INJECTION MOLDING

Introduction

Injection molding is an extensive global manufacturing process for making sim-
ple to intricate plastic, ceramic, and metal parts. Injection molding converts wax,
thermoplastics, thermosets as well as powdered metals, and magnesium into thou-
sands of products (1,2). Commercial processes and equipment have been developed
for a wide variety of materials, though the bulk of what is injection molded are ther-
moplastics. Applications appear limitless, from compact discs to jet plane canopies,
to medical implants. Nearly all consumer products have some injection-molded
components. In the year 2000 injection molding consumed approximately 32 wt%
of all plastics sold worldwide, roughly 4.95

× 10

7

metric tons (3). This is second

to extrusion which consumed 36 wt% of all plastics sold worldwide. The United
States consumes about 33% of the plastic sold worldwide. In the United States,
the plastics industry ranks as the fourth largest after motor vehicles, petroleum
refining, and automotive parts. Plastic processing and products continue to grow
at a faster pace than the national average. This is in spite of the trend of the
United States going toward an information and service industry and away from
manufacturing. In the United States there are about 6000 molding shops, down
from a high of about 8000 in the early 1990s. Consolidation continues as monitored
by the trade journals, for which web sites are listed (4–8). There are an estimated
80,000 injection presses in operation within the United States.

The injection-molding process provides low cost fabrication of large and small

parts to precision tolerances. Many of these parts are nearly impossible to make
in production volumes by other techniques. Injection molding can form parts as
small as a cubic millimeter (micromolding), and parts as large as garbage dump-
sters and 1.2

× 2.4 m (4 × 8 ft) filter panels are in production. A wide range of

1

Encyclopedia of Polymer Science and Technology. Copyright John Wiley & Sons, Inc. All rights reserved.

2

INJECTION MOLDING

Vol. 3

injection-molding processing capability characterizes the industry. Quality and
tolerances can be high if scientific principles are followed. Tolerances near
0.025 mm (0.001 in.) are difficult to achieve but are possible if concurrent engi-
neering is applied to certain parts. One of the strengths of injection molding over
other plastic forming processes is that parts can have fine three-dimensional (3-D)
details in their shape and surface finish. Typical nominal-wall thickness ranges
from 0.5 to 6.4 mm (0.020–0.250 in.); however, thickness of 0.08–50 mm (0.003–2.0
in.) are also possible. Injection molding is the plastic processing method of choice
for large quantities of identical parts. Quantities in the millions are possible with
one or more molds that can form several parts at one time. The efficiency of this
process has changed the quality of life for most Americans, especially low income
households.

Properties of injection-molded parts can be unique. Given a metal or plastic’s

typical properties, injection molding imparts some of its own. Because of flow
patterns, molecular orientation, and other factors, molded part properties are
often different than those of the base material. An example of the benefits of
orientation is the “living hinge”, the flexible plastic hinge found on spice, shampoo,
and dental floss containers. Properties of injection-molded parts can be flexible or
stiff, soft or hard, tough or brittle, clear or opaque, specific chemically resistant
and flame retardant. Parts can be tinted, colored, plated with various metals, and
recycled.

Economic Aspects

Profit margins for the industry are generally low and have averaged around 6%
as shown in Figure 1.

Another perspective to Figure 1 is to exclude resin cost and calculate the cost

to run an injection-molding machine on an hourly basis. Usually this is expressed
as a charge per hour for a given size machine. The range for hourly rates is US
$10–$175/h, depending on machine size ranked in tons of clamp pressure. Machine
clamp pressures range from 69 to 1.4

× 10

5

MPa (5–10,000 tons), most being below

1.4

× 10

4

MPa (1000 tons).

Labor

11%

Sales & Admin

15%

PBT

6%

Overhead

26%

Plastic

42%

Fig. 1.

Costs and profit margins for molders. Source: Facts & Figures of the US Plastics

Industry, published yearly by The Society of the Plastics Industry, Inc. Includes custom
and proprietary molders; PBT

= profits before taxes.

Vol. 3

INJECTION MOLDING

3

$10,400

Cycle time

(s)

20.0

22.5

17.5

$80,000

$89,600

$0

$20,000

$40,000

$60,000

$80,000

$100,000

$20,000

$29,600

$70,400

Mfg. costs

Profits

Fig. 2.

Quoted profit per $100,000 of sales. Courtesy of D. Paulson.

Most businesses operating injection-molding shops (molders) will bid a job

to make 20%, or greater, profit before taxes, yet they realize far less as Figure 1
shows. Resin cost is often included in part quotes, which means that if the price
of the plastic increases the molder must absorb the cost, and profits decrease.
Molders rarely have much of an influence on resin pricing. Quoting, therefore,
should provide a mechanism that allows resin price fluctuations to be passed on
to the customer, yet this is not common practice in the industry. Another reason
for low profit margins is that when quoting, the exact cycle time (the time required
to make a part) is not known or cannot easily calculated; that is, the quote is based
on an estimated cycle time. Often after the mold is built and trialed, production
cycles are longer than estimated or quoted and the molder cannot renegotiate the
contract. This effect of cycle time on profit is known, yet poorly understood (9).
The top bar of Figure 2 shows a typical bidding situation with a 20% return added
to costs for a 20-s cycle.

The middle bar shows how a 10–15% cycle time increase, averaged to 22.5 s,

can cut profits by nearly 50%. Using the same concept, a 10–15% decrease in cycle
time improves profits by

∼50%, for a 2.5-s decrease in cycle time. Few molders

know their actual costs and most are not willing to spend money to shorten cycles.

This low profit margin is the cause for many consolidations and hurts a

molder’s ability to attract capital for machine replacement as new technology
emerges. Figure 1 also points out that molders only sell time that is used to
make acceptable parts. Any time not used to make acceptable parts such as mold
changes, downtime, repairs, and time to make rejects is lost profit opportunity
and must be kept to a minimum to be internationally competitive.

The Injection Molding Process and Machine

Injection molding is a forming process. Material (plastic, metal, ceramic, wax, etc)
is fed into a hopper which delivers it to the feed section of the barrel and screw.
The material is melted usually via a screw that melts or blends the material and
then pushes liquefied material (eg, plastic) into the mold, which forms the part.

4

INJECTION MOLDING

Vol. 3

The injection-molding industry is relatively young when compared to other

manufacturing processes, such as metals, wood, or cement. Although patented
in 1870 by Smith and Locke and 1872 by Hyatt, the first commercial plunger
machines were developed in the late 1920s and 1930s. Egan patented the recip-
rocating screw in 1956. Since then the elements of the machine have stayed the
same but advances continue with the evolution of advanced computer controls, hy-
draulic circuits, and computer numerically controlled (CNC) all electric presses.
There are several variations and extensions of injection molding within the indus-
try that provide unique capabilities to the process and in turn special properties
to parts. Virtually all share the common elements of the following:

(1) Material preparation: The plastic/metal/mixture may be cleaned, dried, col-

ored, blended, heated, cooled, or in some way readied for use in the ma-
chine. This can be one resin, thermoplastic or thermoset, or combination of
base resin and additives. Additives include colors, metal particles, foaming
agents, antistatic agents, fillers, fibers, flow aids, stabilizers, antioxidants,
mold-release agents, binders, flame retardants, etc.

(2) Material, usually dried plastic granules, is fed (usually by gravity) into a

feed port or throat of a heating cylinder or barrel.

(3) Material melting and/or mixing, [usually thermoplastics via heat (heater

bands) and mechanical shear (flights of a screw shearing the plastic at
inside surface of a barrel wall)], preparing it to be pushed into the cavity:
As the screw rotates, it pumps plastic forward to prepare enough material
for injection. The injection unit, barrel, and screw are now something like
a syringe ready to inject fluid.

(4) Filling the cavity by pushing the material under pressure [7 to

∼414 MPa

(1000–60,000 psi)] into a mold cavity. The cavity sees less pressure, between
1.4 and 140 MPa (200–20,000 psi), because of large pressure losses as the
plastic travels the path to the part. This path includes the nozzle of the
injection-molding machine, the sprue (a tapered cone) that connects
the nozzle to the runner, the runner (usually a round channel), and the
gate or entryway to the part.

(5) The mold or tool that contains the cavity that forms the part and provides

heat to cure thermoset parts or cooling to set up or freeze thermoplastic
parts. Typically, the mold cools the molten plastic to a rigid or semirigid
form so that it can withstand the force of ejection (part removal) and retain
its shape. Not all of the energy to melt the plastic is removed by the mold.
Cooling continues after ejection and the part continues to shrink. Parts
made with certain plastics, ie, semicrystalline, may take 3 days to 6 weeks
to stabilize. Post-molding conditioning can be critical to achieve desired
performance, dimensional criteria, or flatness.

(6) The clamp which holds the mold halves together during filling and packing

the part with plastic.

(7) Part removal or an ejection mechanism. This occurs after the clamp opens,

separating the mold at the parting line into halves. The part is pushed
(ejected) out of the mold and drops to a box, conveyor belt, or is taken out
by a robot.

Vol. 3

INJECTION MOLDING

5

Rear platen

Clamp

Injection Unit

Tiebar

Moving
platen

Toggle

Stationary platen

Feed throat

Barrel

Check ring

Hopper

Injection hydraulic
cylinder

Screw rotate
motor

Controller

Frame

Nozzle tip

Toggle clamp
locking cylinder

Screw

Injection unit
pull-in cyclinder

Mold

Fig. 3.

Basic hydraulic (toggle clamp) injection-molding machine components. Illustration

by John W. Bozzelli & Rick J. Bujanowski.

Fig. 4.

Toggle and hydraulic clamps. Illustrations by Rick J. Bujanowski & John W.

Bozzelli.

(8) A controller, usually a computer, that coordinates and controls the various

steps of the process and components of the machine.

Figure 3 depicts many of these components in a typical hydraulic injection-

molding machine with a toggle clamp.

There are a number of variations in machine construction and processing

techniques (10). There are hydraulic (Fig. 4), hydromechanical, and toggle clamp-
ing machines (Fig. 4). There are also tiebar-less clamps as shown in Figure 5,
and two-platen machines that were developed to save floor space over the normal
three-platen machine.

Two-platen machines do away with the rear platen by locking the moving

platen to the four tiebars. Here the moving platen closes on the mold as usual,
then a tiebar locking mechanism locks onto the four tiebars, allowing a hydraulic
pancake cylinder to build clamp pressure by pushing the moving platen toward the
stationary platen. Clamps shown in Figures 3–5 are horizontal clamps; vertical

6

INJECTION MOLDING

Vol. 3

Fig. 5.

Tiebar-less clamp. Courtesy of Engel Machinery Inc.

clamps are also available, where a platen moves up and down. Vertical clamps
are useful in insert molding to use gravity to hold the insert in place as the mold
closes. In all cases except in tiebar-less machines, the tiebars are stretched to build
clamp pressure (tonnage) that holds the mold halves together during injection and
packing of the plastic into the mold.

A significant shift in the industry, which started in the 1990s, is to use electric

motor servo-drives rather than hydraulics to power machines. Greater accuracy,
less noise, 20–60% less power requirements, and no hydraulic fluids are claimed
advantages.

The process for making a part is called the molding cycle. This cycle is de-

picted in Figure 6 with the average time for that function stated as a percentage
of the total cycle time.

A cycle or cycle time is the elapsed time required for the injection-molding

machine to complete the process of forming a part. It can be timed starting at
any point of the cycle to the same point on the next cycle. Cycle times range from
below 4 s for compact discs to hours for jet canopies, both made from polycarbonate.
Machines have been made that can cycle in less than one second.

Fig. 6.

The injection molding cycle.

Vol. 3

INJECTION MOLDING

7

Variations and Extension to the Injection-Molding Process

Injection Blow Molding.

A preform (this looks like a test tube with bot-

tle cap threads) is injection molded in one cavity, removed and then placed into
another where it is pressurized with gas to stretch the hot preform into a thinner-
walled seamless bottle or container such as a milk bottle or gas tank. This is de-
picted in Figure 7. This is an extension of injection molding more than a variation.

Injection Compression/Coining.

With this technique the mold is only

partially closed during injection. At the appropriate time and with the right
amount of plastic in the mold, the clamp is then completely closed, forcing (com-
pressing) the plastic to the shape of the mold cavity. A variation on this is coining.
The clamp is closed but the mold has components that compress the plastic in
the cavity as the plastic cools. Coining is where the cavity volume is changing
during the solidification of the plastic. Plastic is injected into the cavity and then
the movable platen closes completely, or a mold component moves to compress the
plastic to compensate for shrinkage or densification.

Gas-Assist Injection Molding.

Here, plastic is injected into the cavity

until it is 50–85% full, then high pressure gas, usually nitrogen, is injected to finish
filling the cavity by pushing the plastic flow front to the end of the cavity. This
leaves a gas bubble or channel inside the part. This saves plastic, reduces cost,
and often improves part strength especially in thick sections. Gas can be injected
at the nozzle of the machine or directly into the mold as depicted in Figure 8.
Gas-assist molding can be considered as a variation of co-injection molding where
the outer layer or skin of the part is plastic and the core is a gas channel rather
than another type of plastic.

Gas Pack or External Gas-Assist Injection Molding.

This is a varia-

tion of coining but uses a gas, usually nitrogen, to pack out a certain section of a
part to meet dimensional requirements. The gas is injected on the rear (ejector)
side of the part to push the molten plastic against the cosmetic side of the mold
surface.

Air

Core Rod

Direction of Rotation

Injection Mold

3. Bottle Ejection

Station

Blow Mold

2. Blow Mold Station

1. Parison Station

Fig. 7.

Injection blow molding.

8

INJECTION MOLDING

Vol. 3

N

2

gas inlet into nozzle

N

2

gas inlet into mold

Fig. 8.

Gas-assist injection molding (nozzle & mold gas entry with bubble in part). Illus-

tration by Rick J. Bujanowski & John W. Bozzelli.

Gas-Counter Pressure Injection Molding.

Normally, as plastic enters

a mold it displaces the air within the cavity. The air is vented to the atmosphere via
specially designed vents specific to the type of plastic being molded. In gas-counter
pressure molding an O-ring is installed around the perimeter of the cavity and
a gas pressurizes the cavity. During injection this counter pressure gas, usually
between 345 and 2760 kPa (50–400 psi), prevents gases, water, nitrogen, etc from
emerging to the surface. Gas-counter pressure molding will provide a smooth skin
and nice appearance when processing plastic with a foaming agent. The counter
pressure prevents the gas bubbles from developing at the surface.

Structural Foam Molding.

Plastic pellets are blended with a chemical

blowing (foaming) agent, usually in pellet form, and injected into the mold under
normal molding conditions. The foaming agent can also be a gas or liquid. The cav-
ity is filled only to 70–95% full and then the foaming agent releases a gas, usually
nitrogen, to finish fill. The parts surface is usually splayed or rough. The resulting
part has a significant density reduction, saving material costs. Density reduction
ranges from 5 to 35% with an average in the 8–20% range, yet the parts are stiffer
than solid wall parts. In combination with gas-counter pressure the rough surface
caused by the gas bubbles coming to the surface can be eliminated. The counter
pressure gas prevents the foam bubbles from developing at the surface; cosmet-
ics are as good as regular injection molding. Injection and cavity pressures are
lower than that with conventional injection molding and sometimes the term low
pressure molding is applied. Machine and mold costs are lower due to the lower
pressure requirements. Trexel Inc. has recently brought out a new variation to
foam molding, the Mucel

®

Process. It injects high pressure gas (N

2

or CO

2

) into the

barrel achieving supercritical fluid conditions to dissolve the gas into the polymer.
A special screw aids in dispersing the gas into the molten plastic before injection.
Once in the cavity, small bubbles evolve as pressure is lowered as the part cools.

Thin-Wall Molding.

This is identical in concept to normal injection mold-

ing but the parts produced have much thinner nominal walls; notebook computer

Vol. 3

INJECTION MOLDING

9

shells and cellular phone housings are examples. While an exact definition of “thin-
wall” has not yet been defined, the range is often in the 0.08–0.80 mm (0.003–0.030
in.) range. Sometimes thin-wall is defined as a wall thickness to flow length ratio
in the range of 150–300 to 1. Modified injection-molding machines that provide
fast injection rates and higher injection pressures are usually required.

Insert Injection Molding.

Plastic can be injected around another material

or another plastic to form a part with plastic partially or completely encapsulating
the other material or “insert.” A metal grid, screw, or electronic circuit can be
encapsulated by placing the item into a cavity, the mold closing around it forming
the cavity and holding the insert in place and then injecting plastic. The plastic
connectors on the end of a wire harness, automotive battery tops with terminals,
and car windows with molded weather-stripping gasket are insert molded.

In-Mold Decorating.

This is a variation of insert injection molding where

the insert is a label, fabric, or some type of appliqu´e that is mounted flush to one
side of a mold and held in place, sometimes by vacuum, and plastic is injected
behind it. Wood-grain car dashboard pieces, labels on food containers, and fabric
on seats are examples of in-mold decorating.

Multishot Injection Molding.

This is similar to insert molding where the

insert is usually another plastic component. A soft grip over a rigid plastic handle,
many toothbrush handles with soft and hard sections, and an automotive tail-light
with clear, red, and orange lenses are typical examples. State-of-the-art multishot
molding uses an injection press fitted with two to six injection barrels. They can
inject different colors or different materials simultaneously or through a timed
sequence. Normally, the mold pivots or rotates between shots. The first part or
inner section is made and then the second plastic is shot around or through the
first.

Co-injection Molding.

This type is similar to multishot molding. Using

a timed or position sequenced injection, two or more plastics in different barrels
can provide a skin or outside wall made of an expensive plastic with the core or
inside filled with a foam, gas, or recycled or cheaper plastic. Fence rails or railroad
ties can be made with a post-consumer or recycled plastic as the core on the inside
and virgin prime material as the skin on the outside for acceptable appearance
and performance.

Metal/Ceramic/Carbide Injection Molding or Powder Injection Mold-

ing (PIM).

This is similar to plastic injection molding but the plastic resin is

replaced with metal, a plastic–metal blend, or ceramic powder. For example, a
plastic binder, often polypropylene or acetal, encapsulates or is used as a carrier
for fine metal particles. Loading can be near 80% metal. This plastic–metal com-
bination is forced under high pressure into the mold cavity. The “green” fragile
part is ejected and then the plastic or binder is removed with acid, solvents, water,
catalytic vapor, or heat. The resulting part is placed in a sintering oven and a pre-
cisely controlled heating regimen sinters the metal particles together yielding a
metal part. Metal molding is an excellent technique to form complex metal shapes
that would be difficult or impossible to machine. Ceramic molding is similar. Pure
metal molding is also practiced. Magnesium in granular or powder form can be
fed to a screw, melted, and injected into a mold.

Reaction Injection Molding.

Two or more components are fed into the

barrel feed throat, often preblended; the screw continues the mixing and then

10

INJECTION MOLDING

Vol. 3

injects the mixture into a mold. The mold can be heated or cooled. Once mixing
is started a chemical reaction begins. Careful control of barrel and mold temper-
atures as well as residence time in the barrel are critical to prevent premature
curing. It is possible to have the compound cure and seize the screw within the
barrel.

Liquid Injection Molding.

Molding machines with special barrels,

screws, and feed systems mix and then inject liquids into a mold cavity. Simi-
lar to reaction injection molding the two liquids, often silicone or urethane based,
react to form a thermoset part. Baby bottle nipples and keycap return springs are
two current silicone type applications.

Micromolding.

This is an extension of injection molding to extremely

small part molding. Part weights in micromolding are already below 1.0 g. Tech-
nology is evolving to injection-mold parts in the milligram (or 10

− 3

g) range.

Components of a Successful Injection-Molding Product

A successful plastic product depends on the optimization of each of the follow-
ing four principal components: part design, material selection and handling, tool
design and construction, and processing/machine capabilities.

These interact with each other and they are equally important. One or more

is often neglected by most in the molding industry, which causes delays in pro-
duction and significant profit loss throughout the entire life cycle of the product.
To employ all four from the beginning of a project is “concurrent” engineering
and their implementation is scientific injection molding (SIM). SIM is a detailed
strategy of optimizing each one of these components and not allowing errors to be
compensated through processing. The synergy of working with all four from the
beginning of a project has shown profit increases of 100% and faster development
times to market and significantly fewer manufacturing problems.

Part Design.

A successful plastic product must begin with a good part de-

sign. Because injection molding is a comparatively new and a rapidly expanding
industry, it is often the case that an engineer with little plastics training is forced
into plastic part design. This combination of no plastics experience and a tendency
to apply rules for metals to complex shapes is ill-advised and often leads to project
delays, failures, cost overruns, and production problems. Issues such as draft an-
gles, weldlines, and polish in the direction of draw are singular to plastics. Plastics
are unique, different than the materials most designers are familiar with and do
not have a uniform set of precise design rules. A common mistake for the uniniti-
ated designer is to increase nominal wall thickness to gain strength. With plastics,
generally thicker means weaker due to more internal stresses. Thicker walls add
weight and higher production cost because of longer cycle times. To gain stiffness
or strength in a plastic wall, properly designed reinforcing ribs can be added.

Different plastics have different amount of shrink upon cooling, higher for

semicrystalline resins (eg, polypropylene can shrink 0.030 in./in.) and lower for
amorphous plastics (eg, polystyrene can shrink 0.006 in./in.). Further, shrink-
age, especially for semicrystalline materials, varies due to thickness, cooling rate,
and often color. Whatever stays hotter for a longer time will shrink more. There-
fore, changes in nominal wall thickness cause differential cooling, which causes

Vol. 3

INJECTION MOLDING

11

Poor design

Better design

Void

Stress

Sinks

Fig. 9.

Cooling and nominal wall thickness. Courtesy of Glenn Beall.

differential shrinkage that causes warp. Warp distorts the part out of dimensional
tolerance. This distortion may be immediate after ejection from the mold, or after
assembly in the application environment. For example, the fit and shape of an
interior trim piece of a car’s dashboard are fine in production and assembly, but
once on the road and after a few thermal cycles (cool nights and hot days) the
trim piece warps out of shape. With plastics, nominal wall-thickness changes and
sharp corners should be avoided. If nominal wall-thickness changes are necessary
the variation must be minimized. How fast the wall changes thickness will also
influence properties and performance. Blending different thickness wall sections
will minimize internal stresses. The first rule in plastic part design is uniform
nominal wall thickness, yet few parts are designed on this basic premise. Figure 9
shows an application of uniform nominal wall thickness.

During part design one must take into consideration material, tool, and pro-

cessing issues. When designing a part the mold is also being designed. For exam-
ple, to have a dimensionally stable, high performance part it must be cooled as
uniformly as possible. Is it possible to get water channels to all sections of the
part so that it will cool evenly? Can the part be removed from the mold? Part de-
sign forces these and other tool construction details. Often the gate or entry point
for the resin is determined by the cosmetic or performance requirements; how-
ever, it may not be possible because of mold building requirements. A good plastic
part designer knows and comprehends the other three components and makes the
difficult compromises to meet material, tooling, and processing requirements.

Schools and training programs have not kept up with demand for plastic

part designers because of industry growth. Lack of proper training and drive for
short product development times are two of the many causes for “engineering
changes” as the mold is being built. That is, the design is changed after work
has started on making the mold. Often a design change is made and there is no
steel to accommodate the change. The mold builder cuts it to the previous design.
Figures 10 and 11 depict the cost of engineering changes at different stages of a
project and the impact design has on the profit margin of a project. Engineering
changes must be completed in the design phase before production tooling is made.
Unfortunately, the industry norm is to start cutting the tool (mold) before the
design is finalized.

Industry trends will continue to challenge design. Trends to consolidate

multiple functions into few parts for easier and faster assembly results in more

12

INJECTION MOLDING

Vol. 3

$10,000

$10,000

$8,000

$6,000

$4,000

$2,000

$0

$1,000

$100

$10

$1

At CAD

Design Ck

D. Consult

Pre Prod.

Production

Fig. 10.

Costs for design changes. Courtesy of Glenn Beall, data from Martin Marietta

Co.

Design

Material

Labour

Overhead

70

50

5

15

15

5

10

30

90

80

70

60

50

40

30

20

10

0

% Influence on profit

% of project cost

Fig. 11.

Costs vs influence. Courtesy of Glenn Beall, data from Hewlett Packard/Ford

Motor Co.

complex plastic part geometries. Thinner walls to save plastic and longer flow
paths in larger parts are other trends that test design limits. It is imperative that
part design heed fundamental design principles as outlined in References (11)
and (12). Better performance is often tied to tighter tolerances. This forces more
attention to the details in each of the four components for making a successful
plastic part.

Material Selection and Handling.

Material selection is usually done dur-

ing early stages of part design. Involvement of processing and material specialists
are valuable in optimizing this tough decision-making process. With data bases
covering over 32,000 resins the task is daunting (13), especially when most of the
data provided does a poor job of predicting the time and temperature behavior
of plastic materials in end applications. Sepe’s handbook provides practical time
temperature data for 120 families of thermoplastic materials based on dynamic

Vol. 3

INJECTION MOLDING

13

mechanical analysis (14). Although most of the 32,000 resins are not used in large
quantities many have specific properties and applications. During material selec-
tion the rush to production often takes precedence over the necessary thorough
determination of the applications’ physical requirements and exposure to envi-
ronmental factors.

Thorough review of the performance criteria for the application is critical.

Before a plastic is selected a list of criteria should be built on answers to questions
such as transparency, use temperature, thermal cycle, dimensional tolerances,
cosmetic aesthetics, agency approval UL, FDA, NSF etc, and wear and abrasion
needs. Environmental conditions are crucial and must be evaluated under time,
temperature, concentration of the chemical or radiation, and stress within and
applied to the part. Strength, fatigue, rate of loading, elongation, cleaning aids or
solvents, body oils or fluids, shipping and storage conditions, expected misuse, and
liability concerns must also be considered. For all plastic product failures, 32% is
due to environmental stress-crack resistance issues (15). Environmental concerns
for waste disposal and recycling must also be evaluated.

The process of injection molding influences the material’s properties. Parts

made by injection molding retain stress developed during injection, packing, or
cooling. Properties in the flow direction are often different than that in the cross
flow direction for anisotropic behavior. Shrinkage can be greater in the cross flow
direction, while tensile strength is stronger in the flow direction. Addition of addi-
tives often provides one benefit, such as color, mold release, radiation stabilization,
or wear resistance, but compromise another property such as impact or elongation.
Part performance is further complicated by use of temperature range, complex de-
sign features, nominal wall thickness, sharp internal corners, process variations,
etc. The end result is that nearly all the typical property data sheets of a material
supplier dictate that an injection-molded part must be tested in its end application
to verify performance.

Material handling is part of the molding process. Plastic must be properly

prepared before entry into the feed throat of an injection-molding machine. Plas-
tic granules can get contaminated with foreign material during shipping, and
handling especially in regrind operations. Fines, metal, dirt, moisture, or another
plastic must be removed before processing. Typically, the only precaution taken is
to pass the plastic over a magnet to trap ferrous metal fragments before use. For
resins that are hygroscopic it is necessary to dry them to remove the moisture be-
fore processing. Polycarbonates, nylons, urethanes, and polyesters are examples
of plastics that are sensitive to hydrolysis. Hydrolysis occurs as the polymer is
melted in the barrel of the injection-molding machine in the presence of moisture
and results in parts with poor physical properties. The parts may look acceptable
but the molecular chains are shorter due to hydrolysis, and parts will not per-
form to expectations. Once dry, the resin must be conveyed and stored dry until
processed. It is common for the industry to dry the plastic but convey and/or store it
exposed to plant air. Normally, the plastic must be processed within 30 min before
it readsorbs enough moisture to mandate redrying. Some resins especially sensi-
tive to moisture adsorption and hydrolysis present a drying challenge because of
poor dryer design, maintenance, and operation. Figure 12 shows a common layout
of a two-bed desiccant dryer. Two-, three-, and four-bed desiccants are common in
the industry.

14

INJECTION MOLDING

Vol. 3

Heater

Desiccant

in process

Hopper

containing plastic

Angle of hopper

must provide

mass flow

Reg.

Blower

Desiccant Bed

Regenerating

Filter

Process air

blower

Filter

Process air

heater

Fig. 12.

Two-bed desiccant dryer.

Poor dryer maintenance leads to inactive desiccants, air leaks in hoses,

clogged filters, defective heaters, and the inability to maintain a low dew point
of the drying air. Parts made from improperly dried or wet resins should not be
reground for reuse if the plastic is subject to hydrolysis. Degradation by hydrolysis
destroys properties and is not reversible.

Blending in additives is another common resin preparation step before feed-

ing plastic to the injection-molding machine. For most additives this is best done
and usually performed at the plastic resin producer’s facility after polymerization
and before extrusion into pellets. However, at-the-press additives are popular. The
additive can be granules, powders, or liquids, which can modify the base resins
in appearance and performance. Coloring is an example of an additive used to
change appearance. Typically, the natural resin is blended with a coloring agent.
If the resin producer adds the color the resin is “precolored” as delivered to the
molder. If the coloring is done at the molding shop it is called in-house coloring
with natural resin and color concentrate or master batch coloring. The concen-
trated colorant, usually in granular form but can be powder or liquid, is blended
with the natural resin (usually in a ratio of 20:1, ie, 20 kg of resin to 1 kg of col-
orant) and then fed into the feed throat of the molding machine. Colored plastic
parts do not peel or show scratches like painted parts, the color is throughout the
part. Colorants change not only the base resin’s appearance but also its physical
and processing properties. Different colors of the same resin will have different
properties and may require different processing conditions. This is especially true
for semicrystalline resins as colors often change the degree of crystallinity.

Tool Design and Construction.

Mold design and construction is critical

to all molding processes. Injection molding can not be accomplished without a
mold or “tool.” Building a mold is capital and time intensive (16). Tooling cost
breaks down into 10% for engineering, 15% for material, and 75% labor. Current
trends show that material and labor costs are increasing so much that low pressure
forming processes that can function with less expensive tooling are gaining favor,
such as rotomolding, blow molding, and thermoforming.

Vol. 3

INJECTION MOLDING

15

A mold performs many different functions. It can be highly complex leading

to high initial cost and a maintenance burden. However, for the plastics industry
simpler equates to elegance, lower cost, and higher production. The main functions
of a mold are to form the shape and surface finish of the part from the molten
plastic and then cool or heat the plastic to solidify or cure it so that the part can
be removed from the mold. The demand for 3-D complex parts has evolved mold
making into an art form second to none. Novel mold designs and construction
techniques allow ever more complex shapes to be made to exacting tolerances.
The mold must be ruggedly built to withstand repeated cycles, sometimes in the
millions, and enormous pressures. The typical injection mold will be pressurized
to 1.4–140 MPa (200–20,000 psi) to shape and pack a cavity full of plastic.

The simplest mold is the two-plate mold as depicted in Figure 13.
This two-plate mold forms one part with a cavity and core. The molten plastic

enters the mold through the sprue bushing which mates perfectly with the nozzle
of the molding machine. In Figure 13 the sprue bushing delivers molten plastic
directly into the part. The part is “sprue gated.” A “gate” is a restricted entry
point for the plastic to fill the part. There are several gate designs, each with its
unique advantages and disadvantages. In a multicavity tool the flow of plastic
is channeled at right angles from the sprue bushing to the different cavities by
means of a runner system as shown in Figure 14 (cold runner).

If multicavity molds must be built to meet product demand and the part

must be center-gated (eg, round gears) there are two tooling options: a three-plate
mold or a hot runner mold. A three-plate mold provides a runner path as shown
in Figure 15, which allows for center-gating multiple parts.

Since this runner system comes out of the mold as the part is ejected, it is

known as a cold runner. Ideally, the runner system provides the shortest path
possible to each part with equal flow distance and equal pressure-drop to each

Clamping plates

Ejector pin

Part, a rectangular bracket

Sprue

Sprue bushing

Locating ring

Cooling line

Cavity

Core

Leader pin

bushing

Ejector plate

Ejector bar

Fig. 13.

Basic components of a two-plate mold. Illustration by John W. Bozzelli.

16

INJECTION MOLDING

Vol. 3

Nozzle

Hot Manifold

Only parts drop out of mold on

each cycle

Hot Runner

Runner

Sprue

Parts

Sprue, runner and parts drop out

of mold on each cycle

Cold Runner

Fig. 14.

Multicavity (two cavities) Cold and hot runners. Drawing courtesy of John Klees.

Closed Position

Open Position

Core Plate

Cavity
Plate

Runner Plate

Core Plate

Cavity Plate

Runner Plate

Fig. 15.

Three-plate two-cavity mold.

part. Multicavity tools can make identical (Fig. 15) or nonidentical or dissimilar
parts. Tools that make nonidentical parts are called family molds. Parts for a
model airplane are made in one shot from a family tool and are usually packaged
attached to the runner. Making quality parts in a family tool is more difficult
than making parts in a multicavity tool with identical cavities because of flow
imbalances. Usually, plastic will not fill a small part the same way as it will fill a
large part within the same mold.

Multicavity tools can be built to make over 100 parts per cycle. These are

complex molds that can cost hundreds of thousands of dollars. The cold runner
systems in multicavity tools can be extensive, wasting plastic. The runner has to
be separated from the part, handled, discarded, or reground for reuse if possible.
An alternative to cold runners and a trend in the molding industry are hot run-
ners or runnerless molds. Hot runners serve the same purpose as cold runners, ie,

Vol. 3

INJECTION MOLDING

17

to provide a flow channel to each cavity, but unlike cold runners, they are heated
channels that keep the plastic molten during the entire cycle. The molten runner
stays in the tool between shots, always ready to deliver plastic to the cavity. These
hot runners are heated externally by surrounding the channel with an appropriate
electric heating element or internally where the electric element is inside the chan-
nel. While this concept of hot runners may seem fairly easy to accomplish, a hot
runner tool is more expensive, incurs higher maintenance costs, and often more dif-
ficult to start and run than cold runner counterparts. Runnerless mold design and
construction adds several more degrees of complexity to an already intricate task.
Mold making, especially hot runner tools, must be done by specialized mold makers
in combination with mold filling simulation analysis. Figure 14 shows a balanced
hot runner layout for two cavities. Figure 16 shows details of an internally vs
externally heated hot tips used to deliver plastic to parts in hot runner tools.

With mold costs increasing due to complexity and required fast mold build

times, the need to reliably predict plastic flow behavior has intensified. Mold filling
simulation has evolved with the industry to become a significant factor in aiding
the design and construction of molds (17). There are several software programs
available. They range from those that make simple two-dimensional 2-D calcu-
lations quickly and inexpensively to the sophisticated 3-D software suites that
can calculate pressure drops, shrinkage, warp, and cooling characteristics costing
thousands of dollars. Highly trained, experienced engineers or technologists are
needed to obtain reasonable predictions with this software.

Mold making is a complex and intricate undertaking. Skills required include

the knowledge of machining various metals, computers, CNC programming, elec-
tronics, hydraulics, metallurgy, polymer flow, polymer shrinkage, etc. Mold makers
have to make knowledgeable compromises on types of metal used to withstand cor-
rosion, pressure, temperature, heat conduction, thermal expansion, etc. How the
metal is cut, heat-treated, coated, or polished influences the quality and cost of
the mold. The mere direction of polishing strokes can influence the ability for the
plastic part to be ejected. Polish on a mold can be from a lens quality high polish

Externally Heated

Internally Heated

Insulator

Manifold

Heater

Heater

Melt Channel

Melt Channel

Nozzle

Cooling

Cooling

Locator Ring

Heater Probe

Plug

Fig. 16.

Hot runner tips. Courtesy of Husky Injection Molding Systems Ltd.

18

INJECTION MOLDING

Vol. 3

or gloss to a texture that imitates wood or leather grain. Polishing can be 50% of
the tooling cost.

The foremost challenge in tool building is fast response. How fast can the

mold be made and at the least cost? With unrelenting demand for shorter lead
times between concept to production, speed of part design and tool construction
becomes strategic. Prototype tools are sometimes made before production tooling
to test design or performance of both the part and the tool. An entire industry has
developed to service “rapid prototype tooling,” building some tools in days rather
than weeks. The technologies that are currently in use include metal casting,
laser sintering, metal spray-up, and Kirksite. As important as mold makers are to
injection molding, the United States has yet to develop an organized apprentice
program that can be compared to those of Europe and the Pacific rim. More mold
makers are needed because of industry growth, demand for faster build times,
and more complex tools, yet there is a decreasing supply of talented mold makers
within the United States. With plastic a growth industry ranked fourth in the U.S.
manufacturing community, this situation may inhibit growth and our ability to
be internationally competitive.

Processing.

For all plastic processings the goal is to optimize the pro-

cess in order to provide volumes of parts with optimum performance in the least
amount of time. This requires molding within the constraints of the plastic, tool,
machine, and part. To accomplish this goal, molding must be scientific from the
plastic’s point of view and focus on four key processing variables: plastic flow rate
or fill time, plastic pressure, plastic temperature, and plastic cooling rate and time.

These are plastic conditions, but the first consideration is that the injection

molding machine does what is required and expected. Reliability and repeatability
are one set of issues, while proper control of injection rate, molten plastic unifor-
mity, and other variables are another. Because of the number of machine makers
and variations in design and construction it is crucial to evaluate and monitor
actual machine performance. Hydraulic vs toggle clamping, electric vs hydraulic
drives, and different hydraulic architectures are some of the variations possible.
Process or machine monitoring, while not commonly used in the industry, does pro-
vide scientific data to objectively evaluate machine performance (18,19). Hydraulic
pressures, stroke position, temperatures, cavity, and nozzle plastic pressures are
common variables monitored. Figure 17 shows a typical plot of hydraulic and cav-
ity pressure curves vs time for a two-stage process. First stage fills the cavity to
99% full and the second stage finishes filling and packs the cavity to the required
pressure and then holds the plastic in the cavity until the gate cools and freezes.
After the second stage the screw rotates to plasticize or melt the plastic and charge
the front of the screw with the next shot.

Plastic Flow Rate.

Plastic viscosity changes with shear or injection rate.

Thus, injection rate (or the time to fill a cavity 99% full) determines a resin’s
viscosity while filling the cavity. Normally viscosity vs shear rate data is gathered
from a capillary rheometer and plotted on log scales to provide a straight line.
Figure 18 is a viscosity vs shear rate curve for a typical plastic developed by using
an injection-molding machine as a rheometer (20,21). This involves using the
Hagan–Poiseuille equation to derive the relationship that viscosity is proportional
to the pressure drop and time (fill time) as the plastic flows from the nozzle to the
end of fill making a 95–99% full shot (short shot).

Vol. 3

INJECTION MOLDING

19

Hydraulic pressure

R

i

, and Cavity pressure vs time

138

110

82.8

55.2

27.6

0

0

1

2

3

4

5

6

7

8

9

10

Nozzle and Cavity Pressure, Mpa

End of 1

st

stage part is 99% full, switchover

to 2

nd

stage

2

nd

stage, pack and hold

Cavity pressure near the gate

Cavity pressure near the last area to fill

Back pressure as screw rotates
and readies next shot

20.7

13.8

6.9

Time, s

Fig. 17.

Nozzle and cavity pressures vs time, good shot. Illustration by Rick J. Bujanowski

& John W. Bozzelli. To convert MPa to psi, multiply by 145.

Change in Plastic Viscosity

Fill time in seconds

Shear Rate, 1/s

Relative Viscosity, kPa

ⴢs

1,035,000

828,000

621,000

414,000

207,000

Slow

Easy

Stiff

Fast

0.5

10

2

1

0

0.5

1

1.5

2

2.5

3

0

Fig. 18.

Typical plastic flow. The change in plastic viscosity is at constant temperature.

Shear rate correlates to flow rate or fill time. To convert kPa

·s to psi·s, multiply by 0.145.

To convert kPa

·s to P, multiply by 10,000.

These data are then plotted on linear scales to dramatize (for the molder) the

viscosity change in plastics relative to shear rate. For injection-molding processors
(people who run the injection-molding machine) shear rate correlates to injection
speed or fill time. The viscosity of plastics is high at slow injection rates and
low at fast injection rates. The viscosity dependence on shear rate for plastics
mandates that the molding machine accurately controls injection rate or “fill time”
for process control. Injection speed or fill time must be controlled and constant to
within

±0.04 s shot-to-shot and run-to-run. If fill time varies, viscosity will vary;

if viscosity varies parts will vary. The flow-front shear rate is highly transient
during filling, either accelerating or decelerating as it flows into a complex 3-D
cavity during this fill time. Constant fill time is one way to keep these transient
velocities consistent shot to shot. Machines must be set up and run to maintain
a constant injection rate or fill time. Machines often have the ability to store

20

INJECTION MOLDING

Vol. 3

Velocity profile

Orientation

frozen layer

Fountain flow

High Shear

2

nd

flow

Fig. 19.

Fountain flow.

process set points such as injection speed or velocity in millimeters per second.
However, because of calibration errors and normal wear on a press, duplicating
a millimeters per second (mm/s) set point called up from computer memory is no
guarantee that fill time will be repeated. Unfortunately, the industry trend is to
duplicate machine set points, not the actual velocities. It is best to monitor and
control fill time.

Polymer flow through the nozzle, sprue, runner, gate, and part is best de-

scribed with the concept of fountain flow. This describes the movement of the
long-chain molecules. Figure 19 depicts this fountain flow concept where the long
chains freeze at the side walls (there is zero velocity at the side walls of the flow
path), orientate, and provide a velocity gradient and a high shear region just inside
the frozen outer layer.

This explains why in-mold decorating can be done with delicate fabrics or

paper labels without tearing the decoration while plastic flows over it to fill the
cavity. Powdered-metal and metal molding do not exhibit this type of flow behavior.

Historically, most discussions on polymer flow focus only on viscosity. How-

ever, with many of the newer resins that are highly impact modified or have high
elasticity, viscosity tells only part of the story for flow behavior. Two plastics can
have identical viscosity characteristics and yet mold differently. Elastic response
and the polymer’s ability to transmit pressure become increasingly important as
part designs increase in complexity. Both viscous and elastic response are required
to properly characterize flow through complex cavities (22).

Controlling fill time is so important for plastic molding that a specific mold-

ing strategy has evolved. This strategy is based on separating fill (first stage or
boost) from pack and hold (second stage). During fill, abundant hydraulic pressure
(power) must be available to the hydraulic cylinder so that the flow-control valve
on the injection-molding machine controls flow. First-stage hydraulic pressure
set-point or limit must not control injection speed. There must be a pressure dif-
ferential,

P, across the flow control valve so that the screw maintains injection

velocity during fill. This concept of filling on first stage under velocity control is
analogous to cruise-control on a car. If the plastic has to go through a gate or thin
section, there has to be plenty of energy (hydraulic pressure) available to do it, just
like a car going up a hill on cruise control. As with cruise control, if you run out of
power (hydraulic pressure in the case of molding) you slow down. If you slow down,
as the curve in Figure 18 shows, you move to the left and the plastic’s viscosity
gets stiffer. If the hydraulic injection cylinder does not have abundant hydraulic

Vol. 3

INJECTION MOLDING

21

pressure and the material gets stiffer due to a temperature change or different lot,
you have a compounding effect. The injection speed slows, which in turn drives
the viscosity stiffer and again the injection speed slows. This actually can end
up as an unfilled part or short shot. Many molders run with little or no pressure
differential across the flow-control device, preventing state-of-the-art closed loop
machines from correctly controlling the first stage or filling the part. This is called
“pressure-limited molding.” Processing with an appropriate

P, not too high to

flash the mold, allows the machine to attenuate viscosity changes on first stage.
If the machine is set up correctly, it will adapt on any given shot to compensate
for viscosity changes during filling the part. This removes the need for processors
to tweak first-stage settings during production. Also, there is no lost production
while waiting for someone to realize the machine is making unacceptable parts
because of a change in viscosity such as a lot or temperature change.

Predicting the flow through the runner system (hot or cold), the gate and

part is known as mold-filling simulation or mold-flow analysis as referenced in
the tooling discussion. All deal with the viscosity vs shear rate relationship and
continue to evolve. For example, why inside cavities on a multicavity (over four)
mold fill differently than outside cavities has recently been explained ((23)). That
is, some cavities fill first and others fill later during injection or the first stage. This
happens even though the flow distance and pressure drop to each cavity is identi-
cal. This is known as unbalanced filling, which can make nonidentical parts even
though the mold cavities are identical. It also sets up these cavities for nonuni-
form packing during the second stage. Once filling the mold is accomplished, the
cavity must be packed with molten plastic to a pressure that provides appropriate
dimensions and surface finish.

Plastic Pressure.

The pressure gradient across a part establishes most of

the part’s attributes. The plastic pressure during pack and hold (second stage)
influences dimensions, warp, gate area strength, gate vestige, and some visual
characteristics such as sink marks and gloss. There is a pressure drop along the
flow path from the plastic within the nozzle to the part’s last area to fill. This pres-
sure drop provides a measure of the difficulty of molding. The larger the pressure
drop, the larger the amplification of the process variables. Determining this pres-
sure drop is not straightforward, as different machines develop different plastic
pressures at the same hydraulic pressures. By knowing a machine’s intensification
ratio (hydraulic ram area vs the area of the nonreturn valve acting like a piston
pushing the plastic) we can calculate the plastic pressure in the nozzle. Often this
plastic pressure is noted as injection pressure or specific pressure on a machine’s
specification sheet. Figure 20 depicts the hydraulic advantage or intensification
ratio of hydraulic pressure intensified to plastic pressure in the nozzle.

During the early stages of the industry the machine producers made nearly

all machines with a 10:1 intensification ratio with maximum injection pressures
∼140 MPa (20,000 psi). If the maximum hydraulic pressure allowed to the hy-
draulic injection cylinder was 14 MPa (2000 psi), with an intensification ratio of
10:1, the maximum molten plastic pressure available in the nozzle was 140 MPa
(20,000 psi). Today, machines are available with intensification ratios from 6:1 to
43:1 and corresponding plastic pressures up to

∼415 MPa (60,000 psi). Figure 20

depicts a hydraulic cylinder and screw with a 16:1 intensification ratio.

This intensification ratio and resulting maximum available injection pres-

sure were a factor in U.S. machine producers losing a significant portion of the

22

INJECTION MOLDING

Vol. 3

Melt Pressure

221 MPa

(32,000 psi)

Plastic variable

Hydraulic Pressure

13.8 MPa (2000 psi)

Machine variable

Pump

An Intensifier

R

i

 16:1

Screw diameter

d

 25.4 mm (1.00 in.); and hydraulic cylinder D  101.6 mm (4.0 in.)

R

i

 (␲d

2

/4)/(

␲d

2

/4)

The intensification ratio (

R

i

) is 102

2

/25.4

2

 16:1

Fig. 20.

Hydraulic vs plastic pressure. Illustration by John W. Bozzelli & Rick J. Bu-

janowski.

market to foreign competitors. The United States was slow to respond to the need
for higher injection pressures required for long flow paths and complex thin-wall
parts. It is suggested that machines purchased have at least 207 MPa (30,000 psi)
of injection pressure, higher if the intention is for thin-wall molding. This will
provide machines with greater flexibility in precision molding at no extra cost.
They will also have better injection velocity control.

Over 90% of the hydraulic machine producers currently do not report plastic

pressure on the controller or gauges. Hydraulic pressure is the standard. Produc-
tion scheduling often forces a mold to be run in several different presses, each
with a potentially different intensification ratio. Processors need to control this
plastic pressure variable for consistent parts. Many molders reproduce hydraulic
pressures (a machine variable) and not the critically important molten plastic
pressures (the key plastic variable). With the advent of all electric machines this
issue may resolve itself. Electric injection-molding machines use plastic pressure
values for set points and reported pressures on their controllers. Injection pres-
sure is sensed through a transducer mounted on the thrust bearing pushing on
the screw, and it must read in plastic or injection pressure.

Knowing and controlling both injection and cavity pressures during filling

(first stage) and packing (second stage) provides for excellent process control and
part consistency. In the first stage (filling) the goal is to make a 99% full part before
switching or transferring to the second stage (pack and hold). The part is slightly
short or unfilled if the process was stopped at the end of the first stage. Since the
part is short, the pressure near the last area to fill or the end of the flow front is
zero. The pressure at the beginning of flow inside the nozzle can be calculated from
the hydraulic pressure at the end of the first stage and using the intensification
ratio. Thus, the plastic-pressure drop can be calculated for any short shot. By a
series of short shots the pressure loss during the first stage (or filling the part) can
be detailed without the use of cavity transducers. This data can be compared to
mold-filling simulation calculations to check accuracy. A typical short shot study
might show pressure losses as follows.

Vol. 3

INJECTION MOLDING

23

Nozzle

20.6 MPa

3,000 psi

Hot runner and gate

38.0 MPa

5,500 psi

Filling the part

82.7 MPa

12,000 psi

Total

141.3 MPa

20,500 psi

This information allows the processor to find high pressure-loss areas, which

can be modified to provide easier flow and therefore a more robust molding process
that makes acceptable parts. Lower pressure drop leads to better process control
and consistency.

Switching from the first stage or velocity (V) control to the second stage or

pressure (P) control is known as first to second stage transfer, or VP transfer. This
switchover occurs as the screw moves forward and reaches a set point or position
that provides the right amount of material to fill the mold 95–99.9% full. Other
possible switchover conditions can be based on time, hydraulic pressure, or cav-
ity pressure. Switchover by cavity pressure is more repeatable than by position,
which is more repeatable than time. Switchover by hydraulic pressure is not rec-
ommended, as it limits the machine’s ability to control velocity. To properly control
the development of cavity pressure in the mold requires sharp machine response
from the first to the second stage and accurate pressure control (see Figure 17,
nozzle pressure curve).

The purpose of the second stage is to finish filling the mold and build cavity

pressure to sufficient levels to replicate the shape and finish of the mold cav-
ity. This is a different pressure loss than seen in filling or the first stage. Too
much packing pressure will result in parts that are overpacked or flashed, causing
wasted plastic and dimensional variances. Too little pressure or lack of control will
produce nonfilled parts (shorts), dimensional variances, and unacceptable parts.
Cavity pressure measurements (transducers are required) provide a methodol-
ogy to verify part consistency. While cavity pressure measurement capability has
been available for

∼25 years, it is rarely used in the molding industry. Costs,

incorrect placement, lack of large data gathering and storage capability, lack of
training, delicate transducers (pressure sensors), and difficulty in interfacing with
the molding machine have all been negative factors. Focus had only been on peak
cavity pressures. More complete data can be found with cavity pressure curve
integrals or area under the cavity pressure curves. For example, gate seal or plas-
tic backflow out of the cavity is not seen if only peak pressures are monitored.
Process monitoring instrumentation and appropriately located transducers can
be used to inspect parts for 100% acceptance or rejection for dimensional toler-
ances. Cavity pressure measurements will not inspect for cosmetic or aesthetic
criteria, which can be done via computer vision systems. Newer easier-to-use sen-
sors that measure both cavity pressure and temperature are under development
based on ultrasonics (24,25).

Before the amount of pressure to pack the part to the specified tolerances

can be found, second stage (pack and hold) time must be determined. Second
stage pressure is a timed event and must be on either (a) long enough to make
sure the gate is frozen so plastic does not backflow out of the part or (b) for just
long enough to pack the part out but not overpack the gate. Backflow of plastic out
of the part into the runner can occur if second stage pressure timer ends before

24

INJECTION MOLDING

Vol. 3

the gate freezes. This allows the higher pressure that has been developed in the
cavity to push plastic back out the gate so that the gate area of the part does not
have molecules too closely packed or overpacked. Once the mold is filled, plastic
cools and shrinkage requires more plastic to compensate. The amount of plastic
that can be packed into the cavity during the second stage time depends on how
long it takes for the gate to seal or freeze off. Gate freeze or gate seal time can be
determined with a cavity pressure sensor near the gate or by plotting part weight
vs increasing second stage (pack and hold) time. Cavity pressure curves from
measurements near the gate will show changes in slope on the cooling portion
(decreasing pressure vs time) of the curve between gate seal and gate unseal.
However, cavity pressure data is rarely available. Part weight experiments are
more common and require weighing parts, excluding the runners, as the processor
changes the amount of second stage time. Part weight vs second stage time are
recorded and graphed. The resulting graph shows stable part weights when the
gate is sealed and a drop in weight at second stage times where the gate is not
frozen. Part weight will vary as second stage time is changed until gate freeze
occurs. Cycle time must be kept constant during this experiment to keep the
energy load on the mold consistent.

Once gate seal time is established, the processor can optimize the amount of

second stage pressure. It is this pack and hold pressure that establishes the parts
dimensions and density. If running with second stage time long enough achieves
gate seal, raising second stage pressure will increase part weight and size. Lower
pressures produce lighter and smaller parts. A well-made mold that correctly
accommodates for shrinkage (difficult to do) and the correct packing pressure
will provide a part centered on the dimensional specifications. This provides an
optimized process to meet dimensional criteria. The gate area will normally be
packed to a higher pressure than the last area to fill. There is a pressure gradient
from the gate area to the last area to fill. Within the part there is some pressure
loss which can vary considerably on a shot-to-shot basis. The pressure gradient
must be duplicated for consistent parts. This pressure gradient can be minimized
if the processor runs with second stage times shorter than those needed for gate
seal. Plastic can be allowed to backflow out of the cavity to reduce pressure near
the gate after filling. These cavity pressures can be measured with cavity pressure
transducers (see Fig. 17, the cavity pressure curves).

Correctly adjusting screw rotation speed, backpressure, clamp opening, ejec-

tion, and clamp closing will optimize the rest of the cycle. During screw rotation,
there is pressure held against the hydraulic cylinder in the injection unit. This
resistance pressure is called backpressure and must be overcome by the metering
section of the screw acting like a pump pushing plastic in front of it, preparing for
the next shot. Something like filling a syringe for a shot. With a properly designed
screw, minimum screw rotation speeds to keep the cycle constant, and backpres-
sures

∼7.0 MPa (1000 psi) plastic pressure, uniform molten plastic consistency

can be achieved. Clamp movement should be smooth and set appropriately to pro-
tect damaging the expensive mold. Ejection should not distort the part with too
much speed or pressure.

The clamp keeps the mold halves together as cavity pressure develops dur-

ing filling and second stage packing. Clamp force is generated by the stationary
and rear platen stretching the four tiebars after the moving platen pushes the
mold closed. It is important that the tool takes up most of platen area between the

Vol. 3

INJECTION MOLDING

25

tiebars and that the clamping force is uniform. Uniform clamp force around the
parting line prevents excessive tool and clamp wear. Clamping a mold correctly
requires a level machine, parallel platens, and uniform stretch of each of the four
tiebars. Occasionally, high injection pressures will overcome the clamping force,
separate the mold halves, then produce flashed parts. These must be trimmed or
discarded. Clamping forces required are usually estimated and rarely scientifi-
cally calculated or measured. Industry guidelines vary from 20.7 to 68.9 MPa (1.5
to

∼5 tons/in.

2

) of the projected area of the part and runner. Thin-wall applications

can require as much as 110 MPa (8 tons/in.

2

).

Different types of molding machine clamps (toggles, hydraulic, and hydrome-

chanical) have different advantages. Toggles are fast and energy efficient, yet it is
difficult to set clamp tonnage accurately. Better toggle machines use strain gages
to measure stretch of the tiebars which relates clamp tonnage. With toggles, clamp
force changes as mold temperature changes due to the thermal expansion of the
mold and machine components. Hydraulic clamps are easy to set, less energy effi-
cient but are sometimes slower due to the large volume of oil necessary to fill the
hydraulic cylinder. Hydraulic clamps automatically compensate for thermal mold
expansion. Molding machines are ranked by clamp tonnage within the molding
industry. Machines range from 69 to 1.4

× 10

5

MPa (5–10,000 ton/in.

2

).

Plastic Temperature.

Correct temperature and uniformity are crucial for

a consistent process. Monitoring of plastic temperature is difficult and rarely uti-
lized in the industry. Control of temperature is done via thermocouples partially
embedded into the barrel wall, usually three to four along the length of the bar-
rel. The actual molten plastic is not monitored for temperature and

>90% of the

temperature values provided as data are barrel wall temperatures that can be off
by

±25

◦

C (45

◦

F). Temperature control is done via proportional-integral-derivative

(PID) controllers or PID algorithms on computer controlled presses. Calibration of
thermocouples is seldom done. PID temperature control of the nozzle tips is also
important, yet

∼40% of the industry uses variacs.

Heat necessary to melt the plastic granules comes from two sources. The

heater bands start the melting process by getting the plastic to stick to the barrel
wall. Between 10 and 40% of the heat necessary to melt the plastic comes from the
heater bands. Most of the heat comes from the frictional or shear heat developed
as the flights of the screw rotate within close proximity of the barrel wall. The
flights shave plastic from the barrel wall into a spiral pool between the flights.
Between 60 and 90% of the energy necessary to melt the plastic comes from screw
rotation or shear at the barrel wall—flight interface. Temperature override can
occur, and the open space between the heater bands is necessary for barrel cooling.
Some machines have cooling fans placed between the heater bands to aid control.

Screw design and construction are important aspects to any plastic process-

ing, yet design gets little attention in the injection-molding industry. Figure 21
shows a general–purpose screw which is what is specified or accepted for

∼85%

of machines delivered. The screw is divided into three zones: feed, normally 50%
of the flights; transition, 25%; and metering, 25%.

The feed zone conveys plastic forward to the transition zone and has a con-

stant root diameter. The root diameter tapers through the transition zone, which
compresses the granules forcing air back toward and out of the feed throat and
hopper. The main purpose of the transition zone is to compress the plastic and
provide the shear heating to drive the melting process. Ideally, by the end of the

26

INJECTION MOLDING

Vol. 3

Pitch

Diameter

Nonreturn valve
or check ring

Metering

25% L

Transition

25% L

Feed

50% L

Spline

Flight land

Flight depth

Flight length

Fig. 21.

Screw components. Illustration by Rick J. Bujanowski & John W. Bozzelli.

Flow

Cooling

Melt pool

Solids

Solids bed

Heater band

Hopper

Barrel wall

Fig. 22.

Melt Model, solids pass. Illustration by John W. Bozzelli & Rick J. Bujanowski.

transition zone the plastic is melted, but usually the solid bed breaks up. Unmelted
solids along with molten plastic get delivered to the metering section, which acts
like a pump to push the plastic forward, charging plastic for the next shot (see
Fig. 22).

As plastic is pushed forward (overcoming backpressure) the screw is forced to

retract. There is now a cylinder of plastic in front of the screw tip and nonreturn
valve. Upon injection, the screw moves forward seating the nonreturn valve or
check ring, which acts like a plunger, piston, or ram to push the plastic out the
nozzle tip into the sprue bushing of the mold.

Molten plastic uniformity is required for optimum processing performance;

hence, screw specification is another strategic issue. The standard general-
purpose screw is known to produce unmelted solids (26,27). These unmelted solids
clog gates and produce inferior parts. Molten plastic uniformity can be achieved
with state-of-art screws and metering sections as represented in Figure 23.

Screw and barrel specifications deal with type of material to be processed,

shot volume, and plastic uniformity. Shot volume (the part, runner, and sprue)

Metering zone

Flow

Barrier must end

in slow taper

Fig. 23.

Modified metering section. Illustration by Rick J. Bujanowski & John W. Bozzelli.

Vol. 3

INJECTION MOLDING

27

should be 25–65% of the barrel capacity. Barrel and screw hardness and corrosion
resistance depend on the type of material to be processed. Glass fiber or other
abrasive fillers require hard barrels to prevent excessive wear. Materials that may
produce acid gases, HCl from PVC, or other volatiles that may chemically attack
the barrel and screw call for more chemically resistant steels such as stainless or
plating the screw with a chemically resistant metal such as nickel. Screw design
specifications also involve compression ratio, how much is the plastic compressed
within the flights of the screw from feed to metering zone, length-to-diameter ratio
or L/D, and shape of the screw flights. The screw should be highly polished to allow
plastic to slip on the screw and stick to the barrel, have sharp flights with a large
trailing radius so as to prevent dead spots from causing degradation of the polymer.

Plastic Cooling Rate and Time.

Cooling is over 90% of the molding cycle,

so optimization is critical to provide the minimum cycle time and optimum profits.
Analyzing the cycle in Figure 6 shows that cooling for the part actually occurs in
each segment except for filling. Molding is a thermal process and so consistent
steady-state conditions cycle to cycle are crucial. Within the cycle there are large
swings in temperature, yet cycle-to-cycle consistency is paramount. This is the
reason automatic cycles provide better quality and production over manual cycles.
Automatic cycles allow the machine to cycle unattended. Manual cycle depend on
an operator to take parts out of the mold, then re-cycling the process. Manual
cycles do not run at a constant cycle time, allowing heat loads to vary on the
mold and machine. Molds take several cycles to reach equilibrium because of the
cooling process being a combination of radiation, air convection, and conduction
cooling as coolant flows through the channels of the mold. Time to reach this
steady-state mode can be minutes or hours depending on the part, tool size, and
mold-cooling capacity. Often the machine’s platens that hold the mold must come
to temperature equilibrium with the environment before the cycles are in steady
state. Thus, the machine’s environment can and does influence the process. Air
conditioning a molding plant is expensive but it aids in steady-state processing,
leveling season heat and cooling loads on the mold.

Not all of the heat energy put into the resin must be removed by mold cooling.

The part can be ejected warm as long as it is rigid enough to withstand the force
of ejection and hold its shape during post-molding handling by humans, robots,
conveyors, or as they accumulate in a box or bin. Part cooling will continue after
ejection and must be kept consistent to maintain part uniformity and prevent
warping. Whatever the cause, nonuniform cooling will add retained stress to a
part. Nonuniform nominal wall thickness, improper packaging, even inconsistent
part placement on a table after ejection can cause dimensional variations and
warp. This is particularly important for semicrystalline, high shrink resins, which
can take weeks to stabilize because of continued crystallization.

Mold cooling or temperature control is typically done with a thermolator that

pumps water, water and ethylene glycol mixture, or oil through channels in the
mold to heat or cool the plastic. Heat transfer (usually cooling) problems develop if
corrosion or deposits accumulate in the channels (water lines) in the mold. Other
issues are that water lines are not drilled uniformly around the part, coolant lines
are not routed correctly, or coolant flow is not turbulent. Flow turbulence is defined
by a Reynolds number of 3500 or greater and is difficult to achieve in some plants
because of inadequate feed and return line size or restrictions in the tools cooling

28

INJECTION MOLDING

Vol. 3

channels. Molds often have different size cooling lines yet fed by the one multifrost
manifold. Larger lines get more coolant flow at the expense of smaller lines, leading
to nonuniform cooling. Chiller (refrigerated or chilled water) operation, cooling
tower condition, water treatment, and filtration are important areas that demand
constant monitoring to maintain optimum and consistent cooling. Certain cooling
checks should be made on a routine basis. For instance, each cavity must receive
the same coolant flow rate. Additionally, the pressure drop from inlet to outlet for
each mold circuit should be

∼210 to 240 kPa (30–35 psi) pressure drop or greater.

Temperature differential between inlet vs outlet should be below 2

◦

C (4

◦

F).

Handling of the parts after production must also be controlled for consistency.

For high tolerance parts in semicrystalline resins (polyethylenes, nylons, acetal,
and polypropylene), each part must be treated to the same post-mold conditions
of handling, storage, etc. Cooling influences post-molding shrinkage. Parts in the
center of a box do not cool like those on the sides, and the different cooling rate
may be the cause of rejects. The concept is to treat every part the same. All parts
must be ejected, cooled, and packaged identically, or differences in dimensions and
warp develop even though parts were good out of the press.

BIBLIOGRAPHY

“Injection Molding” in EPSE 2nd ed., Vol. 8, pp. 102–138, by I. I. Rubin, Robinson Plastic
Corp.

1. D. V. Rosato and D. V. Rosato, Concise Encyclopedia of Plastics: Fabrication & Industry,

Kluwer Academic Publishers, Boston, Mass., 2000.

2. D. V. Rosato and D. V. Rosato, Injection Molding Handbook, 3rd ed., Kluwer Academic

Publishers, Boston, Mass., 2000.

3. D. V. Rosato, Injection Molding Newsletter (53), 1–24 (Spring 2000).
4. Injection Molding Magazine, www.immnet.com.
5. Modern Plastics, www.modplas.com.
6. Plastics News, www.PlasticsNews.com.
7. Plastics Engineering, Society of Plastics Engineers, www.4spe.org.
8. Modern Mold and Tooling, 110 William St., New York, NY 10038.
9. D. C. Paulson, Injection Molding Magazine, Dec. 1994, p. 7.

10. F. Johannaber, Injection Molding Machines, A User’s Guide, 3rd ed., Hanser Gardner

Publications, Cincinnati, Ohio, 1994, 328 pp.

11. G. Beall, Plastic Part Design for Economical Injection Molding, Seminar Manual,

1982–2000.

12. R. A. Malloy, Plastic Part Design for Injection Molding—An Introduction, Hanser

Gardner Publications, Cincinnati, Ohio, 1994, 460 pp.

13. IDES Prospector, www.idesinc.com, also available through the Society of Plastic

Engineers, www.4SPE.org.

14. M. Sepe, Dynamic Mechanical Analysis for Plastics Engineering, Plastics Design

Library, Norwich, N.Y., 1998, 189 pp.

15. M.Sepe, private communication, Dec. 1999.
16. R. G. W. Pye, Injection Mold Design, Longman Scientific and Technical, Essex, England,

1989.

17. P. Kennedy, Flow Analysis of Injection Molds, Hanser Gardner Publications, Cincin-

nati, Ohio, 1995, 237 pp.

18. D. C. Paulson, Plastics and Rubber Processing, March 1978, pp. 32–36.
19. R. J. Groleau, 1989 ANTEC, pp. 260–263.

Vol. 3

INORGANIC POLYMERS

29

20. J. W. Bozzelli and R. J. Groleau, Presented at the Las Vegas West SPE Conference, Feb.

1991.

21. J. Bozzelli, A. Larsen, and J. McDonnell, 1998 SPE ANTEC, paper number 346 on

compact disc.

22. L. R. Schmidt, 1998 ANTEC paper number 719 on compact disk #9840CD, Printed

volume pp. 3592–3597.

23. J. P. Beaumount, J. Injection Molding Technol. 1(3), 133–143 (Sept. 1997).
24. S. Y. Baek, R. Edwards, and C. L. Thomas, 2000 SPE ANTEC, paper number 1038 on

compact disk #0050CD.

25. E. C. Brown, A. J. Dawson, P. D. Coates, 2000 SPE ANTEC, paper number 592 on

compact disk # 0050CD.

26. Cylinder & Screw Handbook, 6th ed., Westland Corp., Wichita, Kans., 1997, 33 pp.
27. Plasticating Components, Spirex Corp., Youngstown, Ohio, 2000, 114 pp.

J

OHN

W. B

OZZELLI

Midland, Michigan