Reverse Surface Modeling, Mold Design and Validation

Reverse Surface Modeling,
Design & Validation
ME7722 ADVANCED CAD/CAM

Submitted For: Dr. Redha Benhadj
Submitted By: K0907687 & K1135912
Submission Date: 23/03/15

Mold

ABSTRACT
This report aimed to identify the uses of reverse engineering to achieve surface remodelling,
and thereby design and validate a mold based on a selected CAD geometry. The input
required was achieved through 3D laser scanning of a perfume bottle with a curved structure
that fulfilled the challenging aspect of the scope of this project. Further, , this report includes
reconstruction of cloud point scans, defining NURBS surfaces, the generation of IGES files and
the setup of a mold assembly.
One model was produced by each engineer and modelling reconstruction and model selection
was performed.
Modelling reconstruction used the cloud points obtained through laser scanning and
reconstructed using high advanced surface modelling functions in Geomagic Studio 2014 and
Siemens NX 9.0 software. This data was manipulated using meshing and surfacing techniques.

Both authors are referred to as engineers through the course of this report.

i

Table of Contents
Author’s Acknowledgments ............................................................................................ iii
1.0. Introduction .............................................................................................................. 1
Reverse Engineering......................................................................................................................... 1
Uses of Reverse Engineering ........................................................................................................ 1
FAROArm Machine ........................................................................................................................... 1
FAROArm Machine Specification ................................................................................................. 2
Method of Use ............................................................................................................................. 2
Understanding Geomagic Software ................................................................................................. 3
System Requirements .................................................................................................................. 3
Geomagics Capabilities ................................................................................................................ 3
Scan To CAD ................................................................................................................................. 3
Scan To Surface ............................................................................................................................ 3
Scan to Mesh................................................................................................................................ 4
Geomagics Capabilities Conclusion.............................................................................................. 4
3D Scan Data ............................................................................................................................... 4
3D Scan Data Conclusion ............................................................................................................. 5
Understanding Siemens NX 9.0 Software ........................................................................................ 6
2.0. Summary of Tasks ..................................................................................................... 7
2.1. Part A- Surface Reconstruction Using GEOMAGIC & NXCAD .................................................... 7
Stage 1 – Digitise 3D Part ............................................................................................................ 7
Stage 2 - Model Reconstruction ................................................................................................... 7
Stage 3 – Model Selection and Justification ................................................................................ 7
2.2. Part B- Mold Design, Optimization and Validation ................................................................... 7
Stage 1 – Simplifying the Chosen Model...................................................................................... 7
Stage 2 – Mold Setup ................................................................................................................... 7
Stage 3 – Mold Validation ........................................................................................................... 7
3.0. Part A – Surface Reconstruction with Geomagic and NX ............................................. 8
Component Selection Process ......................................................................................................... 8
Unused Components........................................................................................................................ 8
Stage 1 – Digitize 3D Part ............................................................................................................ 9
Engineer A – Initial Scans ............................................................................................................. 9
Engineer B- Initial Scans............................................................................................................. 10
Conclusion for stage 1................................................................................................................ 10
Stage 2 – Model Reconstruction, Engineer A ............................................................................. 11
Stage 2 – Model Reconstruction, Engineer B ............................................................................. 16
Conclusion for stage 2................................................................................................................ 23
Stage 3 – Model Selection and Justification .............................................................................. 24
Conclusion for stage 3................................................................................................................ 24
3.0. Part B – Mould Design, Validation & Optimisation.................................................... 25
Aim and objectives ......................................................................................................................... 25
Stage1 - Simplifying the Chosen Model ......................................................................................... 26
Stage 2 - Mold Setup ...................................................................................................................... 27
Checking regions and assigning Cavity and Core Faces ............................................................. 27
Project initialisation ................................................................................................................... 28
Creating the Workpiece ............................................................................................................. 29
Analysing the Parting Regions in the bottle section solid geometry ......................................... 31
Adding the Mold Base: ............................................................................................................... 33
Adding Standard Parts ............................................................................................................... 35
Stage 3 –Discussion and Mold validation .................................................................................. 41
References..................................................................................................................... 42
ii

Author’s Acknowledgments
The authors would like to thank the following individuals for their expertise in making the
journey in appreciating and understanding the benefits of CAD/CAM both pleasant and
educationally fulfilling.
Dr Benhadji Djillali
Diana (PhD student)
Cliff Searle

iii

1.0. Introduction
Reverse Engineering
Reverse Engineering is a process that measures an object and reconstructs its 3D model.
The measured data is represented by point clouds and is represented as an image. The
image is then modelled in CAD deriving a more useable format that can further be
manipulated and modified.

Figure 1. Reverse Engineering Process

1

Uses of Reverse Engineering
Optimisation: Capturing the current data of the object and further improving its
performance.
Loss of Documentation: Reverse engineering can be useful when CAD files or engineering
drawings become unavailable.
Legacy Equipment: As a form of research and development, the specification of the
current object can be stored and any further modifications can result in updating the
database with the new specification allowing for the manufacture of spares which will
reduce time and money during the manufacturing process.

FAROArm Machine
Kingston University’s FAROArm laser scanner machine will be used as it is readily available
and with the aid of technicians, is the ideal resource for this project.
The FARO Company develops portable Coordinate Measuring machines and 3D imaging
devices to understand and solve metrology problems.
FARO guarantees high-precision 3D measurement, imaging for production and quality
assurance processes.

1

FAROArm Machine Specification

Figure 2. FAROArm Machine

2

Method of Use
The FAROArm is pre-calibrated by the lab technician, therefore calibration is not required.
For method of use, please refer to Laser Scanning Tutorial (Koorosh Khanjani, 2015)
This tutorial provides the step-by-step guidelines on laser scanning and cloud point data
processing using Geomagic 2014.

2

Understanding Geomagic Software
System Requirements
It is important to consider the system requirements and the package edition, as each
edition varies with system requirements.
The following edition was used as this was readily available at Kingston University for the
purpose of this project: Windows Vista 64-bit Edition
Geomagics Capabilities
Scan To CAD
Rather than designing from the beginning, Geomagic allows for the features to be built
directly from the 3D scan data.
Align scans > Extract design > Merge into a single model > Build CAD model > Confirm
accuracy

Figure 3. Scan To CAD Process

3

Scan To Surface
Geomagic supports the surface fitting for when a copy is required and editing at a later
date is not required.
Align scans > Merge into a single model

Figure 4. Scan To Surface Process

3

4

Scan to Mesh
This is a rather complex technique, with a number of stages in order to reach the desired
model.
Hence, this uses the mesh technique.
Align scans > Merge into model > Smooth > Fill holes > Interactive Surfacing > Save as IGES
or STEP format

Figure 5. Scan To Mesh Process

5

Geomagics Capabilities Conclusion
This project will focus on the ‘Scan To Mesh’ technique to fulfil the scope of this project in
effectively exploring Geomagic and NX software.
This will be achieving in identifying the appropriate meshing and surfacing technique to
remodel the component. Hence, the following sections will cover these techniques in
depth and reasons for their use will be outlined.
3D Scan Data
3D scanned data can be of two main types: Point Clouds and mesh.

Mesh/Triangulation: This is a polyhedron 3D data that consists of: edges, points and faces.
A mesh is usually created from point clouds. Meshing is a process that connects 3 points
to construct a surface.

Figure 6. Meshing Process

6

Point Clouds: This is data that is composed of a group of vertices in 3D system. Each vertex
is defined and is correspondent to one position on the object’s surface. Point clouds
require conversion into mesh or CAD models before they can be used effectively.

4

Point Clouds
Types
Line

Random
No relationship
between points.
Only basic
information can be
altered.

Grid
This can be a 2.5D image
which uses a projection
and scanning direction.
Easily converted into
mesh by projecting onto a
planar, cylinder or
spherical coordinate.

Laser scanner is
used that produces
a line type point
cloud. The laser
emits lasers every
second forming
point clouds on the
surface of the
object.

Table 1. Types of Point Clouds

Mesh
Can measure surface curvature
Can be used for visualisation
Easily extract neighbouring point and
triangle information
Accurately calculate alignment
between scan data
Distinguish front and back faces

Point Clouds
Handle larger data files
Used for metrology based inspection
Can be used for visualisation
Used for simple measurements
Used for surveying

Table 2. Mesh and Point Clouds Comparison

3D Scan Data Conclusion
Point clouds will be used by both engineers, as this is the initial process for scanning.
Without the point clouds, the remainder of the project would cease to exist. However,
during the meshing process, both engineers will use different meshing techniques in order
to compare the accuracy and model reconstruction. Certain parameters will be kept
constant in order to ensure that there is some level of continuity and fairness during the
comparison.

5

Understanding Siemens NX 9.0 Software
Siemens NX software 9.0 is used for the purpose of this project as it is readily available and
is required to cover the scope of this project effectively.
NX is software closely related to solid works and covers the following aspects:
i.
Wireframe
ii.
Solid Modelling
iii.
Surfacing
iv.
Parametric design
Surfacing will be the main use of Siemens NX 9.0 software for the purpose of this project.
Surfacing uses the wireframe; this is where the geometry is displayed. When a surface is
not closed, it is known as an open loop. A surface and surface connection is known as a
solid.
Boolean is a technique that makes one model, these are comprised of primitives and can
be found within the library. The wireframe are analytical curves and these will be obtained
from the Geomagic software ready for surfacing sing the NX software.
There are a number of curves and related surfaces:
i.
Coon’s interpolation – Coon’s surfaces
ii.
Bezier curves - Bezier surfaces
iii.
Splines – Surface defined by splines
iv.
LaGrange’s splines – surface defined by l-curves
Approximation: This is creating a curve that does not necessarily pass through all the
points in a given set
Interpolation: This is creating a curve that does pass through all the points in a given set.
Interpolation leads to ‘strong waves’, which is first derivation. This will result in a surface
issue with steep points. Hence, the final result will be of poor quality.

6

2.0. Summary of Tasks
2.1. Part A- Surface Reconstruction Using GEOMAGIC & NXCAD
i. Part A - Aims & Requirement
Task A requires the use of good knowledge in reverse Engineering technology with the
implementation of advanced surface modelling techniques. This will be obtained through
the following 3 stages:
Stage 1 – Digitise 3D Part
Each engineer will use the same component to obtain a 3D digitisation of the component in
Geomagic software.
Stage 2 - Model Reconstruction
Stage 1 will create cloud points, which will be reconstructed using high advanced surface
modelling functions in Geomagic and NX software.
Stage 3 – Model Selection and Justification
The results of each engineer will be discussed and compared.

2.2. Part B- Mold Design, Optimization and Validation
ii. Part B - Aims & Requirement
To prepare the geometry of the selected CAD model and create an optimised mold for it.
Stage 1 – Simplifying the Chosen Model
The chosen CAD geometry is simplified, through the modification a configuration is chosen
to begin the molding process.
Stage 2 – Mold Setup
Setting up the mold assembly, defining the cavity and core in the workpiece, the addition of
standard parts.
Stage 3 – Mold Validation
Validating the mold configuration chosen, part of the validation process, where region
checks are performed will be done on stage2 –mold setup. Further validation techniques
used in industrial situations will also be explored by means of research.

3.0. Part A – Surface Reconstruction with Geomagic and NX
Component Selection Process
The scope of this project required a simple component with the use of curves in order to
effectively challenge the software whilst considering time constraints. Hence, a perfume
bottle was used. Primer was required as the scanner would not be able to effectively identify
the cloud points. The problem encountered in recognising unique points in this surface was
overcome by using small point markers that could omitted from the scan later.
Front

Back

Unused Components
The components shown below were scanned but not used for the reasons given:

Figure 7. Unused Components

Intricate features
were a main
concern as these
would prove
difficult to scan
effectively.

The thrusters and small
intricate details of the de-latch
mechanisms would prove
difficult. Also, the engineers
assumed that during the
smoothing phase, some of the
more intricate features would
be lost.

Interesting component as it
contained various curves and the
material was perfect to work with,
as it did not require priming.
However, the engineers
established that the moulding
process for this component would
be troublesome.

Stage 1 – Digitize 3D Part
Having scanned the perfume bottle, both engineers sought to assemble the parts to
reconstruct the model.
Assumptions and Considerations during this stage were as follows:
i.
Least number of cells will provide a faster process for surfacing and meshing
ii.
Standard Deviation must be as close to 0 as possible, this suggests that the data
is exactly the same, perfect alignment. A negative value is not possible and
therefore of this occurs, realignment should be reattempted.
iii.
The engineers used a previous component to practise the alignment process and
were advised that 0.001in Standard Deviation is ideal. However, as the scanned
component is in mm, this would convert to approx. 0.03mm. Hence, both
engineers will aim for this Standard Deviation.
iv.
Each stage should be saved as a different file to ensure that if any amendments
are required, they can be made. This is because Geomagic does not allow for
more than one undo process and there is no timeline to show the various
stages.
v.
During alignment, ‘n-point registration’ will be used to allow for three points to
be aligned. This option is not available in ‘1-point registration’ and would
otherwise prove inefficient in deriving the required model.
Engineer A – Initial Scans

Figure 7. Engineer A - Initial Scans

9

Engineer B- Initial Scans

Figure 8. Engineer B - Initial Scans

Conclusion for stage 1
Both engineers deduced that the primer coating on the perfume bottle was an effective
method to collect the points. It was also deduced that the slower the pace of the scanning,
the greater the quality of the points obtained in Geomagic.
In addition, the scanner position had to remain to the normal of the perfume bottle in order
to obtain cloud points. When black patches on the surface of the previously collected points
appeared, it was concluded that scanning over these points corrected the appearance and
quality of the scan.

10

Stage 2 – Model Reconstruction, Engineer A
Applying the Wrap tool and Mesh Doctor on the points:
The wrap tool dialogue box

After applying the wrap tool

Figure 9. engineer A - Applying Wrap and Mesh

Using the Mesh doctor tool to refine the wrapped geometry and remove holes and spikes
(smoothen out the surface) or remove unwanted sections:

Figure 10. Mesh Doctor

NOTE: but not smoothening too much to the point where the features on the side of the
bottle start to disappear.
Checking the surface for spikes or small holes that were not automatically identified by the
mesh doctor tool, and manually removing them using the fill holes feature
Before

After

Figure 11. Meshing Before and After

11

Trimming the top and the bottom of the wrapped geometry to achieve flat surfaces for the
bottle. Using the Trim with plane tool:

Bottom

Top

Figure 12. Trimming Process

Figure 13. Final Model

12

Creating a surface in Geomagic (Method 1):
Using the fit surface tool in the mesh created, this feature generates a NURBS surface on the
object, which can then be exported as CAD files on to other software

Figure 14. Wrapped geometry and conversion

Extracting planar curves from the Geomagic Wrapped Geometry (in order to be exported to
Siemens NX)
Using the ‘create curves by section’ tool in Geomagic 45 sections were initially created on
the horizontal axis of the surface in order to extract curves relating to the surface on each of
the created planes

Figure 15. Creating Curves

Shown below are the curves that were projected on the surface of the bottle. These curves
can now be saved as an .igs file
Note: the spacing and number of planes used were decided as a compromise between the
following key points:
Number of curves generated should not be too high in number, since generating a surface
from a large volume of lines is time consuming.
Number of curves should not be too low, some of the features of the original geometry face
may not be accurately captured as a result.
The first and last lines should be as close to the wrapped geometry as possible, so as to
ensure as much of the geometry can be described using these curves.

13

Figure 16. Projected Curves

Creating a surface in NX (Method 2):
Importing the .igs file on to NX, the curves created on the surface of the Geomagic geometry
have now been opened in NX

Figure 17. IGS File

Using the ‘through curves’ tool in NX, a surface is created, by selecting the curves through
which the surface must be in intersection with.

Figure 18. Through curves in NX

14

Creating a surface in NX (Method 3):
The same extracted curves from method one were imported to NX to create surfaces using
the swept curves tool.
Drawing splines – each vertical spline connects 3 curves (that will be used as the guide
curves)

Figure 19. Splines

Drawing splines in this method was continued along the length of the imported curves.
The reason one long spline was not drawn to be used as the swept curve is because the
geometry of the bottle surface looked too distorted when it was attempted.

Figure 20. Spline Vs Swept

Appearance of resulting surface that was created:

Figure 21. Final model

15

Stage 2 – Model Reconstruction, Engineer B
Group 3 showed an error; therefore this error (bulge) was removed from the structure,
which resulted in a loss of 26,990 cells hence allowing for a faster process.

Figure 22. Unnecessary Scan Identified

The structure appears as shown below without the scanned error.

Figure 23. Unnecessary scan Removed

Lasso Selection Tool allowed the engineers to remove the spikes that were used as
alignment points, using blue tack.

16

Figure 24. Spikes identified and Removed

Each part had to be combined into a single point object as they were made of two or more
point objects which would not be ideal for alignment. Group 5 was not Combined into Points
because it was a single point file, and there were no other scanned points to combine it
with.
The engineer then applied the Manual Registration Tool which creates a rough registration
of two overlapping scans by defining corresponding points in overlapping regions. The peaks
from the blue tack were removed and ‘n-point registration’ was chosen. This allows for the
alignment points to be defined.

Figure 25. Applying Alignment Method

17

Figure 26. Alignment of left hand side with front face

Upon ‘Registering’ the alignment, the following Standard Deviation was retrieved.
This deviation was compared to a previous component during
preliminary testing. The advised alignment requirement was0.001inches which equates to
0.0254mm. Hence, the deviation obtained for the first part of alignment, is ideal.
Final alignment, with 311,806 points. Excess points were removed and the structure was
redefined using the Loop Selection Tool.
The cells were then reduced to 311,620.
The following steps were then conducted in order to obtain the required reconstruction:
i.
Wrap – this tool allowed the engineer to create a more refined structure using
triangles
ii.
Remove Spikes – this tool allowed the engineer to flatten single point spikes
iii.
Relax – allowed the engineer to smoothen the mesh by minimising the angles
between individual polygons. The smoothness applied was the maximum.
Remove spikes was reapplied and this reduced the cell count by 30,000
iv.
Decimate – reduced the number of triangles without comprising the detail of
the surface.
v.
Trim – this cut the top and bottom of the model and rewrap was applied.

18

Original Mesh

Figure 27. Original Mesh before modifications

Refined to 0.0909mm, this resulted in the software not responding at each attempt. Hence,
it was deduced that a higher value of remesh length would be more appropriate based on
time constraints. There had to be a balance between the number of triangles and the mesh
produced.

Figure 28. Remesh attempt 2

19

Figure 29. Remesh Attempt 3

Figure 30. Remesh Attempt 4

The meshing with Total Edge Length of 2.071mm was chosen based on the following:
i.
The triangles are not too large in comparison
ii.
The triangles are mainly equilateral triangles
iii.
Redundant triangles are not considered
iv.
Boundaries are excluded means the mesh is not accurate
v.
Reversed triangles which show opposite direction to neighbouring triangles

20

Figure 31. Applied chosen mesh

However, the engineer chose to smooth out the defined mesh by using the Relax icon to
minimise the angles between the polygons, the result is shown below.

Figure 32. Engineer B - Final model

21

Creating a surface in NX (Method 4):
Curves were added to the structure in order to remodel in NX software. This was done by
selecting Curves by Section for the XY plane. The second engineer chose to work with XZ
plane, therefore, as a way of comparing, the XY plane was selected.
56 curves were produced, as shown in the figure below.

Figure 33. curves Added in Geomagic

This file was then saved as an iges file for use in NX software where the curves would be
used for reconstruction of the model, as shown in the figure below.

Figure 34. Curves as shown in NX

22

Figure 35. surfacing in NX

Though a greater number of curves would have greatly defined the model, it was concluded
that the selected 56 curves would be efficient for the purpose of this project as a greater
number than this would result in: slow software and remodelling process.
Conclusion for stage 2
This stage required the reconstruction of the scans obtained by each engineer. This was
achieved by using the FAROArm laser scanner, which scanned the component with the use
of a primer on the surface of the component, making the scanning process more efficient.
The scanned data was processed using Geomagic software, which identified the data as
cloud points.
The scanned points were realigned to reconstruct the component, meshing and surfacing
was conducted in order to refine the reconstructed component.
Each engineer then exported their scanned and remodelled component into NX software
and with the use of curves, remodelled the structure in NX.
This proved rather testing for each engineer as they had chosen to work in different planes
for the NX reconstruction. Engineer A worked in the XY plane, whilst Engineer B worked in
the XZ plane. The XZ plane proved more challenging during the surfacing phase.

23

Stage 3 – Model Selection and Justification
The base dimensions of the reconstructed components in NX is shown below,
Engineer A

Engineer B

Figure 36. Dimensional Analysis

Original Bottle base
Engineer A
Engineer B

40mm (4cm measured)
41.6511mm
34.0346mm

The bottom and top of the component were trimmed in both cases; therefore, the
dimensions cannot be as accurate. Therefore the chosen measurement is not a 100%
reliable means of checking the distances with that of the original bottle.
Since the ‘through curves’ surface generated with curves generated in multiple XY planes
yielded a smoother surface geometry, Engineer A’s initially generated surface (Method 1)
was selected for the progression to Section B (Mold Design)
Conclusion for stage 3
The measurements taken for the purpose of comparison was the longitudinal length of the
base of the bottle, however since the top and bottom edges of the bottle merged scan were
trimmed off in Geomagic, during the geometry generation from cloud points. It is not
feasible to assume that the generated surfaces will have exact dimensions as the original
bottle. However the measurement values from the generated surfaces are not too far off
from the original.
The main reason for choosing Method 1, is because the method in which the surface was
generated from curves was relatively simple and clean-cut compared to method 2, method 1
produced a much more smooth surface in comparison to method 4. Although the Geomagic
surface was the closes to the accurate solid geometry, Method one was chosen for part 2 of
this report, since NX is much more user friendly and produces a much more smooth curve.

24

3.0. Part B – Mould Design, Validation & Optimisation
Aim and objectives
Aim: To prepare the geometry of the selected CAD model and create an optimised mold
for it.
Objectives:
 Selecting a CAD model
 Validating the model for moldability and initialising the molding process
 Define the coordinate system for the mold (selecting the mold CZ relative to the
WCS)
 Defining the workpiece in which the cavity and core will be set.
 Performing the parting of the workpiece.
 Adding a mold base.
 Adding other standard parts to the mold: a sprue, bushings, ejector pins, runners,
gates etc.
 Creating pockets.
The selected CAD model was the surface created using the ‘through curves’ feature using
Siemens NX. The CAD geometry chosen is to be simplified for toe the purpose of molding.
The results of the mold design have been validated using built in HD3D tool in the
Siemens NX software.
In this section the design of the mold assumes that the type of molding used is ‘injection
molding of thermoplastics’. According to (Rebling Custom Molding, 2015), this is the most
common method of manufacturing plastic parts. In this process a pelletized thermoplastic
material is gravity fed into a heated barrel and screw, the screw rotation feeds the molten
thermoplastic into a closed mold at a high pressure through the runners and gate system.
Once the mold cavities are filled the cooling cycle commences. Once the part is rigid
enough it is ejected and removed.

25

Stage1 - Simplifying the Chosen Model
After the Reverse engineering processes performed in task A; and a replica (as close to the
original geometry as possible) of the scanned object was created as a CAD model, it must
now be simplified in order to make it more convenient to reproduce the geometry through
the molding process.
Figure 378. Closing the Bottom of the hollow bottle surface

The surface was created as a sheet
not a solid (for practical reasons,
considering that a bottle is hollow)
This meant that the bottom and
the top sections were left as open
holes, using the ‘N-sided curves’
tool and selecting the ‘trim to
boundary’ setting, the bottom of
the bottle can be closed up

Figure 389. Adding a thickness

It was not possible to measure the
thickness of the glass bottle
without shattering/damaging it,
therefore, an inferred (inner)
thickness of 2.5mm was given to
the bottle surfaces in order to carry
on with mold design.
A solid geometry can now be seen

Figure 40. Cutting in half

Since the bottle is symmetric only
half of it is necessary to be molded.
The most appropriate plane in
which to cut geometry was chosen
as shown to the left.
It was ensured that the plane
bisecting the bottle was symmetric
to its geometry.

26

Figure 41. Appearance of sectioned bottle geometry

After
sectioning
the
bottle
geometry in half, the picture to the
left represents what a single mold
output would look like.
There fore when choosing the mold
layout it is necessary to be able
produce two bottle sections at
once.

Stage 2 - Mold Setup
Checking regions and assigning Cavity and Core Faces
An initial check is performed to see if
the cavity region has been defined
accurately.
Cavity
region
is
highlighted in orange. Blue is the
core regions.
The default selection in the check
regions is reversed, so that the inner
surfaces of the geometry were
chosen as the core region and the
outer region as the cavity section.

Figure 42. Region check

27

Project initialisation
To initialize a project, it’s necessary to supply the name, units, material, and template
configuration. In the mold initialization stage, a new tooling assembly is created based on a
template built in to NX. The material chosen is ABS (a type of resin), the choice of material
will automatically update the shrinkage – in this case a shrinking factor of 1.006
(MiSUMi, 2009) Mold shrinkage is when the molten plastic filled inside the cavity of a mold
is shrinking at the time as being cooled and solidifying. The amount of shrinkage is called the
“molding shrinkage factor”, and if this is known accurately, by preparing the mold making
the increasing the dimensions of the cavity by amount of shrinkage, it’s possible to have the
objects intended dimensions.

Figure 43. Project initialisation attributes

(Chilson, 2013) ABS is a high grade rapid prototyping resin (plastic), as a polymer ABS can
be engineered to have many desired properties.
In this case, the ABS material used (once the mold has been set and cooled) is a strong
plastic with mild flexibility. Colouring can be added to ABS and it can be easily sanded or
machined. ABS also possesses a high temperature resistance and is often the preferred
plastic for engineers and professional applications. ABS is soluble in Acetone so it allows
one to easily weld parts together (that will be beneficial in this case since the 2 bottle
section parts are meant to be welded together)

28

Creating the Workpiece
This is the solid block that will be machined/milled and made into the core and the cavity
parts.

Figure 44. workpiece

Setting WCS on a good position of the bottle geometry

Figure 45. Assigning WCS

Defining the workpiece and positioning it relative to the bottle section geometry

Figure 46. Workpiece positioning

29

In this case a symmetric layout is chosen for the two halves of the bottle.

Figure 47. Defining Cavity Layout

Adding pockets into the workpiece, in this case a type 2 pocket configuration was chosen
from the defaults list in NX, each of the 4 pockets on the corners of the workpiece will
contain a radius of 15mm

Figure 48. Adding Pockets initially

Appearance of the workpiece after adding Type 2 pockets with a radius of 15 mm

Figure 49. Symmetric workpiece with pockets

30

Analysing the Parting Regions in the bottle section solid geometry
It is important to use the ‘Check Regions’ function to check the progress and accuracy of the
process so far, the check at this stage validates the following:


The parting loop around the bottom perimeter of the bottle section – to ensure that
the correct regions have been chosen for where the cavity and the core will part in
order to remove the ABS bottle section once the mould has been set.



Faces formed by the core and cavity – a check must be performed to see if the right
surfaces have been selected for each of the cavity and core regions

Figure 50. Check regions function

NOTE: under the undefined region, there are no faces identified, this validates that all the
faces of the bottle section surface has been accounted for and assigned to either the core or
the cavity regions. At this stage is necessary to ensure there are no undefined regions to
proceed with the creation of the mold. The cavity regions are displayed in orange and the
core regions are displayed in blue.
Checking core and cavity regions separately:
Figure 51. Cavity region

Figure 52. Core region

31

Creating the parting surfaces:
Parting surfaces where created by manually selecting the outer bottom perimeter of the
bottle section as parting lines, as NX defaults cannot recognise the best parting selection
for the bottle geometry.

Figure 53. Parting surface

Mold Wizard selects the appropriate method for the part (in this case the ‘Bounded Plane’
tool is selected) and automatically generates the surface and the appropriate parting
loops.

Figure 54. Parting surface selection to a bounded plane

Once the parting surface has been created the NX software copies the sheets onto the other
symmetric block of the mold and sews them together to form a continuous parting section
for both bottle sections.
Displayed below are the final appearances of the cavity and core blocks that have now been
created. These two will fit together to create the two ABS bottle parts.
Figure 55. Cavity has been created

Figure 56. Core region created

32

Figure 57. Workpiece final assembly with parting line.

NOTE: the parting line (light blue line) is now visible throughout the whole workpiece.

Adding the Mold Base:
A standard mold base is added to the mold assembly (using the default built in library in
NX) The 2A type mold base was selected.

Figure 58. Selecting 2A mould base

The mold base index size has been
automatically selected from the sizes
generated by the ‘Cavity Layout’ tool

Figure 59. Selecting plate thickness for cavity

33

Representation of the mold layout at this point (with a mold base):

Figure 60. Mold base layout

34

Adding Standard Parts
In the Mold Wizard Standard Parts Library, you can find many purchased components
from various manufacturers, including locating rings, sprue bushings, and core pins
1) Adding the injector from the MW Standard Part Library (from the Injection folder)
Inserting locating ring with screws:

Figure 61. Injector setup

1) Inserting Sprue

Figure 62. Sprue setup

The translucent false bodies (dotted line circles below the sprue) show the tap diameter
and drill depth for the mounting screws.

2) Adding Ejector Pins (straight ejector pins)

Figure 63. Ejector pin dimensions

35

Four ejector pins were added to the geometry in 4 locations as shown below

Figure 64. Ejector pin locations

Trimming ejector pins – ejector pins have to be trimmed where they are in contact with
the molded part so that they don’t interfere in the cavity between the mold parts.

Figure 65. Trimming ejector pins

3) Creating pockets and using the HD3D tool to check the work done so far
Top three plates in green will be
cut to all the locating ring, the
screws related to the locating ring,
ejector pins and other inserts

Figure 66. Selecting top planes to cut
Reviewing selection to verify if this
step has been done correctly

Figure 67. Reviewing selection

36

Figure 68. Mold base with pockets

Results of the HD3D tool: passed with information, therefore there are no nonfeasible design features in this model up to this point.

37

4) Adding gates - submarine type gates were chosen from the default templates
available in the NX parts library.
Figure 69. Selecting the gate point

One of the product subassemblies
chosen as the work part

Figure 70. Position of gates

Figure 71. Gate position – front view

Another more clearer view of the gate
positions from the front view

38

5) Runner Design – Cylindrical runners were chosen, it is easier for molten
thermoplastics to flow through a these types of runners
Figure 72. Drawing runner guide curve

Figure 73. Runner specifications

Figure 74. Final view of runner system

Runner system would look like this,
connected to the sprue and ring

39

Getting started with the mold – additional pockets for the added parts have to now be
created.
Figure 75. Selecting targets to
create pockets

To house the gates and
runners (mold base is
hidden)

Figure 76. Appearance of gate
and runner pockets created
on cavity and core

Pockets for runners
displayed on the cavity,
the same was applied for
the core
Figure 77. Creating pockets
for the ejector pins on the
core.

Figure 78. Final appearance of
all the pockets on the cavity
and core parts.

40

Stage 3 –Discussion and Mold validation

Throughout the course of the mold design
using Siemens NX, several checks were
performed where both the cavity and core
regions were checked using the Check
Regions tool. Later on after the mold base
was added the HD3D tool was used to
check if there are any problems with the
mold design. The aim is to pass all the
HD3D tests with the end mold design.
More than one of these region and HD3D
tests were performed.
According to (Polymer Processing, 2009)
injection mold validation is done step by
step with the aid of an injection mold
validation flowchart as shown in Figure 79.
The 12 steps that are required for this type
of mold validation are as follows:
 Mold certification
 Dry cycle mold
 Process stability test
 Gage repeatability & reproducibility
(R&R) test
 Mold viscosity test
 Balance of fill analysis
 Gate freeze test
 Commissioning (multi-cavity analysis)
 Design of experiments
 Qualification (process capability study)
 Mold metal Adjustments - cantering
process
 Verification (30-day run)

Figure 79. Injection Mold Validation Flowchart

7

41

References
1

Reverse Engineering . (2014). Reverse Engineering Services.
Available:
http://www.reverseengineeringoutsourcing.com/images/reverse-engineeringservices.jpg. Last accessed 2015.
2

FAROArm, (2015). FAROARm Measuring Arm.
Available:
http://www.faro.com/en-us/products/metrology/measuring-arm-faroarm/overview
Last accessed 2015.
3, 4, 5, 6

Geomagic User Guide, (2014). 3D Data Types
Available:
http://dl.geomagic.com/Geomagic2014/Rapidform/GeomagicDesignX2014UserGuide.pdf
Last accessed 2015.
7

Polymer Processing. (2009, 08 11). Mold certification - Validation procedure for injection
molds. Retrieved 03 19, 2015, from Polymer processing:
http://polymerprocessing.blogspot.co.uk/2009/08/injection-mold-validation-flowchart.html
McMahon C & Browne J, CADCAM: Principles, Practice and Manufacturing Management,
2000, ISBN: 0201565021.
Beaumont J.P, Runner and gating design handbook: tools for
successful injection moulding, 2004, ISBN: 1569903476
Beaumont J.P, Nagel R, Sherman R, Successful injection moulding: process, design, and
simulation, 2002, ISBN: 1569902917.
Fuh J.Y.H, Computer-aided injection mould design
and manufacture, 2004, ISBN: 0824753143.
Chilson, L. (2013, 01 27). ProtoParadigm. Retrieved 03 19, 2015, from The Difference
Between ABS and PLA for 3D Printing: http://www.protoparadigm.com/news-updates/thedifference-between-abs-and-pla-for-3d-printing/
MiSUMi. (2009, 07 13). Plastic Molding Tutorial . Retrieved 03 19, 2015, from Misumi
Technical Tutorial: http://www.misumi-techcentral.com/tt/en/mold/2009/07/0001-what-isthe-molding-shrinkage-phenomenon.html
Rebling Custom Molding. (2015). Rebling Plastics. Retrieved 03 19, 2015, from Thermoplastic
Injection Molding: http://www.reblingplastics.com/processing-tpim.htm

42