Article

Prediction of the Stability of the Casting Process by the HPDC

Method on the Basis of Knowledge Obtained by Data
Mining Techniques
Marcin Brzeziński 1, * , Jakub Wiśniowski 1 , Mariusz Łucarz 1 , Karolina Kaczmarska 1 , Alena Pribulová 2
and Peter Futáš 2

1 Faculty of Foundry Engineering, AGH University of Krakow, Reymonta 23 St., 30-059 Krakow, Poland;
wisniowskij@agh.edu.pl (J.W.); eumar@agh.edu.pl (M.Ł.); karolina.kaczmarska@agh.edu.pl (K.K.)
2 Faculty of Materials, Technical University of Kosice, Metallurgy and Recycling, Letna 1/9,
042 00 Kosice, Slovakia; alena.pribulova@tuke.sk (A.P.); peter.futas@tuke.sk (P.F.)
* Correspondence: mbr@agh.edu.pl

Abstract: High-pressure die casting (HPDC) of aluminum alloys is one of the most efficient manu-
facturing methods, offering high repeatability and the ability to produce highly complex castings.
The cast parts are characterized by good surface quality, high dimensional accuracy, and high tensile
strength. Continuous technological advancements are driving the increase in part complexity and
quality requirements. Numerous parameters impact the quality of a casting in the HPDC process.
The most commonly controlled parameters include plunger velocity in the first and second phases,
switching point, and intensification pressure. However, a key question arises: is there a parameter
that can predict casting quality? This article presents an exploratory analysis of data recorded in
a modern HPDC casting machine, focusing on the thickness of the biscuit. The biscuit is the first
component of the casting runner system, with a diameter equivalent to that of the injection chamber
and a height linked to various processes and mold characteristics. While its diameter is fixed, the
thickness varies. The nominal thickness value and tolerances are defined by the process designer
Citation: Brzeziński, M.; Wiśniowski, based on calculations. Although the thickness of the biscuit does not affect the casting geometry,
J.; Łucarz, M.; Kaczmarska, K.; it influences porosity and cold-shot formation. This study aimed to determine the relationship
Pribulová, A.; Futáš, P. Prediction of
between biscuit thickness and casting quality parameters, such as porosity. For this purpose, a series
the Stability of the Casting Process by
of injections was produced using automated gating, and biscuit thicknesses were examined. This
the HPDC Method on the Basis of
article presents quality assessment tools and statistical analyses demonstrating a strong correlation
Knowledge Obtained by Data Mining
between biscuit thickness and casting quality. The knowledge gained from the methodology and
Techniques. Materials 2024, 17, 5935.
https://doi.org/10.3390/
analyses developed in this study can be applied in support systems for the quality diagnostics of
ma17235935 HPDC castings.

Academic Editors: Xuefeng Tang,

Keywords: HPDC; porosity; aluminium alloys; quality; SPC
Hongchao Ji and Zhen Jia

Received: 2 November 2024

Revised: 26 November 2024
Accepted: 1 December 2024 1. Introduction
Published: 4 December 2024
Casting is one of the most efficient methods for large-scale product manufacturing [1–4]
and one of the least expensive technologies for producing metallic components or finished
elements in a broad market [4,5]. Castings are often foundational parts of machinery and as-
Copyright: © 2024 by the authors.
semblies, and serve as a starting point across many industrial fields [3]. Compared to other
Licensee MDPI, Basel, Switzerland. production methods, casting offers numerous advantages, including minimized or even
This article is an open access article eliminated machining processes required for final products, the ability to produce parts
distributed under the terms and with highly complex geometries that would otherwise require the assembly of multiple
conditions of the Creative Commons parts in alternative methods, and adaptability to mass production requirements [2,4,6]. The
Attribution (CC BY) license (https:// foundry industry is continuously advancing to meet the increasingly challenging demands
creativecommons.org/licenses/by/ of customers [2,7]. Additionally, given the extensive possibilities offered by a variety of
4.0/). metal alloys and their properties, specific branches of casting have distinct, specialized

Materials 2024, 17, 5935. https://doi.org/10.3390/ma17235935 https://www.mdpi.com/journal/materials

Materials 2024, 17, 5935 2 of 23

characteristics [8]. Consequently, both research institutions and companies are directing
more financial resources toward the discovery of new alloys and specific casting methods,
fostering new design solutions for equipment and manufacturing technologies to meet the
stringent demands of various sectors, particularly the automotive industry, which is the
most challenging one [2,6,8–10].
The automotive industry, one of the largest consumers of foundry products, is also the
most demanding client [2,4,11]. Beyond technological, quality, and safety requirements,
the industry imposes parameters related to repeatability and high production efficiency
due to its mass-production nature [6]. Each year, the automotive sector produces millions
of cast engine blocks, gearbox housings, and other automotive components. The industry
shows growth trends in the coming years [12]. To meet such demands, the industry
essentially mandates high-pressure die-casting (HPDC) technology, which is one of the
most rapidly growing sectors of the casting industry [12]. Ensuring high productivity in
this casting process requires controlling numerous factors, such as mold shape dimensions,
die coating, chemical composition, and metal temperature, with particular attention paid
to preventive actions to avoid potential defects [1,4,6,8]. HPDC castings are known for
their high geometric precision, good mechanical properties, and low cost [11,13]. They are
also widely used to produce electrical, hydraulic, and lighting components. Aluminum die
casting offers numerous advantages over other metalworking processes, such as the ability
to produce highly complex shapes that cannot be effectively achieved through extrusion
or machining [13–15]. Additional benefits include the production of textured or smooth
surfaces, casting of both large and small parts, and thin-walled components [16]. A prime
example is the production of complex automotive parts, such as gearboxes and engine
blocks [10].
The high injection speed of liquid metal and short mold filling time ensure that every
area and hard-to-reach space in the mold is filled, precisely replicating the mold’s shape.
These parameters enable the production of intricate castings, while a high intensification
pressure improves the fill quality of the products [16,17]. Die-casting products meet
stringent appearance and dimensional tolerance requirements [13] and are ideally suited for
high-quality standards. Among the many benefits of die casting, one of the most significant
is its exceptionally short casting cycle time. The short manufacturing time is especially
well-suited for the mass production of complex, lightweight components [18,19]. High
production efficiency allows for the rapid production of a large number of units, including
both compliant and non-compliant parts, according to client specifications [2,13,20]. Minor
undetected changes or incidents in the casting process can result in significant quality
issues in large quantities [21]. Such instances significantly increase foundry costs related
to sorting, scrapping, and implementing additional defect detection measures, making
high repeatability the most desirable characteristic [4,6,9,21]. Both peripheral equipment
and the casting machine, when operating correctly, have tolerances that generate “natural
process variability” [9,22,23]. A major issue arises with variables that push equipment
performance beyond this “natural variability” [21,24,25]. This necessitates detailed analyses
and identification of the causes of disturbances, which require considerable effort, time, and
data [22,25,26]. For high-volume production, preventive actions are essential [24]. Viewing
casting as a set of parameters that must be met by the foundry, the process engineer should
consider the casting process in a similar manner [27]. Casting represents the work of the
entire team, and its inspection can confirm compliance with customer requirements [21,25].
Additionally, continuous process changes disrupt production stability, contrary to the need
for repeatability [22,24,28]. To prevent this, the process designer should define the range of
the monitored process parameters, establish their tolerances, develop control plans, and
organize peripheral equipment inspections [28,29]. This approach enables an early response
to emerging variability, allowing analytical actions to influence production quality while
the process remains under control. Predicting potential changes in the production process
allows for timely intervention in response to anomalies, thereby eliminating defective
products and their causes [21,25,30].
Materials 2024, 17, 5935 3 of 23

However, like any manufacturing process, HPDC has drawbacks. The most important of
these is the potential porosity due to air entrapment in the form of very fine bubbles, which
affects the structural integrity of the casting and causes problems with subsequent mechanical
and heat treatment. This problem is particularly relevant for components that must satisfy
high-quality requirements, such as those in the aerospace and automotive industries. In
addition, porosity limits heat treatment options, as gas trapped in the pores during heating
can lead to deformation of the casting or the formation of defects. Another problem is the high
start-up and operating costs of HPDC foundries. HPDC requires sophisticated equipment
such as steel molds, casting machines, and cooling and lubrication systems. The cost of
molds (dies) for HPDC is very high, making this process cost effective, mainly for mass
production. In addition, HPDC machines are expensive to purchase and maintain. HPDC
works best for metals and alloys with low melting points and good flowability, such as
aluminum, magnesium, or zinc. Metals with high melting points, such as steel, are difficult to
process in HPDC due to rapid mold wear and limitations in cooling technology. In addition,
metals with low fluidity may not accurately fill the mold, resulting in underfilling and surface
defects [1,8,13,20,26]. Porosity, primarily due to air entrapment during the filling phase, greatly
affects casting quality [3,23,24,26,29,31,32]. HPDC technology involves injecting liquid metal
under high pressure into a casting mold [19]. The injection chamber is filled with liquid
metal manually—with an operator pouring a specified amount of metal using a ladle—or
automatically—with a system dosing a precise amount into the chamber. The process then
proceeds through three phases of plunger movement. The first movement occurs in the shot
sleeve, where the alloy is forced forward by the plunger at a low speed of approximately
0.2–0.3 m·s−1 (depending on chamber filling and plunger diameter) until it nears the mold
gate (for as long as the mold design allows for low-speed flow without disturbances, usually
not beyond the gate) [24,32,33]. In the second movement, the alloy is transported at a high
plunger speed of approximately 2–4 m·s−1 , filling the casting and overflow/venting system
to complete the mold filling. The third very short plunger movement at low speed occurs
during the intensification stage and is aimed at compensating for shrinkage by pushing the
last solidifying alloy under high pressure, typically between 20 and 100 MPa [8,18,22,31].
Additionally, during intensification, most of the entrapped air generated by the fluid flow in
the shot chamber is compressed to relatively small sizes due to the applied pressure [31,34].
The presence of trapped air in the final porosity content is crucial for assessing the suitability
of the cast component for quality and acceptance in the subsequent finishing stages [32,35].
Shrinkage porosity occurs in the final solidification phase, and its content depends on the
design of the gating system and cavity-filling parameters. Voids in the casting structure
arise from alloy shrinkage during solidification, gases present directly in the alloy, or a
combination of these factors, including gas porosity from trapped air during high-pressure
casting [31,32,34,36]. Unfortunately, complete elimination of porosity in HPDC castings
is impossible [26]. Various techniques are employed to reduce it to acceptable levels,
including increased third-phase pressure values, higher second-phase velocities, use of
reduced pressure in the mold system, and optimization of the first (slow) phase to minimize
entrapped air in the chamber and liquid metal-air contact [16,37]. Some of these methods
may accelerate mold and plunger/cylinder mechanism wear and intensify unfavorable
cavitation phenomena in the mold [18,19,29].
Another important factor affecting porosity in the finished die casting is mold de-
sign [27,28]. Designing the gating system, positioning individual castings, and venting
are challenging and crucial, often preceded by extensive simulations verifying the design
accuracy [27,33]. Key parameters guiding designers include rapid mold cavity filling
with liquid metal, avoiding sharp angles that cause abrupt flow direction changes, and
proper mold venting [16,17,27,31]. In addition to the design parameters, calculating the
correct volume of metal needed to fill the mold cavity is critical. This is the total metal
volume, including the vents, casting, and gating system. Depending on casting size, the
metal-to-casting ratio ranges from 25 to about 80% [32,34]. In analyzing each mold system
design, the first component is the biscuit, which directly contacts the plunger. It has a
 Materials 2024, 17, x FOR PEER REVIEW 4 of 8
Materials 2024, 17, x FOR PEER REVIEW 4 of 8

including
including thethe vents,
vents, casting,
casting, andand gating
gating system.
system. Depending
Depending on on casting
casting size,
size, thethe metal-to-
metal-to-
Materials 2024, 17, 5935 casting ratio ranges from 25 to about 80% [32,34]. In
casting ratio ranges from 25 to about 80% [32,34]. In analyzing each mold system design,analyzing each mold system design,
4 of 23
the first component is the biscuit, which directly contacts
the first component is the biscuit, which directly contacts the plunger. It has a diameter the plunger. It has a diameter
matching
matching thatthat of the
of the injection
injection chamber
chamber andand a height
a height tiedtied to various
to various processes
processes andand moldmold
characteristics.
diameter
characteristics. matchingIts Its diameter
that
diameter of the is fixed;
injection
is fixed; however,
chamber
however, its
its and thickness
a height
thickness can
cantied vary.
to
vary. TheThe
various nominal
processes
nominal thick-
and
thick-
ness
mold and tolerances
characteristics. are
Its defined
diameter byis the process
fixed; designer
however,
ness and tolerances are defined by the process designer based on simulations and calcu-its based
thickness on simulations
can vary. The and calcu-
nominal
lations
thickness
lations [32,36,38].
and tolerances
[32,36,38]. Although
Although are biscuit
defined
biscuit thickness
by thedoes
thickness doesnotnot
process directly
designer
directly affect
based
affect the the
on geometry
simulations
geometry of
of theandthe
calculations
finished [32,36,38].
product, it Although
impacts biscuit
non-geometric thickness does
characteristics,
finished product, it impacts non-geometric characteristics, such as mold cavity fill quality not directly
such as affect
mold the
cavity geometry
fill quality
of theporosity
finished[38]. product, it impacts non-geometric characteristics, such as mold cavity fill
andand porosity [38].
qualityThisandstudyporosity [38]. the correlation between biscuit thickness and the quality of fin-
analyzes
This study analyzes the correlation between biscuit thickness and the quality of fin-
ishedThis study The
castings. analyzes the correlation
correlation between biscuitofthickness and the quality of
ished castings. The correlation cancan determine
determine thetheeffecteffect
of thisthis parameter
parameter on on predicting
predicting
finished
porosity castings.
in cast The correlation
elements at a very can determine
early production the effect
stage. of thisparameter
This parameterserves on predicting
as a basis
porosity in cast elements at a very early production stage. This parameter serves as a basis
porosity in cast elements at achanges
very early production stage.presented
This parameter serves asaacritical
basis
forfor implementing
implementing potential
potential changes [21,30]. [21,30].
TheThe casting
casting presented in this in this
studystudy is aiscritical
for implementing
component of potential safety
automotive changes [21,30].The
systems. Theanalyses
casting presented
performed inwill
thisexpand
study is a critical
knowledge
component of automotive safety systems. The analyses performed will expand knowledge
component
regarding of automotive
production safety systems.
stability, define The analyses performed will expand knowledge
regarding production stability, define thethe quality
quality of the
of the casting
casting directly
directly responsible
responsible forfor
regarding
human production
safety, and stability,
contribute define
to the
minimizing quality
castingof the casting
scrap. directly responsible for
human safety, and contribute to minimizing casting scrap.
human safety, and contribute to minimizing casting scrap.
2. 2.2. Materials
Materials and and Methods
Methods
Materials and Methods
The The analysis
analysis and and assessment
assessment of the
of of
the impact
impact of ofof biscuit
biscuit height
height ononon porosity,
porosity, and and conse-
conse-
The analysis and assessment the impact biscuit height porosity, and conse-
quently
quently onon on
the the quality
quality ofof of
the the
finalfinal product,
product, were were conducted
conducted using
using a casting
a casting (Figure
(Figure 1)1) 1)
mademade
quently the quality the final product, were conducted using a casting (Figure made
with with
HPDCHPDC technology
technology inin in a robotic
a robotic production
production cell.
cell. The The dimensions
dimensions ofofof the
thethe biscuit
biscuit (Figure
(Figure 2) 2)2)
with HPDC technology a robotic production cell. The dimensions biscuit (Figure
were were analyzed,
analyzed, focusing
focusing onon on
theirtheir correlation
correlation with with the
thethe porosity
porosity observed
observed in the
in in
the final final castings.
castings.
were analyzed, focusing their correlation with porosity observed the final castings.

Figure1.1.Tested
Figure Testedcasting.
casting.
Figure 1. Tested casting.

(a) (a) (b)(b)

Figure 2. Casting biscuit: (a) die casting tree with two castings—socket C1 and C2, (b) HPDC
biscuit—H: thickness, D: diameter.

The basic design parameters of the biscuit are presented in Table 1.

 Figure
Figure 2.
2. Casting
Casting biscuit:
biscuit: (a)
(a) die
die casting
casting tree
tree with
with two
two castings—socket
castings—socket C1
C1 and
and C2,
C2, (b)
(b) HPDC
HPDC bis-
bis-
cuit—H:
cuit—H: thickness,
thickness, D:
D: diameter.
diameter.
Materials 2024, 17, 5935 5 of 23
The basic design parameters of the biscuit are presented in Table 1.

Table
Table 1.
1. HPDC Table 1.
HPDC biscuit
biscuit HPDC biscuit
parameters
parameters for parameters
for the
the analyzed for
analyzed the analyzed casting.
casting.
casting.

Parameters of HPDC Parameters of HPDC Biscuit

Biscuit
Property Property Value Value
Biscuit height, mm
Biscuit height, mm 25 (+/− 10) 25 (+/−10)
Biscuit diameter,Biscuit
mm diameter, mm 70 70
Biscuit volume, cm3 96.21 (+/ −38.48)
Biscuit volume, cm 33 96.21 (+/− 38.48)

Porosity evaluation in the

Porosity castings was
evaluation performed
in the castingsatwas
twoperformed
critical points (Area
at two A andpoints (Area A
critical
Area B), which and
are planned
Area B),locations
which are forplanned
CNC machining. These
locations for CNC critical areas were
machining. exam-
These critical areas were
ined using X-ray inspection,
examined as illustrated
using in Figure
X-ray inspection, 3.
as illustrated in Figure 3.

Figure
Figure 3.
3. Porosity
Porosity analysis
analysis
Figure areas
areas (area
(area
3. Porosity A
A and area
andareas
analysis B—critical
area(area A andpoints).
B—critical points).
area B—critical points).

To ensure consistency
To ensureof consistency
results, the castings were
of results, produced
the castings using
were pressure
produced die cast-
using pressure die casting
ing with fixed technological parameters throughout
with fixed technological parametersthe cycle. Temperature
throughout the cycle.measurements
Temperature measurements
of the metal mold were
of the conducted
metal mold were continuously
conductedtocontinuously
stabilize casting conditions.
to stabilize Relevant
casting conditions. Relevant
thermograms are shown in Figure
thermograms 4. Information
are shown in Figureon 4.
theInformation
camera usedon andthe
thecamera
survey used
meth-and the survey
odology is provided in Section
methodology 2.3.
is provided in Section 2.3.

Figure
Figure 4.
4. Thermal analysis
Figure
Thermal of
of the
the casting
4. Thermal
analysis mold.
analysis
casting of the casting mold.
mold.

The gating system characteristics of the analyzed casting are presented in Table 2, and
the technological parameters of the production process are summarized in Table 3.
Materials 2024, 17, 5935 6 of 23

Table 2. Characteristics of the gating system.

Shot Chamber Properties

Property Value
Plunger diameter, mm 70
Active chamber length, mm 420
End of the filling hole, mm 70
Compression surface, cm2 310
Cast volume, cm3 83.33
Gating system volume, cm3 266.67
Overflow volume, cm3 55.56
Branch volume, cm3 488.89
Dosing volume, cm3 550

Table 3. Technological parameters of casting cycle.

Technological Parameters of the Casting Cycle

Property Value
Alloy EN-AC 46000
Pouring temperature, ◦ C 690
Die temperature, ◦ C 190
Speed at the filler slot, m/s 53.3
Intensification pressure, MPa 83
Maximum pressure in the dosing furnace, kPa 7

2.1. Material
The base material was the EN AC-46000 AlSi9Cu3(Fe) casting alloy. The metal was
melted in an industrial shaft furnace with a capacity of 3000 kg. The alloy composition
was determined using a SPECTROMAXx SPECTRO emission spectrometer (SPECTRO
POLAND, Kleve, Germany) on samples taken directly from the casting and in accordance
with EN 1706:2020 + A1:2021 standards [39]. The tested composition is presented in Table 4.
The metal was held in a Stotek Pro Dos 3 furnace (a pressure dosing system for supplying
casting alloy to the machine and molds).

Table 4. Chemical composition of the applied alloy.

Element Value, %
Al 86.1
Si 8.8
Fe 0.87
Cu 0.87
Mn 0.195
Mg 0.15
Cr 0.044
Ni 0.08
Zn 1.03
Ti 0.051
Pb 0.084
Sn 0.021

2.2. Test Casting and Process Parameters

Foundries vary greatly in their degree of mechanization and automation. In HPDC,
two methods of filling the pressure chamber are used: manual by the operator and fully
automated. In the analyzed case, the casting machine was a modern automated die-casting
machine—the Buhler Evolution 84 model. It is fully computerized, allowing for data
 2.2. Test Casting and Process Parameters
Foundries vary greatly in their degree of mechanization and automation. In HPDC,
Materials 2024, 17, 5935
two methods of filling the pressure chamber are used: manual by the operator and7fully of 23

automated. In the analyzed case, the casting machine was a modern automated die-cast-
ing machine—the Buhler Evolution 84 model. It is fully computerized, allowing for data
collection
collection and export to
and export totext
textfiles.
files.The
The parameters
parameters of the
of the pressure
pressure casting
casting machine
machine are
are pro-
provided in Table 5, and its appearance is shown in Figure
vided in Table 5, and its appearance is shown in Figure 5. 5.

Table5.5.Technical
Table Technicalparameters
parametersof
ofBuhler
BuhlerEvolution
Evolution84.
84.
Parameter
Parameter Specification
Specification
Type
Type Die
Diecasting
casting machine—metal injection
machine—metal injection
Model
Model Buhler Evolution 84
Buhler Evolution 84
Maximum casting weight 5.03 kg (Al)
Maximum casting weight 5.03 kg (Al)
Clamping force 850 tonnes
Clamping force
Machine weight 850 tonnes
36.5 tonnes
Machine weight 36.5 tonnes

Figure5.
Figure 5. HPDC
HPDC machine
machine Buhler
Buhler Evolutrion
Evolutrion84.
84.

The
The study
study waswas conducted
conducted on on aa robotic
robotic production
production cell cell comprising
comprising aa casting
casting machine,
machine,
dosing furnace,robot
dosing furnace, robotfor forremoving
removing thethe casting
casting along
along withwith the gating
the gating system,
system, spraying
spraying robot,
robot,
conveyorconveyor belt,trimming
belt, and and trimming press.press. The furnace
The furnace places places a specified
a specified amount
amount of material
of material into
into the shot
the shot chamber,
chamber, followed
followed by aby a shot,
shot, according
according to the
to the technological
technological settings.
settings. TheThedata data
for
for the process are shown in Table 3. Next, the mold is halved separately,
the process are shown in Table 3. Next, the mold is halved separately, the ejector plate is the ejector plate
is activated,
activated, andand
thethe sprue
sprue is picked
is picked up byup the
by the removing
removing robot.robot. Meanwhile,
Meanwhile, the spraying
the spraying robot
robot
applies a protective coating to the mold surface. The removing robot then transfers transfers
applies a protective coating to the mold surface. The removing robot then the sprue
the sprue
to the to the
sensor sensor
area area to confirm
to confirm its completeness
its completeness and places anditplaces
on theit conveyor
on the conveyor belt,
belt, which
which transports
transports the product
the product to the operator.
to the operator. The worker
The worker places the places the casting
casting in a trimming
in a trimming press to
press
remove to remove the residual
the residual air vent airandvent andsystem
gating gating and
system and visually
visually inspectsinspects the trimmed
the trimmed casting.
casting. Based on the assessment, the casting is placed
Based on the assessment, the casting is placed in either the OK or NOK bin. in either the OK or NOK bin.
The
The Buhler
Buhler chamber
chamber is is supplied
supplied withwith metal
metal by
by aa pressure
pressure dosing
dosing system
system forfor casting
casting
machines
machines and molds Stotek Pro Dos 3 (Figure 6). It is a resistance electric furnace that
and molds Stotek Pro Dos 3 (Figure 6). It is a resistance electric furnace that
maintains
maintains the temperaturefor
the temperature fordosing
dosingaluminum
aluminumalloys alloys without
without a crucible.
a crucible. Filling
Filling the
the shot
shot chamber
chamber with with
such such
devices devices involves
involves air pressure
air pressure exerted exerted
on theon the liquid
liquid metal. metal.
The dosingThe
dosing furnace is connected to the casting machine, where the dosing
furnace is connected to the casting machine, where the dosing value is set in mass units— value is set in mass
units—kilograms.
kilograms. The gas Thepressure
gas pressure increase
increase exerts
exerts pressure
pressure on onthethe liquid
liquid material,
material, causinga
causing
apressure
pressurerise.
rise.ToToachieve
achievepressure
pressure equilibrium,
equilibrium, aa flow
flow is is generated, moving the
generated, moving the materialmaterial
towards the riser pipe. The liquid metal is then transferred to a heated transfer trough,
towards the riser pipe. The liquid metal is then transferred to a heated transfer trough,
flowing directly into the gating chamber. After this operation, the casting process proceeds
flowing directly into the gating chamber. After this operation, the casting process pro-
according to the set technological parameters. The nominal biscuit thickness was 25 mm,
ceeds according to the set technological parameters. The nominal biscuit thickness was 25
with an allowable tolerance of +/−10 mm.
mm, with an allowable tolerance of +/− 10 mm.
Automatic recording of biscuit thickness was conducted for 100% of the injections via
the casting machine, and the data were exported to MS Excel. The number of measurement
points: 4995.
Materials2024,
Materials 2024,17,
17,5935
x FOR PEER REVIEW 88ofof238

Figure 6. Stotek Pro Dos 3 pressure dispensing machine.

Automatic recording of biscuit thickness was conducted for 100% of the injections via
the casting machine, and the data were exported to MS Excel. The number of measure-
ment 6.points:
Figure
Figure Stotek4995.
6. Stotek Pro Dos
Pro Dos 33 pressure
pressuredispensing
dispensingmachine.
machine.

2.3.
2.3.Methodology
Methodology
Automatic recording of biscuit thickness was conducted for 100% of the injections via
The experimental
the casting methodology
machine, and is presented
the data were intoFigure 7.
The experimental methodology isexported
presented MS
in Excel.
Figure 7.The number of measure-
ment points: 4995.

Melting alloy
Preparing alloy

CTGII/600
Furnance
Pouring

Analysis
Alloy refining

URM-1/A

Alloy holding,
automatic dosing

Stotek Pro Dos 3

Casting

Buhler
Evolutrion 84

Removing cast,
measuring
Buhler
Evolutrion 84

RTG analysis

YXLON Y.MU2000-D

Data analysis,
statistical analysis
MS Excel,
Statistica 13

Figure7.7.Diagram
Figure Diagramofof the
the research
research methodology
methodology applied
applied in experiment.
in the the experiment.
24, 17, x FOR PEER REVIEW
Materials 9 of 23
2024,17,
Materials2024, 17,x5935
FOR PEER REVIEW 9 of 23
9 of 23

The recorded data Thefrom

The the data
recorded
recorded Buhler
data Evolution
from
from the the 84Evolution
Buhler
Buhler machine 84were
Evolution 84exported
machine
machine towere
a com-
exported
were exported to a com-
to a computer,
puter, and the results
and are
the shown
results in
are Figure
shown in8.
puter, and the results are shown in Figure 8.
Figure 8.

Figure 8. Sample data from

Figure
Figure
Buhlerdata
8.8.Sample
Sample Evolution
datafrom 84.
fromBuhler
Buhler Evolution
Evolution 84.84.

The data were statistically

Thedata
The datawereanalysed
were usinganalysed
statistically
statistically Statistica
analysed 13.3
using
using byStatistica
StatSoft
Statistica Europe
13.3
13.3 by by GmbH.
StatSoft
StatSoft Europe
Europe GmbH.
GmbH.
The data cleaning
The process
data was carried
cleaning out
process entirely
was in
carried Statistica.
out entirely The
in methodology
Statistica. The of data
The data cleaning process was carried out entirely in Statistica. The methodology of data
methodology of data
cleaning is presented in Figure
cleaning
cleaning 9.
isispresented
presented ininFigure
Figure9. 9.

Loading data (Statistica)

Loading data (Statistica)

Profiling (data quality assessment)

Profiling (data quality assessment)

Data cleaning (removal of outliers)

Data cleaning (removal of outliers)

Preparation of the final collection

Preparation of the final collection

Statistical analysis
Statistical analysis

Figure 9. DiagramFigure
of data
Figure 9.9.cleaning
Diagramapplied
Diagram ofofdata incleaning
the experiment.
datacleaning applied
applied in in
thethe experiment.
experiment.

The porosity of The

the
Thecastings
porositywas
porosity evaluated
ofofthe
the castings
castings using
was
was a non-destructive
evaluated
evaluated using
using method withmethod
a non-destructive
a non-destructive an method
withwithan an
RTG machine—YXLON
RTG Y.MU2000-D. The
RTG machine—YXLON
machine—YXLON castings were
Y.MU2000-D.
Y.MU2000-D. The TheX-rayed,
castingsand
castings were weretheX-rayed,
X-rayed,resulting
and im-the resulting im-
and
the resulting images
ages were assessed
were based on
assessed ASTM
based E
on 505—“Standard
ASTM E Reference
505—“Standard Radiographs
Reference for
Radiographs
ages were assessed based on ASTM E 505—“Standard Reference Radiographs Inspec-
for Inspection of
for Inspec-
tion of Aluminum tionand
Aluminum Magnesium Die Magnesium
and Magnesium
of Aluminum and Castings” and
Dieinternal
Die Castings” standards.
and internal
Castings” The obtained
standards.
and internal
The obtained results
standards. The obtained
were used
results were usedresults for qualitative
for qualitative evaluation evaluation of the
of theevaluation castings.
castings. The The
detailed detailed
parametersparameters
of the of the ma-
were used for qualitative of the castings. The detailed parameters of the
chine are summarized in Table 6, while the casting analysis methodology is illustrated in
machine are summarized
machine arein Table 6, whilein
summarized the casting
Table analysis
6, while the methodology
casting analysisis illustrated
methodology is illustrated
Figure 10.
in Figure 10. in Figure 10.

Table 6. Technical Table

parameters of X-Ray
6. Technical Machine of
parameters YXLON
X-Ray Y.MU2000-D.
Machine YXLON Y.MU2000-D.

Parameter Parameter Specification Specification

Manufacturer Manufacturer Comet Yxlon, Hamburg, Germany
Comet Yxlon, Hamburg, Germany
Type Type X-Ray Machine X-Ray Machine
Model Model YXLON MU2000- D Standard
YXLON MU2000- D Standard
Materials 2024, 17, 5935 10 of 23

Table 6. Technical parameters of X-Ray Machine YXLON Y.MU2000-D.

Parameter Specification
Materials 2024, 17, x FOR PEER REVIEW 10 of 23
Manufacturer Comet Yxlon, Hamburg, Germany
Type X-Ray Machine
Model YXLON MU2000-D Standard
RTG Lamp
RTG Lamp Y.TU
Y.TU 160-D06 160kV
160-D06 160 kV
Rotary table
Rotary table 600 mm
600 mm
Software including ASTM window Y.IMAGE 2500-D
Software including ASTM window Y.IMAGE 2500-D
Frequency of basic calibration before each test
Frequency of basic
Main frequency calibration
calibration before
annually test
each
Main frequency calibration annually

Selection of the sample (casting) to be

examined, placement in the X-ray machine

Positioning of casting in chosen position, basic

X-ray calibration

Taking of the X-ray image and transfer to the

computer

Digital processing - Y.Image x500 and PXM 2500

software

Comparison of porosity with reference images

of ASTM E 505 standard and porosity evaluation

Figure 10.
Figure 10. Schematic
Schematic diagram
diagram of
of the
the X-ray
X-ray research
research methodology
methodology used
used in
in the
the experiment.
experiment.

When the
When the casting
casting machine
machine waswas stopped due to
stopped due to maintenance
maintenance or or breakdown,
breakdown, aa thermal
thermal
analysis of
analysis ofthe
themould
mouldwas wasperformed
performedusing
using a FLIR
a FLIR E53E53
fromfrom Teledyne
Teledyne Flir Flir (Wilsonville,
(Wilsonville, OR,
OR, USA).
USA). The The Camera
Camera parameters
parameters are given
are given in Table
in Table 7, and
7, and thethe methodology
methodology forfor perform-
performing
ing thermal
the the thermal analysis
analysis is shown
is shown in Figure
in Figure 11. 11.

Table 7.
Table 7. Technical
Technical parameters
parameters of
of FLIR
FLIR E53.
E53.

Parameter
Parameter Specification
Specification
Type
Type Thermal Imaging
Thermal Camera
Imaging Camera
Model
Model FLIR E53
FLIR E53
◦ C @ 30 ◦ C (86 ◦ F), 42◦ lens
Features
FeaturesE53
E53 Thermal Sensitivity/NETD
Thermal Sensitivity/NETD <0.03
<0.03 °C @ 30 °C (86°F), 42° lens
IR Resolution 240 × 180 (43,200 pixels)
IR Resolution 240 × 180 (43,200 pixels)
Object Temperature Range 0 ◦ C to 650 ◦ C
ObjectImage
Temperature
FrequencyRange 0 °C30 toHz
650 °C
Image Frequency 30 Hz

Scheduled or emergency stop of the machine

Maintenance or fault diagnosis, execution of start-up

injection, achievement of set temperature by mould

Taking pictures with FLIR E53 -Analysis of

thermogram

Thermogram analysis

Decision to start the production process

Figure 11.
Figure 11. Schematic
Schematic diagram
diagram of
of the
the thermal
thermal imaging
imaging methodology
methodology used
used in
in the
the experiment.
experiment.

3. Results
The obtained results are from the ongoing production of the foundry, and the data
cover a 7-day production process. The analysis was divided into three parts. The first part
includes statistical data analysis, descriptive statistics, data cleaning, normality
Materials 2024, 17, 5935 11 of 23

3. Results
The obtained results are from the ongoing production of the foundry, and the data
cover a 7-day production process. The analysis was divided into three parts. The first
part includes statistical data analysis, descriptive statistics, data cleaning, normality dis-
tribution analysis, and time-course analysis. The second part is the RTG analysis, which
determines the relationship between porosity and biscuit thickness. The third part contains
an assessment of the dimensional stability of the biscuit.

3.1. Data Analysis

Given the large number of data points (4995 samples), preliminary statistical analysis
was conducted to verify their accuracy. The analysis included basic descriptive statistics
parameters, allowing the estimation of the distribution location, variability, and data
structure. Measures such as the mean, median, standard deviation, variance, higher-order
moments, and standard errors were used.
The mean, as a measure of central tendency, indicates the average biscuit thickness,
while the median indicates the middle value, eliminating the impact of extreme values.
The standard deviation and variance describe the dispersion level of the results around
the mean, with the standard deviation being the square root of the variance, allowing for a
data dispersion expression in the same units as the measurement.
Higher-order moments, such as skewness, measure the asymmetry of the distribution
relative to the mean, identifying deviations towards higher or lower values. Kurtosis de-
scribes the degree of concentration of data around central values: higher kurtosis indicates
more data clustered around the mean, reflected in a more peaked distribution curve and
possibly indicating outliers.
The analysis also includes standard errors, assessing the precision of the mean and
other parameters, indicating how representative the sample mean is for the entire popula-
tion. The results of the statistical analysis are presented in Table 8.

Table 8. Statistical data analysis results.

Parameter Specification
Valid N 4995
Percent of Valid Observation 100
Mean 24.6580581
Median 25
Minimum 10
Maximum 40
Variance 10.7577019
Standard deviation 3.27989358
Coefficient of Variation 13.301508
Standard Error 0.0464079096
Skewness −0.189708083
Kurtosis 0.300340094

Analysis of the data presented in Table 8 shows that both the mean and median
biscuit thicknesses correspond to nominal values, suggesting central clustering of the
results around the target. The measurements exhibit low variability, with athe coefficient of
variation (CV) of 13%. The CV, defined as the ratio of the standard deviation to the mean,
expresses data dispersion relatively independent of the measurement units.
The values of skewness and kurtosis confirm that the distribution of results is close
to a normal distribution. Skewness close to zero indicates an even distribution around
the mean, while kurtosis suggests a shape similar to a normal distribution, indicating the
absence of extreme values.
However, extreme biscuit thickness values, from 10 mm (minimum value) to 40 mm
(maximum value), indicate the presence of some anomalies and deviations from the desired
Materials 2024, 17, 5935 12 of 23

biscuit thickness, suggesting the influence of random factors or process errors.

Materials 2024, 17, x FOR PEER REVIEW As all the
12 of 23
Materials 2024, 17, x FOR PEER REVIEW 12 analyze
data contain a precise record of the time it was produced, it was possible to of 23 why
such values were being produced: the analysis showed that the outlier data came from
injections made were
why such values after being
the machine
produced: had
thestopped. In suchthat
analysis showed cases, the machine
the outlier data came carries out
why
the such values were being produced: the analysis showed that the outlier
from injections made after the machine had stopped. In such cases, the machine carries reduced
process with special care for the mold, using reduced injection data
speeds came
and
from
out theinjections
compression madespecial
processpressure.
with afterThese
the
caremachine
for thethe
are had
mold, stopped. In such
using reduced
so-called start-up cases,
injection the
speeds
injections machine carries
and reduced
(castings from these
out the
injectionsprocess
compression with special
arepressure.
scrapped),Thesecare for
are the
which the
can mold,
so-called using
deviatestart-upreduced injection
injections
significantly from speeds
(castings and reduced
from these
the standard in-
tolerances. In
compression
jections
the caseare pressure.
scrapped),
analyzed, Thesewere
which
these are deviate
can the
less so-called15start-up
significantly
than orinjections
mm from (castings
the standard
more than from
35tolerances.
mm these
(out In in-
ofthe
tolerance).
jections
Taking are
intoscrapped),
case analyzed, account which
these were lesscan
these datadeviate
than 15 mm
(whose significantly
orvalues
more thanfrom
35 the
differ mm standard tolerances.
(out of tolerance).
significantly from those In the
Taking assumed)
case
into analyzed,
account these
these were
data less than
(whose values15 differ
mm orsignificantly
more than 35 mmthose
from (out assumed)
of tolerance). Taking
could lead
could lead to situations in which the interpretation and conclusions drawn from the control
into account these
to situations data the
in which (whose values differ
interpretation significantly from those assumed) couldcards
lead
cards could lead to incorrect decisionsand conclusions
regarding drawn
the production from the control
process.
to situations
could lead to in which the
incorrect interpretation
decisions regarding and
theconclusions
production drawn
process.from the control cards
For
couldFor
better
lead
data visualization,
to incorrect decisions regarding
a histogram (Figureprocess.
the (Figure
production
12) and a box plot (Figure 13) were
better data visualization, a histogram 12) and a box plot (Figure 13) were
used.For The histogram assesses the shape of the data distribution, while the box plot visually
used. Thebetter data visualization,
histogram a histogram
assesses the shape (Figure
of the data 12) and awhile
distribution, box plot (Figure
the box plot13) were
visually
represents
used.
represents
the rangeofofvariability,
The histogram variability,
the range assesses the shape quartile
of the
quartile
positions,
data and while
distribution,
positions,
outliers.
and outliers. the box plot visually
represents the range of variability, quartile positions, and outliers.
Histogram of HPDC Biscuit thickness
3500 Histogram of HPDC Biscuit thickness
3500

3000
3000

2500
2500

2000
of obs

2000
of obs
No No

1500
1500

1000
1000

500
500

0
0 5 10 min 20 25 30 max 40 45
5 10 min 20
HPDC 25 thickness
Biscuit 30 max 40 45
HPDC Biscuit thickness
Figure 12.
Figure 12. Histogram
Histogram ofof
HPDC biscuit
HPDC thickness.
biscuit thickness.
Figure 12. Histogram of HPDC biscuit thickness.
Box Plot of HPDC Biscuit thickness
45 Box Plot of HPDC Biscuit thickness
45

40
40

35
35
thickness

30
thickness

30

25
Biscuit

25
Biscuit
HPDC

20
HPDC

20

15
15 Median = 25
25%-75%
25% − 75%
Median = 25
10 =25%-75%
(22,−27)
25% 75%
10 =Non-Outlier
(22, 27) Range
=Non-Outlier
(15, 34) Range
5 =Outliers
(15, 34)
5 Outliers

Figure 13. Box Plot of HPDC biscuit thickness.

Materials 2024, 17, x FOR PEER REVIEW 13 of 23
Materials 2024, 17, x FOR PEER REVIEW 13 of 23

Materials 2024, 17, 5935 13 of 23

Figure 13. Box Plot of HPDC biscuit thickness.

Figure 13. Box Plot of HPDC biscuit thickness.

Analysis of
Analysis of the
the obtained
obtained plots
plots revealed
revealed outliers,
outliers, whose
whose origin
origin needs
needs toto be
beanalyzed
analyzed
Analysis ofsource
the obtained plots revealed outliers, whose origin needs to be analyzed
concerning the source of such deviations. Additionally, the “T-shaped whiskers”are
concerning the of such deviations. Additionally, the “T-shaped whiskers” areclose
close
concerning the source of such deviations. Additionally, theBased
“T-shaped whiskers” are close
to the maximum and minimum biscuit thickness values. Based on this information, the
to the maximum and minimum biscuit thickness values. on this information, the
to thewere
data maximum and minimum biscuit thickness values. Based on this information, the
data werecleaned,
cleaned,andandtheir
theirgraphical
graphicalanalysis
analysisisispresented
presentedininFigures
Figures14
14and
and15.
15.
data were cleaned, and their graphical analysis is presented in Figures 14 and 15.

Histogram of HPDC Biscuit thickness

1400 Histogram of HPDC Biscuit thickness
1400

1200
1200

1000
1000

800
obs

800
obs
of of
NoNo

600
600

400
400

200
200

0
0 15.0 16.9 18.8 20.7 22.6 24.5 26.4 28.3 30.2 32.1 34.0
15.0 16.9 18.8 20.7 22.6 24.5 26.4 28.3 30.2 32.1 34.0
HPDC Biscuit thickness
HPDC Biscuit thickness

Figure 14. Histogram

Histogram of HPDC
HPDC biscuitthickness
thickness afterdata
data cleaning.
Figure 14. Histogram of
Figure 14. of HPDC biscuit
biscuit thickness after
after data cleaning.
cleaning.

Box Plot of HPDC Biscuit thickness

36 Box Plot of HPDC Biscuit thickness
36

34
34

32
32

30
30
thickness

28
thickness

28

26
26
Biscuit
Biscuit

24
24
HPDC
HPDC

22
22

20
20

18
18 Median = 25
Median
25%-75%
25% − 75%= 25
16 25%-75%
=25%
(22,−27)
75%
16 =Non-Outlier
(22, 27) Range
14 =Non-Outlier
(15, 34) Range
14 = (15, 34)
Figure 15. Box Plot of HPDC biscuit thickness after data cleaning.
Figure 15.
Figure 15. Box
Box Plot
Plot of
of HPDC
HPDC biscuit
biscuit thickness
thickness after
after data
data cleaning.
cleaning.

After data cleaning, 25 measurements of biscuit thickness were excluded from further
After data
After data cleaning,
cleaning, 25
25 measurements
measurements ofof biscuit
biscuit thickness
thickness were
were excluded
excluded from
from further
further
analysis, constituting approximately 0.5% of the total, which, from a statistical perspec-
analysis, constituting
analysis, constitutingapproximately
approximately0.5%
0.5%ofofthe
the total,
total, which,
which, from
from a statistical
a statistical perspec-
perspective,
tive, is negligible and does not significantly impact the results. Including them could cause
tive,
is is negligible
negligible and and
doesdoes
not not significantly
significantly impact
impact thethe results.Including
results. Includingthem
themcould
could cause
cause
significant measurement results, distortions, uncertainties in interpretation, leading to false
conclusions and inappropriate decisions.
 Materials 2024, 17, x FOR PEER REVIEW 14 of 2
Materials 2024, 17, 5935 14 of 23

significant measurement results, distortions, uncertainties in interpretation, leading to

The nextfalse
stepconclusions
in the analysisand inappropriate
was to assessdecisions.
the changes in biscuit thickness during
The next
the production process. stepanalysis
This in the analysis
aims to was to assess
reveal the changes
potential trends,influctuations,
biscuit thicknessandduring th
production
unforeseen changes overprocess.
a longThis analysis aims
measurement to reveal
period, potential
which wouldtrends,
notfluctuations, and unforeseen
be visible using
earlier tools. changes over a long measurement period, which would not be visible using earlier tools.
Production
Production processes oftenprocesses oftenvarious
experience experience various
failures thatfailures that affect
affect product product parameter
parameters
and sometimesand sometimes
cause machinecause machine
stoppages. stoppages.
Figure 16 shows Figure 16 shows the
the thickness thickness
changes duringchanges
the during
theCharacteristic
analysis period. analysis period. Characteristic
trends trends and
were identified, weretheir
identified,
causesand weretheir causes
verified were verified
using
using the tool
the tool shop department andshop department
machine monitoringand records.
machineThreemonitoring records.stages
characteristic Threewerecharacteristi
identified. Atstages
the end were identified.
of stage At the end
1, a machine of stage 1,
stoppage a machine
occurred duestoppage occurred
to the need due to
to clean a the need
mold elementtoand clean a moldmaterial
unblock element and
flowunblock
throughmaterial flow through
the trough. This recordthe trough.
indicatesThis record indi
that
cates that
the heated transfer the transporting
trough heated transfer trough
liquid transporting
metal liquid metalreducing
was contaminated, was contaminated,
its flow reduc
ing its flow
capacity and affecting capacity
the amount andofaffecting the amount
metal dosed. of metal dosed.
Any blockage in theAny blockage
trough resulted in the trough
resulted
in less material in the in lesschamber,
shot material lowering
in the shotthechamber,
biscuitlowering
thickness. theAtbiscuit
the endthickness.
of stageAt2,the end o
problems withstage 2, problems
the sealing of thewith the movable
mold’s sealing ofelement—the
the mold’s movable element—the
slider—were recorded.slider—were
This rec
orded. This led to a stoppage during which the tool shop
led to a stoppage during which the tool shop department performed the regeneration work. department performed the re
generation work. After remounting and fitting the sliders,
After remounting and fitting the sliders, production continued. The effectiveness of the production continued. The ef
fectiveness of the operation is confirmed by the absence
operation is confirmed by the absence of further sealing issues for the remainder of the of further sealing issues for th
measurementremainder
period. of the measurement period.

Figure
Figure 16. Box Plot of 16.
HPDCBox biscuit
Plot of HPDC biscuit
thickness thickness
after after data cleaning.
data cleaning.

3.2. Analysis of3.2. Analysis

X-Ray of X-Ray Inspection
Inspection
The castingsradiographic
The castings underwent underwent radiographic
inspection to inspection
verify theto porosity.
verify the Due
porosity.
to theDue to th
volume of thevolume
produced of the produced components,
components, it was not
it was not feasible feasible
to X-ray to X-ray
every everyInstead,
product. product. Instead
representativerepresentative
samples were samples were
selected, selected, considering
considering productionproduction process variability
process variability and and
spanning a widespanning a widetotime
time range, range,
ensure thattothe
ensure that
results the results
closely closely approximate
approximate what would what
be would
be observed
observed if every productifhadevery product had
undergone undergoneinspection.
radiographic radiographic inspection.
Porosity levels Porosity levels were
were evaluated evaluatedtoaccording
according ASTM E505 to ASTM E505 [35] guidelines
[35] guidelines and interna
and internal
factory
factory standards, standards,
with with porosity
the adopted the adopted porosity classification
classification detailed in detailed
Table 9. in Table 9.
Figures 17–19 show example X-ray images for three thicknesses of the biscuit as per
tolerance criteria. Figure 17 depicts the thickness within the tolerance, Figure 18 shows
the thickness above the upper tolerance, and Figure 19 shows the thickness below the
 Materials 2024, 17, x FOR PEER REVIEW 15 of 23
Materials 2024, 17, x FOR PEER REVIEW 15 of 23
Materials 2024, 17, 5935 15 of 23

Table 9. Detailed classification of adopted porosity.

Table 9. Detailed classification of adopted porosity.
lower tolerance. Two critical areas (A X-Ray and B)Inspection
(marked in Figure 3) were highlighted
Criteria
regarding further processing, along with
Rate X-Ray Inspection
two sockets Criteria
(Socket C1 and C2) related to the
Value
casting tree structure. Rate
Fine porosity Value
1
Numerous porosities Fine
andporosity
small, isolated shrinkages 12
Table 9. Detailed classification of adopted porosity.
Numerous porosities
Increased andand
porosity small, isolated
large shrinkages
shrinkages 23
Increased porosityX-RayandInspection
large shrinkages
Criteria 3
Figures 17–19
Rate show example X-ray images for three thicknesses
Value of the biscuit as per
Figures 17–19 show example X-ray images for three thicknesses
tolerance criteria. Figure 17 depicts the thickness within the tolerance, Figureof the biscuit as per
18 shows
tolerance
Finecriteria.
porosityFigure 17 depicts the thickness within 1the tolerance, Figure 18 shows
the thickness above the upper tolerance, and Figure 19 shows the thickness below the
Numerous porosities and small, isolated
the
lowerthickness above the upper tolerance, and Figure 19 shows
tolerance. the thickness below the
shrinkagesTwo critical areas (A and B) (marked in Figure 3) were highlighted re-
2
lower
garding
Increased tolerance.
further
porosity Two critical
processing,
and large along with two sockets (Socket3 C1 and3)C2)
shrinkagesareas (A and B) (marked in Figure were highlighted
related re-
to the cast-
garding further processing, along with two sockets (Socket C1 and C2) related to the cast-
ing tree structure.
ing tree structure.

(a) (b)
(a) (b)

(c) (d)
(c) (d)
Figureinspection
Figure 17. X-ray 17. X-ray inspection for(about
for nominal nominal (about
26 mm) 26 mm)
value value of
of biscuit biscuit thickness
thickness in specificinsockets:
specific sockets:
Figure
socket 17.
C1: X-ray
(a) inspection
area A—rate for nominal
value 1, (b) (about
area 26
B—ratemm) value
value 2; of biscuit
socket C2: thickness
socket C1: (a) area A—rate value 1, (b) area B—rate value 2; socket C2: (c) area A—rate valuevalue
(c) area in specific
A—rate 2, sockets:
2, (d)
socket C1: (a)value
area B—rate area A—rate
2. value 1, (b) area B—rate value 2; socket C2: (c) area A—rate value 2, (d)
(d) area B—rate value 2.
area B—rate value 2.

(a) (b)
(a) (b)
Figure 18. Cont.
Materials 2024,Materials
17, 59352024, 17, x FOR PEER REVIEW 16 of 23 16 16
of 23
Materials 2024, 17, x FOR PEER REVIEW of

(c) (d)
(c) (d)
Figureinspection
Figure 18. X-ray 18. X-ray for
inspection
maximumfor (about
maximum (about
35 mm) 35 mm)
value valuethickness
of biscuit of biscuitinthickness in specific sock-
specific sockets:
Figure 18. C1:
X-ray inspection forvalue
maximum (about 35 mm) value of biscuit thickness in specific soc
socket C1: (a) area A—rate value 3, (b) area B—rate value 2; socket C2: (c) area A—rate value 3, value 3,
ets: socket (a) area A—rate 3, (b) area B—rate value 2; socket C2: (c) area A—rate
ets:area
(d) socket C1: (a)
B—rate area2.A—rate value 3, (b) area B—rate value 2; socket C2: (c) area A—rate value
value
(d) area B—rate value 2.
(d) area B—rate value 2.

(a) (b)
(a) (b)

(c) (d)
Figure(c)19. X-ray inspection for minimum (about 10 mm) value (d)of biscuit thickness in specific sockets:
socket C1: (a) area A—rate value 3, (b) area B—rate value 3; socket C2: (c) area A—rate value 3, (d)
Figure
Figure 19. X-ray 19. X-ray
inspection
area B—rate
inspection
for
value
for minimum (about 10 mm) value of biscuit thickness in specific socket
3.minimum (about 10 mm) value of biscuit thickness in specific sockets:
socket C1: socket
(a) areaC1: (a) areavalue
A—rate A—rate value
3, (b) area3,B—rate
(b) areavalue
B—rate value 3;
3; socket socket
C2: C2: A—rate
(c) area (c) area A—rate
value 3,value 3, (
area B—rate value 3.
The analysis in Figure 16 enabled the segmenting of the process into phases for qual-
(d) area B—rate value 3.
itative evaluation of casting quality regarding defect levels. This analysis utilized radio-
The analysisTheinanalysis
Figure in16Figure 16 enabled
enabled the segmentingthe of the process into phases for qua
graphic (X-ray) inspection data. the segmenting
Figures 20 and 21ofpresent
process into phases
these results. Figurefor
20 shows
itative
qualitativethe evaluation
evaluation of of casting
casting quality
quality regarding
regarding defect
defect levels.
levels. This This analysis
analysis utilized radio
utilized
average rejection rate (in ASTM classification) at each stage, while Figure 21 indicates
graphic
radiographic (X-ray) inspection
(X-ray) data.data. Figures 20 and 21 present these results. Figure
20 20 show
defect levelsinspection
due to porosity Figures 20 and
inconsistencies 21surfaces
on present these results.
designated Figure
for machining.
the average rejection rate (in ASTM classification) at each stage,
shows the average rejection rate (in ASTM classification) at each stage, while Figure 21 while Figure 21 indicate
defect levels due to porosity inconsistencies on surfaces designated
indicates defect levels due to porosity inconsistencies on surfaces designated for machining.for machining.
Materials 2024, 17, x FOR PEER REVIEW 17 of 23
Materials 2024, 17, 5935 17 of 23

Figure 20. Average quality ratings based on X-ray images.

Based on Figure 20, phases 1 and 2 exhibit similar cumulative quality levels according
to X-ray assessment, reaching around 2.31/3. The third, final phase shows a decrease to
around 1.9/3, marking an improvement of approximately 18%. This rejection rate reduc-
tion resulted from efforts to seal the movable mold part—the slider—ensuring that previ-
ously
Figureleaking
Figure 20. material
20.Average
Average stayed
quality
qualityratingswithin
ratingsbased the
basedon mold,
onX-ray
X-ray improving material feed for the part.
images.
images.

Based on Figure 20, phases 1 and 2 exhibit similar cumulative quality levels according
to X-ray assessment, reaching around 2.31/3. The third, final phase shows a decrease to
around 1.9/3, marking an improvement of approximately 18%. This rejection rate reduc-
tion resulted from efforts to seal the movable mold part—the slider—ensuring that previ-
ously leaking material stayed within the mold, improving material feed for the part.

Figure
Figure 21.
21. Defective
Defectivecastings
castingsdue
dueto
toporosity
porosity issues
issues on
on machining
machining surfaces.
surfaces.

Figure
Based on 21 Figure
presents 20,aphases
percentage
1 andanalysis
2 exhibitofsimilar
castings failing to quality
cumulative meet porosity require-
levels according
ments. Phase one had a relatively high rejection rate for porosity on machined
to X-ray assessment, reaching around 2.31/3. The third, final phase shows surfaces
a decrease to
(the average value for sockets was 1.74%). This level decreased across subsequent phases,
around 1.9/3, marking an improvement of approximately 18%. This rejection rate reduction
resulted
from 0.92%fromto efforts
0.79%, to seal theamovable
marking mold part—the
55% reduction from the slider—ensuring
initial value. Thisthat previously
improvement
leaking material
was achieved
Figure stayed
by stabilizing
21. Defective within
castings duethe the mold, improving
process issues
to porosity and adjusting material
on machining feed
the surfaces. for the part.
biscuit thickness closer to the
Figure
nominal value. 21 presents a percentage analysis of castings failing to meet porosity require-
ments. Phase one had a relatively high rejection rate for
Figure 21 presents a percentage analysis of castings failing to meet porosity on machined surfaces
porosity require-
(the average
3.3. Dimensional
ments. value for
Phase oneStability sockets was 1.74%).
Assessmenthigh
had a relatively of BiscuitThis level
Thickness
rejection decreased across subsequent
rate for porosity on machined surfaces phases,
from 0.92%
(the average to 0.79%, marking
value forstatistical
sockets was a 55% reduction
1.74%). from the initial value. This improvement
A preliminary analysis of This level decreased
the biscuit thicknessacross subsequent
measurements phases,
identified
was achieved
from 0.92% to by stabilizing
0.79%, marking the
a process
55% and adjusting
reduction from the the biscuit
initial thickness
value. This closer to the
improvement
three production process phases. An XmR (Shewhart’s Control Chart; individuals and
nominal
was value. stabilizing the process and adjusting the biscuit thickness closer to the
achieved
moving range)bycontrol chart was chosen to assess biscuit thickness stability, enabling var-
nominal
iability value.
3.3. Dimensional Stabilityidentifying
monitoring and Assessment the causesThickness
of Biscuit of fluctuations. Control chart analysis in-
volves Aplotting
preliminarya selected statistic
statistical for aofquality
analysis characteristic
the biscuit by examining identified
thickness measurements values against
three
3.3. Dimensional
control lines overStability
time Assessment
[40]. of Biscuit Thickness
production process phases. An XmR (Shewhart’s Control Chart; individuals and moving
A preliminary
range) control chartstatistical
was chosen analysis of the
to assess biscuit
biscuit thickness
thickness measurements
stability, enabling identified
variability
three production process phases. An XmR (Shewhart’s Control
monitoring and identifying the causes of fluctuations. Control chart analysis Chart; individuals and
involves
moving range) control chart was chosen to assess biscuit thickness stability,
plotting a selected statistic for a quality characteristic by examining values against control enabling var-
iability monitoring
lines over time [40].and identifying the causes of fluctuations. Control chart analysis in-
volves plotting a selected statistic for a quality characteristic by examining values against
control lines over time [40].
 Materials 2024, 17, x FOR PEER REVIEW 18 of 23
Materials 2024, 17, x FOR PEER REVIEW 18 of 23
Materials 2024, 17, 5935 18 of 23

Given the identified process phases, separate control charts are recommended for
Given the identified process phases, separate control charts are recommended for
each,Given
allowing precise stability
the identified process assessments
phases, separate for control
each phase.
chartsAn
areXmR control chart
recommended was
for each,
each, allowing precise stability assessments for each phase. An XmR control chart was
chosen due
allowing to thestability
precise data characteristics
assessments(individual
for each phase.measurements).
An XmR control chart was chosen
chosen due to thecharacteristics
data characteristics (individual measurements).
due toFigure
the data22 uses a box-and-whisker plot measurements).
(individual to analyze the variability in each phase, showing
Figure 22 uses a box-and-whisker plot to analyze the
thevariability inineach phase, showing
a gradual
Figure stabilization of biscuit thickness.
22 uses a box-and-whisker plotPhase one displays
to analyze significant
variability outliers,
each phase,indicating
showing
aagradual stabilization of biscuit thickness. Phase one displays significant outliers, indicating
substantial thickness dispersion caused by clogged transport channels and metal feed
gradual stabilization of biscuit thickness. Phase one displays significant outliers, vari-
indicat-
substantial thickness dispersion caused by clogged transport channels and metal feed vari-
ability. Phase two shows greater stability, although a new source of variability—mold seal-
ing substantial thickness dispersion caused by clogged transport channels and metal feed
ability. Phase two shows greater stability, although a new source of variability—mold seal-
ing issues—arose.
variability. After
Phase two correcting
shows greaterthis issue, the
stability, processaachieved
although new sourceits highest stability.
of variability—mold
ing issues—arose.
sealing issues—arose.After After
correcting this issue,
correcting the process
this issue, achieved
the process its highest
achieved stability.
its highest stability.
Box Plot of HPDC Biscuit thickness grouped by Stage
36 Box Plot of HPDC Biscuit thickness grouped by Stage
36
34
34
32
32
30
30
HPDC Biscuit thickness

28
HPDC Biscuit thickness

28
26
26
24
24
22
22
20
20
18
18
16
16 Median
Median
25%-75%
25% − 75%
14 25%-75%
25% − 75% Range
14 Non-Outlier
1 2 3 Non-Outlier
1 2 3 Outliers Range
Stage Outliers
Extremes
Stage Extremes

Figure 22.
Figure 22. Box
Box Plot
Plot of
of biscuits
biscuits grouped
grouped by
by stage.
stage.
Figure 22. Box Plot of biscuits grouped by stage.

XmR control
XmR control charts
charts are
are shown
shown in
in Figures
Figures 23–25.
23–25.
XmR control charts are shown in Figures 23–25.

Figure 23.
Figure 23. X
X and
and Moving
Moving R
R Chart—stage
Chart—stage 1.
1.
Figure 23. X and Moving R Chart—stage 1.
Materials 2024, 17, x FOR PEER REVIEW 19 of 23
Materials 2024,17,
Materials2024, 17,x5935
FOR PEER REVIEW 19 of 23
19 of 23

Figure
Figure24.
24.XXXand
andMoving
MovingRRRChart—stage
Chart—stage2.
Figure 24. and Moving Chart—stage 2.
2.

Figure
Figure25.
25.XXand
andMoving
MovingRRChart—stage
Chart—stage3.3.
Figure 25. X and Moving R Chart—stage 3.
The data in Figure 23 reveal high biscuit thickness variability, especially during the
The data in Figure 23 reveal high biscuit thickness variability, especially during the
The
initial data in Figure
production phase.23Both reveal high
lower biscuit
(LCL) andthickness variability,
upper control limits especially
(UCL) were during the
breached,
initial production phase. Both lower (LCL) and upper control limits (UCL) were breached,
initial productionmeasurements
with individual phase. Both lower (LCL)
nearing USLand
andupper
LSL control limits (UCL)
(Upper/Lower were breached,
Specification Limits).
with individual measurements nearing USL and LSL (Upper/Lower Specification Limits).
with individual
Notably, valuesmeasurements
below the LCLnearing USL andthe
could increase LSL (Upper/Lower
porosity Specification
of the castings. Limits).
The X-chart’s
Notably, values below the LCL could increase the porosity of the castings. The X-chart’s
Notably,
center line values below athe
(CL) shows LCL could
significant gapincrease
betweenthetheporosity of the castings.
mean thickness The X-chart’s
and the nominal value,
center line (CL) shows a significant gap between the mean thickness and the nominal
center line (CL) shows a significant gap between the mean thickness
resulting from contaminants in the pouring channel. This problem escalated, prompting and the nominal
value, resulting from contaminants in the pouring channel. This problem escalated,
value,
operator resulting from and
intervention contaminants
a process haltin to
theclear
pouring channel. This problem escalated,
the channel.
prompting operator intervention and a process halt to clear the channel.
prompting operator
Subsequent intervention
stages saw process and stabilization.
a process haltA tomoving
clear theaverage
channel.curve on the chart
Subsequent stages saw process stabilization. A moving average curve on the chart
Subsequent
allows for long-termstages sawobservation,
trend process stabilization.
revealing A moving decrease
a gradual average in curve on the
biscuit chart
thickness
allows for long-term trend observation, revealing a gradual decrease in biscuit thickness
allows
due to for long-term
channel trend observation, revealing a gradual decrease in biscuit thickness
clogging.
due toFigure
channel 24 clogging.
due to channel clogging. the production process after unclogging the transport pipe. The
illustrates
Figure 24 illustrates the production process after unclogging
a clear the transport pipe. The
Figure 24 illustrates
moving average analysis the production
indicates process after
stabilization unclogging
without the transport
trend; however,pipe. The
biscuit
moving average analysis indicates
both stabilization
nominal and without a clear trend; however,
furtherbiscuit
moving
thickness average
variationsanalysis
exceedingindicates stabilization without a clear
undersized valuestrend; however,
require biscuit
investi-
thickness variations exceeding likelyboth nominal
from and undersized values require further
thickness
gation. These variations exceeding
fluctuations both nominal
resulted and undersized
intermittent pouringvalues require
channel further
cleaning by
Materials 2024, 17, 5935 20 of 23

the operator, which was performed without stopping the machine. Toward the end of this
stage, a concerning downward trend and increased thin biscuits were noted, attributed to a
leak in the movable mold element—the slider.
Loss of mold sealing causes flash, the thickness and spread of which depend on the
leak severity. The molten metal intended for the part flows into the mold seams, creating a
flash that hinders casting and may cause dimensional nonconformities. Significant flash
along mold seams reduces cavity fill, requiring the piston to travel further during dosing,
thus reducing the biscuit thickness. The observed trends suggest that reduced mold sealing
is directly correlated with decreased biscuit thickness during repeated dosing.
The third stage began once the mold sealing was restored. Maintenance records
indicate occasional issues stemming from furnace malfunctions (heater element replacement
and channel cleaning) (Figure 25). Control charts reflect significant production parameter
stabilization, affirming that corrective actions achieved the desired outcome. However, the
downward trend in the final phase requires further analysis to prevent future deviations.
The causes of the intervention are the subject of ongoing analysis. The phenomenon
of trough clogging is mainly due to the direct contact between the liquid metal and air,
resulting in the formation of oxides that block the flow in the trough. This results in a
reduction in metal throughput and increased variability in biscuit thickness. As part of the
optimization efforts, projects are being implemented to minimize the contact area between
the molten metal and the air. In addition, the problem of mold sealing and leakage can
be a consequence of suboptimal mold design, which is also an area of ongoing research
and development. The result of this work is, among other things, the identification of the
causes of their occurrence.
Gluing of the casting mold is most often the result of a reproducing component that is
too hot. The causes of this problem may include the following:
• lack of a cooling system (design error),
• ineffective cooling resulting from
✓ insufficient cooling range,
✓ inadequate design of cooling channels (e.g., too small diameter),
✓ use of unsuitable cooling medium,
✓ clogging of existing cooling channels during operation.
In the case of mold leaks, e.g., on sliders, the problem is not always related to insuffi-
cient mold short-circuiting. Possible causes are:
• too high a post-pressure or too high a pressure in the mold cavity,
• wrong slide fit,
• insufficient force generated by the slider actuators,
• too high a melt temperature,
• too early, a switch from phase II to phase III injection.
The above problems can be taken into account when designing a new tool to improve
the performance of the mould and increase its productivity and service life.

4. Discussion
The conducted studies align with the existing literature, demonstrating that biscuit
thickness—and consequently the volume of poured metal—significantly affects casting
porosity. Each analysis confirmed that reducing biscuit thickness increases casting poros-
ity. This study aimed not only to describe the impact of biscuit height on porosity but
also to analyze the correlation between biscuit thickness and porosity, enabling quality
improvement by predicting porosity levels based on control chart trends for biscuit thick-
ness. A standard control chart from the SPC system was employed to monitor process
stability, allowing early detection and response to deviations. Dividing the process into
phases (considering incidents) permitted a more detailed statistical analysis for each phase.
However, whole-process analysis shows that the results align with phase-based analyses,
including those based on X-ray images. Both types of analyses complement each other,
Materials 2024, 17, 5935 21 of 23

providing comprehensive tools for continuous quality improvement in foundry production

and defect reduction. Reducing the variability of biscuit thickness is very important for a
company’s quality policy. This phenomenon is related to the fact that greater variability
in process parameters, such as material dosage, as seen, for example, in the variation of
biscuit thickness, leads to an increased number of non-conforming products. A mold is
characterized by a certain service life, expressed, for example, as 100,000 injection cycles,
which translates into 200,000 castings. With low variability in biscuit thickness, lower scrap
rates can be expected, allowing more full-value parts to be produced and sold from a single
mold. However, the high variability in process parameters results in more waste (scrap),
which reduces the number of saleable parts. Process stability, measured by controlling
biscuit thickness variability, therefore, has a direct impact on mold utilization efficiency,
reducing production costs, and increasing profitability.

5. Conclusions
In summary, the study findings lead to the following conclusions.
1. The high-pressure casting under study has high technological complexity and is prone
to porosity in critical areas, affecting the quality and CNC processing outcomes.
2. X-ray analyses confirm a strong porosity dependence on the biscuit height, especially
when deviating from the design specifications.
3. Precise dosing of molten metal into the shot chamber is crucial for reducing porosity.
4. Control charts enable porosity prediction in the final castings, thereby enhancing
foundry production quality.
5. Control charts also allow the early detection of process incidents affecting porosity,
supporting high product quality.
6. While automatic pouring ensures high repeatability, it does not eliminate the risk of
tolerance deviations, necessitating process safeguards.
7. Dosing fluctuations lead to variable chamber fill, potentially resulting in suboptimal
metal volume despite meeting the technological requirements, which may cause air
occlusion in the shot chamber.
8. Ongoing adjustments to technological parameters must align with the current state of
the pressure die-casting machine to optimize the production process.

Author Contributions: Conceptualization, M.B.; methodology, M.B. and P.F.; validation, M.Ł., K.K.
and P.F.; formal analysis, M.B.; investigation, J.W.; resources, J.W.; data curation, J.W.; writing—original
draft preparation, M.B. and J.W.; writing—review and editing, M.Ł. and K.K.; visualization, M.B. and
K.K.; supervision, M.B. and A.P.; project administration, M.B. and A.P.; funding acquisition, M.Ł and
J.W. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by the Polish Ministry of Science and Higher Education (Grant
Number 16.16.170.654/B407).
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: The original contributions presented in this study are included in the
article. Further inquiries can be directed to the corresponding author.
Conflicts of Interest: The authors declare no conflicts of interest.

References
1. Kridli, G.T.; Friedman, P.A.; Boileau, J.M. Chapter 7—Manufacturing Processes for Light Alloys. In Materials, Design and
Manufacturing for Lightweight Vehicles, 2nd ed.; Updated by J.M. Boileau and P.K. Mallick; Mallick, P.K., Ed.; Woodhead Publishing
in Materials; Woodhead Publishing: Sawston, UK, 2021; pp. 267–320. ISBN 978-0-12-818712-8.
2. Burggräf, P.; Bergweiler, G.; Kehrer, S.; Krawczyk, T.; Fiedler, F. Mega-Casting in the Automotive Production System: Expert
Interview-Based Impact Analysis of Large-Format Aluminium High-Pressure Die-Casting (HPDC) on the Vehicle Production.
J. Manuf. Process 2024, 124, 918–935. [CrossRef]
Materials 2024, 17, 5935 22 of 23

3. Brzeziński, M.; Łucarz, M.; Trela, A.; Pribulova, A.; Futáš, P. The Effect of Vacuum Forming on the Quality of Refractory Materials.
Materials 2023, 16, 7260. [CrossRef] [PubMed]
4. Riedel, E.; Ahmed, M.; Hellmann, B.; Horn, I. Foundry 4.0: Digitally Recordable Casting Ladle for the Application of Industry
4.0-Ready Manual Casting Processes. Procedia CIRP 2023, 116, 95–100. [CrossRef]
5. Josan, A.; Pinca-Bretotean, C.; Ratiu, S. Management of the Regeneration Process of the Moulding Mixtures in Order to Reduce
the Costs of the Foundry Type Industrial Enterprises. Mater. Today Proc. 2021, 45, 4161–4165. [CrossRef]
6. Casarotto, F.; Franke, A.J.; Franke, R. 6—High-Pressure Die-Cast (HPDC) Aluminium Alloys for Automotive Applications.
In Advanced Materials in Automotive Engineering; Rowe, J., Ed.; Woodhead Publishing: Sawston, UK, 2012; pp. 109–149.
ISBN 978-1-84569-561-3.
7. US Census Bureau. Non-Ferrous Metal Die-Casting Foundries Revenue in the U.S. 2012–2022 (in Million U.S. Dollars); US Census
Bureau: Washington, DC, USA, 2023.
8. Campbell, J. Complete Casting Handbook: Metal Casting Processes, Metallurgy, Techniques and Design; Butterworth-Heinemann:
Oxford, UK, 2015.
9. Roos, H.J.; Lagler, M.; Quintana, L. The Future of Structural Components in HPDC; Bühler AG: Uzwil, Switzerland, 2019.
10. ROLAND; BERGER; GMBH. High-Pressure Die Casting for Large Auto Parts A Game-Changer for the Industry? Roland Berger:
Munich, Germany, 2023.
11. Abdu, Y.A.; Shafii, T.M.; Dubey KKGupta, U.K. Design and Analysis of Pressure Die Casting Die for Automobile Component.
Glob. J. Res. Eng. 2016, 16, 1–8.
12. Automotive Aluminum Parts High-Pressure Die Casting (HPDC) Market Size & Share Analysis—Growth Trends & Forecasts
(2024–2029). Available online: https://www.mordorintelligence.com/industry-reports/automotive-aluminum-parts-high-
pressure-die-casting-hpdc-market (accessed on 10 October 2024).
13. Peter, I.; Rosso, M. Focus on Development of Quality, High Pressure Die Casting Process. Arch. Metall. Mater. 2019, 64, 125–128.
[CrossRef]
14. Jung, J.G.; Ahn, T.Y.; Cho, Y.H.; Kim, S.H.; Lee, J.M. Synergistic Effect of Ultrasonic Melt Treatment and Fast Cooling on the
Refinement of Primary Si in a Hypereutectic Al–Si Alloy. Acta Mater. 2018, 144, 31–40. [CrossRef]
15. Cleary, P.W.; Ha, J.; Prakash, M.; Nguyen, T. 3D SPH Flow Predictions and Validation for High Pressure Die Casting of Automotive
Components. Appl. Math. Model. 2006, 30, 1406–1427. [CrossRef]
16. Carnegie, C.R.; Grigas, A.; Taylor, S.; Bamber, K.; Hughes, D.J. Evaluating the Joinability of Thin-Walled High Pressure Die Cast
Aluminium for Automotive Structures Using Self-Piercing Rivets. J. Adv. Join. Process. 2024, 9, 100222. [CrossRef]
17. Niu, Z.; Liu, G.; Li, T.; Ji, S. Effect of High Pressure Die Casting on the Castability, Defects and Mechanical Properties of Aluminium
Alloys in Extra-Large Thin-Wall Castings. J. Mater. Process Technol. 2022, 303, 117525. [CrossRef]
18. Murray, M.T.; Murray, M. High Pressure Die Casting of Aluminium and Its Alloys. In Fundamentals of Aluminium Metallurgy:
Production, Processing and Applications; Elsevier: Amsterdam, The Netherlands, 2011; pp. 217–261. [CrossRef]
19. Butler, W.A. Die Casting (Permanent Mold). In Encyclopedia of Materials: Science and Technology; Elsevier: Amsterdam, The
Netherlands, 2001; pp. 2147–2152. [CrossRef]
20. Anglada, E.; Meléndez, A.; Vicario, I.; Idoiaga, J.K.; Mugarza, A.; Arratibel, E. Prediction and Validation of Shape Distortions in
the Simulation of High Pressure Die Casting. J. Manuf. Process 2018, 33, 228–237. [CrossRef]
21. Kozłowski, J.; Jakimiuk, M.; Rogalewicz, M.; Sika, R.; Hajkowski, J. Analysis and Control of High-Pressure Die-Casting Process
Parameters with Use of Data Mining Tools; Springer: Berlin/Heidelberg, Germany, 2019; pp. 253–267.
22. Lordan, E.; Zhang, Y.; Dou, K.; Jacot, A.; Tzileroglou, C.; Wang, S.; Wang, Y.; Patel, J.; Lazaro-Nebreda, J.; Zhou, X.; et al.
High-Pressure Die Casting: A Review of Progress from the EPSRC Future LiME Hub. Metals 2022, 12, 1575. [CrossRef]
23. Ji, S.; Yang, H.; Watson, D.; Fan, Z. Weibull Analysis for the Repeatibility of Die Castings Made by an Al-Mg-Si-Mn Alloy. In Light
Metals 2016; Williams, E., Ed.; Springer International Publishing: Cham, Switzerland, 2016; pp. 681–685. ISBN 978-3-319-48251-4.
24. Dong, X.; Zhu, X.; Ji, S. Effect of Super Vacuum Assisted High Pressure Die Casting on the Repeatability of Mechanical Properties
of Al-Si-Mg-Mn Die-Cast Alloys. J. Mater. Process Technol. 2019, 266, 105–113. [CrossRef]
25. Brzeziński, M. Application of I/MR and CUSUM Control Charts to Evaluate the Quality of Cast Steel in Induction Furnaces.
Arch. Metall. Mater. 2018, 63, 99–104. [CrossRef]
26. Hsu, Q.C.; Do, A.T. Minimum Porosity Formation in Pressure Die Casting by Taguchi Method. Math. Probl. Eng. 2013, 2013,
920865. [CrossRef]
27. Brzeziński, M.; Wiśniowski, J. Effect of Cavitation Phenomenon on the Quality of High-Pressure Aluminium Alloy Castings.
J. Cast. Mater. Eng. 2023, 7, 27–30. [CrossRef]
28. Lordan, E.; Zhang, Y.; Dou, K.; Jacot, A.; Tzileroglou, C.; Wang, S.; Wang, Y.; Patel, J.; Lazaro-Nebreda, J.; Zhou, X.; et al. High-
Pressure Die Casting Processing Parameters and Die Design|Encyclopedia High-Pressure Die Casting Processing Parameters
and Die Design. Available online: https://encyclopedia.pub/entry/30691 (accessed on 10 October 2024).
29. Zhang, Y.; Patel, J.B.; Lazaro-Nebreda, J.; Fan, Z. Improved Defect Control and Mechanical Property Variation in High-Pressure
Die Casting of A380 Alloy by High Shear Melt Conditioning. JOM 2018, 70, 2726–2730. [CrossRef]
30. Huang, J.; Meng, Y.; Liu, F.; Liu, C.; Li, H. Modeling and Predicting Inventory Variation for Multistage Steel Production Processes
Based on a New Spatio-Temporal Markov Model. Comput. Ind. Eng. 2022, 164, 107854. [CrossRef]
Materials 2024, 17, 5935 23 of 23

31. Zhang, Y.; Lordan, E.; Dou, K.; Wang, S.; Fan, Z. Influence of Porosity Characteristics on the Variability in Mechanical Properties
of High Pressure Die Casting (HPDC) AlSi7MgMn Alloys. J. Manuf. Process 2020, 56, 500–509. [CrossRef]
32. Gašpár, Š.; Majerník, J.; Kolínský, J. Analysis of Causes of Porosity Change of Castings under the Influence of Variable Biscuit
Height in the Filling Chamber. Materials 2021, 14, 6827. [CrossRef]
33. Mednanský, M.; Bruna, M.; Matejka, M. Electromotor Housing Microstructure Dependence on HPDC Process Parameters.
In Transportation Research Procedia; Elsevier B.V.: Amsterdam, The Netherlands, 2023; Volume 74, pp. 680–687.
34. Kowalczyk, W.; Dańko, R.; Górny, M.; Kawalec, M.; Burbelko, A. Influence of High-Pressure Die Casting Parameters on the
Cooling Rate and the Structure of EN-AC 46000 Alloy. Materials 2022, 15, 5702. [CrossRef] [PubMed]
35. ASTM E505-22; Standard Reference Radiographs for Inspection of Aluminum and Magnesium Die Castings. ASTM International:
West Conshohocken, PA, USA, 2020.
36. Wang, B.S.; Xiong, S.M. Effects of Shot Speed and Biscuit Thickness on Externaly Solidified Crystals of High-Pressure Diet Cast
AM60B Magnesium Alloy. Trans. Nonferrous Met. Soc. China 2011, 21, 767–772. [CrossRef]
37. Prasad, D.; Srinivasa Rao, P. Defect Analysis and Remedies in the High Pressure Diecasting Process with ADC-12 Alloy—A
Technical Review. Am. J. Multidiscip. Res. Dev. 2022, 4, 1–8.
38. Sharma, B. Biscuit Thickness Validation (HPDC). Int. J. Res. Appl. Sci. Eng. Technol. 2021, 9, 1987–1992. [CrossRef]
39. European Norms EN 1706:2020+A1:2021; Aluminium and Aluminium Alloys—Castings—Chemical Composition and Mechanical
Properties. British Standard: London, UK, 2021. Available online: https://webstore.ansi.org/preview-pages/BSI/preview_3042
6403.pdf (accessed on 10 October 2024).
40. Adam Hamrol Quality Management and Engineering. With a Look at Reality 4.0, 1st ed.; PWN Scientific Publishing House:
Warsaw, Poland, 2022. (In Polish)

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.