguyễn Phương Nam, Lê Hồng Phúc (2023) Chế tạo máy thử độ bền cho chi tiết nhựa (Đồ án tốt nghiệp, trường Đại Học SPKT TP.HCM)
[9] Nguyễn Anh Minh, Võ Tấn Quí, Nguyễn Duy Quỳnh (2023) Thiết kế và chế tạo khuôn phun ép nhựa sản phẩm “nắp bịt ống thủy lực” (Đồ án tốt nghiệp, trường Đại Học SPKT TP.HCM)
[10] MISUMI Coporation (2015) MISUMI Standard Components of Plastic Mold
[11] Trần Quốc Hùng (2012) Giáo trình Dung sai – Kỹ thuật đo Việt Nam Nhà xuất bản Đại Học Quốc Gia TP.HCM