See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/228520972

A GENERIC FRAMEWORK FOR RAPID DEVELOPMENT OF INJECTION MOLDS

AND PRESSURE DIE CASTING DIES

Article

CITATION READS

1 1,356

3 authors, including:

Nagahanumaiah , B. Ravi
Central Manufacturing Technology Institute, Bengaluru, India Indian Institute of Technology Bombay
80 PUBLICATIONS   761 CITATIONS    128 PUBLICATIONS   2,152 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Reinventing Metal Casting: SMART Foundry 2020 View project

Bacterial adhesion on nano and micro patterned surfaces View project

All content following this page was uploaded by B. Ravi on 30 April 2017.

The user has requested enhancement of the downloaded file.

 A GENERIC FRAMEWORK FOR RAPID
DEVELOPMENT OF INJECTION MOLDS AND
PRESSURE DIE CASTING DIES

Nagahanumaiah
Scientist, Central Mechanical Engineering Research Institute, Durgapur, India
Email: [email protected]

Bhallamudi Ravi
Associate Professor, Mechanical Engineering Department, IIT Bombay, India
Email: [email protected]

Narayan Prasad Mukherjee

Scientist, Central Mechanical Engineering Research Institute, Durgapur, India
Email: [email protected]

ABSTRACT
Rapid product development (RPD), a strategic solution to survive in global competition is
characterized as the process chain that ensures development of flexible products in shorter
lead-time at competitive cost. Even though die and mold development is recognized as a major
step, tooling development procedure is not yet very well documented and is experience based,
resulting in higher cost and longer lead-time. A research attempt has been made to develop a
generic framework for rapid development of dies and molds, which comprises the
methodologies for entire tooling development process starting from tooling requirements
modeling, process modeling to rapid hard tooling development. This paper presents the
methodology developed for each stage particularly for injection molding and pressure die-
casting. The methodology can be easily customized for any tool room, enabling reduction in
lead-time and cost through minimization of tooling iterations and backflows, demonstrated by
industrial examples. Implementation of this framework is proposed for future tool rooms.

1. INTRODUCTION
Near net shape (NNS) manufacturing is gaining wide acceptance for discrete parts
manufacturing to meet the changes in market demand. This includes injection
molding and pressure diecasting. Currently, most injection-molded parts are directly
assembled, without any secondary machining. However, pressure diecast parts have
an accuracy up to ±20µm and surface finish Ra 0.8 - 1µm. One main concern of
these developments is tooling, which currently consumes significant amount of time,
and cost, and requires expertise.

The worldwide tooling industry is currently dominated by USA, Japan, Germany,

Korea, Taiwan, China, Malaysia, Singapore and India. With increasing
globalization, the tooling users (product manufacturers) are demanding rapid
die/mold development, lower cost, better accuracy, better surface finish, shorter
production runs and flexibility to accommodate future design changes. To meet
these requirements, tool rooms need to explore and evaluate new approaches. In line
with the above, we propose a generic methodology for rapid development of dies
and molds for functional prototypes and small order production.
 2
CARs&FOF’2003

2. PREVIOUS WORK
Tooling is regarded as a key bottleneck in rapid product development (RPD) and
several researchers have focussed their attention on this activity. Based on the
review of available literature, research progress on rapid development of dies and
molds can be classified into two broad categories: computer applications and
manufacturing systems.

2.1. Computer Applications in Tooling Development

In die and mold making, computer programs have mainly been used for tool design,
process simulation and tool path generation. The initial work on this was started late
in 1980. Ravi and Srinivasan established the decision criteria for automatic parting
line selection using CAD model as input [Ravi, 1990]. Nee developed computer-
aided program for automatic extraction of tooling elements of injection molds [Nee,
et al, 1998]. Lu developed integrated manufacturability evaluation using geometric
reasoning and knowledge base system (KBS) to support concurrent engineering [Lu,
et al, 1997]. Lee developed a concurrent injection mold design system using KBS
[Lee, et al, 1997]. However, there is still a long way to go before fully automated
die/mold design systems are developed for practical application.

Mold filling and solidification simulation have become indispensable for predicting
potential quality problems and for optimizing the process parameters [Barkhudov,
1997a]. Most of the simulation tools use either finite difference or finite element
methods for discritizing the product geometry. Devey used FEM for filling
simulation and BEM for heat transfer simulation of pressure die casting dies [Devey
et al, 1997]. Coupling these two continues to be a challenging problem, since they
require different mesh size and time span. The computational fluid dynamics
program ‘Flow-3D’ incorporated fractional area/volume obstacle representation
method into the Navier stokes equation of fluid motion, to solve this problem to a
large extent [Barkhudov, 1997b]. These programs are however, computationally
expensive and require an accurate database of thermo-physical properties as well as
metal-mold heat transfer coefficients, which limits their widespread application.

A recent development is the application of virtual prototyping approach for mold

design so that geometric moldability of NNS parts can be assessed [Yin, et al, 2001].
After the geometric design of the die/mold, the NC tool path generation is well
established and widely practiced in the industry.

2.2. Research Progress in Die and Mold Manufacturing Systems

The major research contribution in die and mold manufacturing systems are:
incorporation of concurrent engineering, high speed machining, hardened die steel
machining and rapid tooling. The survey in 1996 among the die and mold
manufacturing industries of USA, Japan and Germany indicated that die and mold
manufacturing system must be considered as a complete system comprising process
planning, part and process design, NC tool path generation, machining, polishing
and tryout [Fallbohmer, et al, 1996]. The linear approach, where activities were
performed in isolation and in a serial manner, is now changed to radial approach, in
which a multidisciplinary team is responsible for successful delivery of dies and
molds [Altan, et al, 1993].

High speed machining of hardened die steel emerges as a cost effective technology
and offers several advantages like reduction of finishing operations, elimination of
Framework for Rapid development of Dies and Molds 3

thermal distortion induced in heat treatment, lower machining cost and increased
surface finish [Fallbohmer, et al, 2000]. Five-axis machining is another promising
approach for lead-time reduction in die making. However, it suffers from
insufficient support from tool path generation algorithms, cutting tool geometry
optimizations, which are the current research topics in this field.

Soft and bridge tooling (rapid tooling) developed by non-conventional material

(other than die steel) has proved to be a cost effective and rapid solution to meet the
changes in demand. However, many of them failed to produce functional prototypes
(without sacrificing the tool material), near-net shape and quality level. This has led
to the development of newer methodologies for rapid hard tooling. This includes 3D
Keltool, SLS (selective laser sintering) mold, SDM (shape deposition modeling)
steel mold, cast mold and RP integrated HIP (hot isostatic pressing). The 3D Keltool
is able to produce inserts either in A6 composite tool steel or stellite, which can
withstand even up to 0.5 millions of injection molding shots [Campbell, 1997]. The
SLS can build the Ni based alloy or stainless steel metal mold [Radstok, 1999]. The
SDM builds the tool steel mold with hardness up to 40HRc, which is even tested
under pressure die-casting. These molds produce die-castings with quality
comparable to that with H13 [Link, et al, 1998]. A few researchers have attempted to
incorporate HIP with RP to develop denser mold inserts [Agarwala, et al, 1999].
However, the acceptability of these rapid hard tooling needs exhaustive data both on
process development and their performance in conventional NNS processes, which
is currently missing.

In summary, there is a growing body of research on rapid development of molds and

dies, covering different facets of tooling. However, these developments can reach
the toolrooms only when their benefits are quantified and enabling technologies to
utilize them in an integrated manner are made available. Some research progress like
process selection, process modeling, quality assurance techniques and web based
collaborative engineering has been found in product development. However, to the
best of our knowledge, no research attempt has been made in this direction to suit
the tooling domain. This is taken up for investigation in the present work.

3. A FRAMEWORK FOR FUTURE TOOL ROOMS

Rapid manufacturing is going to be the future-manufacturing concept, in which tool
rooms have a major role to play. Indeed, they need to shift from conventional
experience-oriented approach to a more scientific and flexible approach. However,
more than 60% of toolrooms are small-scale industries, and it is difficult to expect
all of them to rapid absorb the new developments. We therefore, propose a two-tier
framework for future toolrooms, called as ‘Tooling Solution Center (TSC)’ as shown
in figure 1. Smaller tool rooms can collaborate with such centers to expedite die
development. Presently, many of the components of the proposed TSC are relatively
new for the tooling domain, and will need further development. Our research
focuses on developing a collaborative engineering framework for the effective
functioning of these TSCs for rapid development of injection molds and pressure die
casting dies. This is discussed in detailed in the following sections.
 4
CARs&FOF’2003

Figure 1: Tooling Solution Center

3.1 Tooling Requirements Modeling

Conventionally, a tool designer starts his design by considering the product
geometry, however, success of tooling depends on many more factors like desired
accuracy, expected die life, scope for future changes, etc. We have summarized the
factors associated with tooling development in figure 2.

Figure 2: Die and mold development process variables

Framework for Rapid development of Dies and Molds 5

During close interactions with tool and die making industries, we found that failure
of many projects is mainly because of improper approach or less importance given
to initial requirements analysis and prioritization. This prompted us to develope a
Quality Function Deployment (QFD) model for tooling requirements analysis. The
methodology is as follows.
1. Use product data model (CAD model or 2D drawings) as input and identify the
critical geometric factors.
2. Set the desired objectives for specific die and mold project. For example, we
identified NNS, tool design automation, die design evaluation, better accuracy,
minimum cost, shorter lead time, die life, flexibility and information
management (data handling between design & manufacturing) as desired
objectives for rapid development of molds and dies. Any other requirement,
which is project specific or customer specific can be added in the first column of
the QFD chart.
3. Assign the weights (customer preference) for each requirement on 1-5 or 1-10
scale depending on the case.
4. Identify the tooling development process variables (please see the figure 2) and
represent in columns.
5. Establish the relationship matrix considering the impact of each requirement on
individual process variables. Use 1-3-9 scales to represent weak, medium and
strong relations.
6. Construct the correlation matrix (roof of QFD chart) considering the correlation
among the process variables, use 0.1-0.3-0.9 scale to represent weak, medium
and strong relations.
7. Normalize the QFD matrix, either by weighted average method or by using
Wasserman method that considers coefficients of both relationship matrix and co
relationship matrix. In Wasserman method the co-efficient of normalized matrix
is given by the following equation

∑ (r i. j .γ k . j )
r norm
i. j = k =1
m m

∑∑ (r
j =1 k =1
i. j .γ j .k )

where,
ri.j = co-efficient of relationship matrix
γj.k = co-efficient of correlation matrix
8. Rate and prioritize the process variables based on their technical importance

This analysis will help the tool development team to identify the critical areas,
which need further study or special attention and also plan the activity so that
iterations and thereby delays are minimized.
 6
CARs&FOF’2003

Figure 3: QFD model for tooling requirements analysis

3.2. QFD-AHP Framework for Tooling Process Selection

In rapid manufacturing, selection of the appropriate tooling process is most critical,
which ultimately ensures minimum iterations and rapid development. A QFD-AHP
(quality function deployment - analytical hierarchy process) framework is proposed
for rapid hard tooling process selection. Integration of these two mathematical tools
increases the reliability of decision analysis and this can be used when the
preferences are subjective as well as objective. The methodology is as follows:
1. Identify the desired objectives/criteria for tooling process selection. For
example, minimum cost, shorter lead-time, better accuracy, good surface finish,
more strength and flexibility are considered as major criteria.
2. An Analytical Hierarchy Process model is used to prioritize toolmakers
requirements. In this case toolmaker’s satisfaction would be the top level, while
second level consists of the six criteria (cost, lead time, accuracy, surface finish,
Framework for Rapid development of Dies and Molds 7

strength and flexibility). Each of these criteria would have attributes like
geometric features, type of material, material phase, number of stages etc.
3. The pairwise comparisons are made in all levels of AHP structure using Saaty’s
methodology to prioritize the tooling selection criteria.

Figure 4: Hierarchical structure for tooling process selection criterion

4. The QFD chart is built with rapid hard tooling processes as its rows and
toolmaker’s requirements as columns. The relationship between the rows and
columns will be measured as ‘very weak, ‘weak’, ‘average’, strong, ‘very
strong’ and mapped in 1-9 scale.
5. The roof of the QFD chart, the correlation between selection criteria, is
completed. If the toolmaker/appraiser feels that an increase in one factor results
in the increase of another factor then two factors are positively correlated. The
four types of correlation are ‘strong negative’, ‘negative’, ‘positive’, ‘strong
positive’ mapped in 0.1 to 0.9 scales.
6. Normalize the relationship matrix by weighted average method or Wasserman
normalization technique as described earlier.
7. Calculate the weighted sum importance and relative importance.
8. Select the best rapid hard tooling process.

3.3. Rapid Hard Tooling Development

As discussed in section 2 all rapid hard tooling processes are yet to receive wide
acceptance by toolmakers, demanding detailed experimental investigations and
technical advancements. We are investigating the available rapid tooling methods
for suitability to injection molding and pressure die-casting. The manufacturing
knowledge generated on tool manufacturability (geometry, distortion, size
limitations, accuracy, etc) and performance under conventional injection molding
and pressure die casting process (die life, product quality, failure analysis) will be
used for developing a decision support system for tooling process modeling and
selection. We have already completed a few such experimental investigations, and
the results on an industrial example are presented in the next section.

4. INDUSTRIAL EXAMPLE
The above said methodologies were implemented for industrial cases to evaluate
their application to real tooling problems. One such case is discussed here. A small-
scale industry approached CMERI for developing the mold for the component ‘Hub
Gear’, which is a part of a washing machine. The customer requirement was
development of initial batches of 2000 injection moldings, without injuring the
 8
CARs&FOF’2003

designed material and the process. We used our proposed framework to develop
these injection molds rapidly, described below.

Requirements modeling: After a detailed deliberation, customer requirements were

tabulated and nine primary requirements were considered for further analysis. Mold
development process variables were identified based on the geometric inputs of
product (CAD model is used). QFD model as shown in figure 3 is developed for
prioritizing the tooling development process. The product geometry calls for twelve
thin ribs (2.28mm) to lie on both the halves (core & cavity), and parting surface
splits the critical dimension. The QFD model identified that tool manufacturability
and parting surface are the crucial task.

Process modeling and selection: The cost of a conventional mold was estimated to
be about US$3000 and the lead-time as 10-12 weeks. We considered four alternative
rapid tooling processes: SLA-AIM, SLS metal mold, spray metal tooling and
conventional tooling. The QFD-AHP framework is developed as discussed in
section 3.2. This analysis suggested spray metal tooling as more cost effective to
produce about 2000 functional prototypes (database is purely based on literatures).

Spray metal tooling (bridge tooling): The spray metal mold is produced using
TAFA metal spray system using MCP400 material (figure 8). These mold inserts
were assembled and processed in injection molding. The first set of molds collapsed
under the clamping force 65KN. Then a second set of molds were produced and
used for injection molding HDPE. After about 120 shots, the sprayed shell started
peeling off, and the moldings had several sink marks. A third set of molds were
tested by injecting Nylon 66. This time the sprayed shell started peeling off after just
40 shots, and the core features ultimately got pulled off along with molding during
ejection before 70th shot. The failure analysis identified three types of failure:
metallurgical (poor bonding between spray shell and backup material), compression
failure (rough parting surface) and tensile failure (low tensile strength and need for
more draft).

Rapid hard tooling development: This involved developing SLS metal molds using
EOS selective laser sintering system followed by injection molding. These molds
were satisfactory except for surface roughness due to the ‘staircase’ effect and
distortion (produced during sintering of the mold). The molds is currently in use.

a. Components b. Spray metal mold c. SLS metal mold

Figure 8: Industrial Example
5. CONCLUSIONS AND FUTURE WORK
Rapid manufacturing has become essential for manufacturers to survive in the global
market. A collaborative engineering framework called ‘Tooling Solution Center’ has
been proposed for future toolrooms to enhance their overall competitiveness. It is
based on methodologies like QFD for tooling requirements analysis and QFD-AHP
algorithm for tooling process selection. Few experimental investigations on rapid
 Framework for Rapid development of Dies and Molds 9

tooling performance analysis have been presented using an industrial example. The
initial applications of the methodologies developed under this framework have
received a positive response from the industry. However, the main limitation of this
approach has been identified as being cumbersome for a typical product
designer/tool designer. This is being tackled by the development of a decision
support system. Further work is needed to quantify the benefits of the approach and
enable the end-users to access the tooling solution centers over the web.

6. ACKNOWLEDGEMENTS
The assistance of Central Manufacturing Technology Institute, Bangalore, is
acknowledged in developing the SLS molds.

7. REFERENCES
1. Agarwala Mukesh, Kolosterman Donald, Osboren Nora, Lightman Allon, “ Hard Metal Tooling via
SFF of Ceramics and Powder Metallury”, Solid Freeform Fabrication Symposium, Austin, Texas,
August, 1999.
2. Altan.T, Lilly.B.W., Kruth.J.P., Leaven.K.U., Konig.W., Tonshoff.H.K., Luttervelt Van.C.A,
Khairy.A.B, “Advanced Tecnologies for Die and Mold Manufacturing”, Annals of the CIRP, Vol.42,
No.2, 1993, pp 707-716.
3. Barkhudorov.M.R. “Is the Fluid Flow is Important for Predicting Solidification” Presented at the
solidification Proceedings 97 Conference, 7-10, July 1997a, Sheffield, UK.
4. Barkhudorov.M.R., “Advanced Simulation of the Flow and Heat Transfer Process in Simultaneous
Engineering”, presented at the casting 1997-International ADI and Simulation Conference, Helsinki,
Finland, May 28-30, 1997b
5. Campbell Clark, Blain Colin, “Virtual Capacity- Rapid Hard Tooling of Experiences”, 9th ESUA,
Nov. 2-5, 1997, Florance, Italy, pp 1-5
6. Davey.K,Bounds.S., “Modeling the Pressure Die Casting Process Using Boundary and Finite
Element Methods”, Journal of Materials Processing Technology, Vol. 63, 1997, pp 696-700.
7. Fallbohmer.P., Altan.T., Tonshoff.H.K., Nakagawa.T, ‘Survey of the Die and Mold Manufacturing
Industry- Practices in Germany, Japan and United States’, Journal of Materials Processing
Technology, Vol.59, 1996, pp 158-168.
8. Fallbohmer.P., Rodriguiz.C.A., Ozel.T., Altan.T., “High Speed Machining of Cast Iron and Alloy
Steel for Die and Mold Manufacturing”, Journal of Materials Processing Technology, Vo.98, 2002,
pp104-115.
9. Lee Rong-Shean, Chen Yuh-Min, Lee Chang-Zou, “Development of a Concurrent Mold Design
System: A Knowledge –Based Approach” Computer Integrated Manufacturing Systems, Vol.10, No.
4, 1997, pp 287-307.
10. Link Ramdeen Goyle, Fesser John, Nickel Alax, Prinz Fritz, “Rapid Tooling Die Cast Inserts using
Shape Deposition Manufacturing”, Materials and Manufacturing Processes, Vol.13, No.2, 1998, pp
263-274.
11. Lu.S.F., Rebello.A.B., Miller.R.A., Kinzel.G.L., Yagel.R., “Applications- A Simple Visualization
Tool to Support Concurrent Engineering Design” Computer Aided Design, Vol.25, No.10. 1997, pp
727 –735.
12. Nee.A.Y.C., Fu.M.W., Fuh J.Y.H, Lee.K.S., Zhang.Y.F., “Automatic Determination of 3-D Parting
Lines and Surfaces in Plastic Injection Mold Design” Annals of the CIRP, Vol.47/1, 1998, pp 95-98.
13. Ravi.B, Srinivasan.M.N., “Decision Criteria for Computer –Aided Parting Surface Design”,
Computer Aided Design, Vol.22, No.1, 1990, pp 11-18.
14. Radstok Eric, “Rapid Tooling”, Rapid Prototyping Journal, Vol.5, No.4, 1999, pp 164-168.
15. Yin ZhouPing, Ding Han, Xiong YouLun, “Virtual Prototyping of Mold Design: Geometric
Moldability Analysis for Near-Net-Shape Manufactured Parts by Feature recognition and Geometric
Reasoning”, Computer Aided Design, Vol.33, 2001, pp 137-154.

View publication stats