ArticlePDF Available

Causes of Desktop FDM Fabrication Failures in an Open Studio Environment

DOI:10.1016/j.procir.2018.12.007
Authors:
Ruoyu Song
Ruoyu Song
Ruoyu Song
  • This person is not on ResearchGate, or hasn't claimed this research yet.
Cassandra Telenko at Georgia Institute of Technology
Cassandra Telenko
Download full-text PDF
Read full-text
Download citation

Abstract and Figures

The usage phase often dominates the lifetime environmental impact of energy and material consuming products. Improper human behaviors could increase the environmental impacts during the use stage of the product’s life cycle. The use phase of desktop fused deposition modeling (FDM) printers involves large energy and material consumption. In university makerspaces, many novices choose to use desktop FDM printers because of its easy operation, but novices often make improper design and operation decisions. The human errors increase the fabrication failures leading to higher material and energy consumption. Therefore, human behaviors should be investigated in order to identify the causes of fabrication failures in an open studio environment. Three types of failure causes are studied, which are designer error, operator error and machine error. In this research, computer-aided design (CAD) models, printing settings, user experience information and printing results were tracked and analyzed. In addition, the energy and material consumption were recorded from the printings. These data are analyzed to identify the key factors affecting printing failure rates and to improve the sustainability and efficiency of desktop FDM. From the collected data, a failure rate of 41.1% was observed. The failures caused by human error accounted for 26.3% of the total prints, which shows that human behaviors could influence the environmental impacts of FDM. For the factors impacting failure rates, user’s experience level and printing parameters were analyzed. We found that that experience did not result in higher expertise or lower failure rates.
Figures - available via license: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
Content may be subject to copyright.
Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
ScienceDirect
Available online at www.sciencedirect.com
Available online at www.sciencedirect.com
ScienceDirect
Procedia CIRP 00 (2017) 000–000
www.elsevier.com/locate/procedia
2212-8271 © 2017 The Authors. Published by Elsevier B.V.
Peer-review under responsibility of the scientific committee of the 28th C IRP Design Conference 2018.
28th CIRP Design Conference, May 2018, Nantes, France
A new methodology to analyze the functional and physical architecture of
existing products for an assembly oriented product family identification
Paul Stief *, Jean-Yves Dantan, Alain Etienne, Ali Siadat
École Nationale Supérieure d’Arts et Métiers, Arts et Métiers ParisTech, LCFC EA 4495, 4 Rue Augustin Fresnel, Metz 57078, France
* Corresponding author. Tel.: +33 3 87 37 54 30; E-mail address: paul.stief@ensam.eu
Abstract
In today’s business environment, the trend towards more product variety and customization is unbroken. Due to this development, the need of
agile and reconfigurable production systems emerged to cope with various products and product families. To design and optimize production
systems as well as to choose the optimal product matches, product analysis methods are needed. Indeed, most of the known methods aim to
analyze a product or one product family on the physical level. Different product families, however, may differ largely in terms of the number and
nature of components. This fact impedes an efficient comparison and choice of appropriate product family combinations for the production
system. A new methodology is proposed to analyze existing products in view of their functional and physical architecture. The aim is to cluster
these products in new assembly oriented product families for the optimization of existing assembly lines and the creation of future reconfigurable
assembly systems. Based on Datum Flow Chain, the physical structure of the products is analyzed. Functional subassemblies are identified, and
a functional analysis is performed. Moreover, a hybrid functional and physical architecture graph (HyFPAG) is the output which depicts the
similarity between product families by providing design support to both, production system planners and product designers. An illustrative
example of a nail-clipper is used to explain the proposed methodology. An industrial case study on two product families of steering columns of
thyssenkrupp Presta France is then carried out to give a first industrial evaluation of the proposed approach.
© 2017 The Authors. Published by Elsevier B.V.
Peer-review under responsibility of the scientific committee of the 28th CIRP Design Conference 2018.
Keywords: Assembly; Design method; Family identification
1. Introduction
Due to the fast development in the domain of
communication and an ongoing trend of digitization and
digitalization, manufacturing enterprises are facing important
challenges in today’s market environments: a continuing
tendency towards reduction of product development times and
shortened product lifecycles. In addition, there is an increasing
demand of customization, being at the same time in a global
competition with competitors all over the world. This trend,
which is inducing the development from macro to micro
markets, results in diminished lot sizes due to augmenting
product varieties (high-volume to low-volume production) [1].
To cope with this augmenting variety as well as to be able to
identify possible optimization potentials in the existing
production system, it is important to have a precise knowledge
of the product range and characteristics manufactured and/or
assembled in this system. In this context, the main challenge in
modelling and analysis is now not only to cope with single
products, a limited product range or existing product families,
but also to be able to analyze and to compare products to define
new product families. It can be observed that classical existing
product families are regrouped in function of clients or features.
However, assembly oriented product families are hardly to find.
On the product family level, products differ mainly in two
main characteristics: (i) the number of components and (ii) the
type of components (e.g. mechanical, electrical, electronical).
Classical methodologies considering mainly single products
or solitary, already existing product families analyze the
product structure on a physical level (components level) which
causes difficulties regarding an efficient definition and
comparison of different product families. Addressing this
Procedia CIRP 80 (2019) 494–499
2212-8271 © 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/)
Peer-review under responsibility of the scientific committee of the 26th CIRP Life Cycle Engineering (LCE) Conference.
10.1016/j.procir.2018.12.007
© 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/)
Peer-review under responsibility of the scientic committee of the 26th CIRP Life Cycle Engineering (LCE) Conference.
Available online at www.sciencedirect.com
ScienceDirect
Procedia CIRP 00 (2018) 000–000
www.elsevier.com/locate/procedia
2212-8271 © 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Peer-review under responsibility of the scientific committee of the 26th CIRP Life Cycle Engineering (LCE) Conference.
doi:10.1016/j.procir.2017.04.009
26th CIRP Life Cycle Engineering (LCE) Conference
Causes of Desktop FDM Fabrication Failures in an Open Studio
Environment
Ruoyu Song, Cassandra Telenko*
George W. Woodruff School of Mechanical Engineering, Georgia Insitute of Technology, Atlanta GA 30332, USA
* Corresponding author. Tel.: +1-404-385-3801. E-mail address: cassandra@gatech.edu
Abstract
The usage phase often dominates the lifetime environmental impact of energy and material consuming products. Improper human behaviors
could increase the environmental impacts during the use stage of the product’s life cycle. The use phase of desktop fused deposition modeling
(FDM) printers involves large energy and material consumption. In university makerspaces, many novices choose to use desktop FDM printers
because of its easy operation, but novices often make improper design and operation decisions. The human errors increase the fabrication
failures leading to higher material and energy consumption. Therefore, human behaviors should be investigated in order to identify the causes
of fabrication failures in an open studio environment. Three types of failure causes are studied, which are designer error, operator error and
machine error. In this research, computer-aided design (CAD) models, printing settings, user experience information and printing results were
tracked and analyzed. In addition, the energy and material consumption were recorded from the printings. These data are analyzed to identify
the key factors affecting printing failure rates and to improve the sustainability and efficiency of desktop FDM. From the collected data, a
failure rate of 41.1% was observed. The failures caused by human error accounted for 26.3% of the total prints, which shows that human
behaviors could influence the environmental impacts of FDM. For the factors impacting failure rates, user’s experience level and printing
parameters were analyzed. We found that that experience did not result in higher expertise or lower failure rates.
© 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Peer-review under responsibility of the scientific committee of the 26th CIRP Life Cycle Engineering (LCE) Conference.
Keywords: Additive manufacturing; fused deposition modeling; failure analysis, makerspaces, building cost
1. Introduction
Nowadays, people can engage in product design and
manufacturing easily because of the development of additive
manufacturing (AM). Compared to traditional manufacturing
processes, AM is more easily implemented in a home
environment and can minimize material waste, reduce
shipping costs and make parts more efficiently in a small
batch [1,2]. Therefore, AM machines have become popular in
makerspaces in universities and schools. However, the
environmental impacts of AM could increase dramatically due
to inexperienced users and malfunctions of inexpensive
machines.
Fused deposition modeling (FDM) is the most widespread
AM technologies in university makerspaces. Barrett et al.
found that desktop FDM machines such as MakerBots, are the
most common piece of equipment by studying 40
makerspaces that were identified from 127 top undergraduate
institutes in United States [3]. The FDM machine produces
parts by extruding molten material to form layers as the
material hardens. Desktop FDM printers are popular because
of compact sizes, affordable prices and low maintenance
costs. Desktop FDM printers are also novice-friendly because
of its easy operation. With free open-source slicer software
tools such as Cura, Slic3r and Repetier, the users can upload
their 3D files, set the printing parameters and print parts
easily.
Available online at www.sciencedirect.com
ScienceDirect
Procedia CIRP 00 (2018) 000–000
www.elsevier.com/locate/procedia
2212-8271 © 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Peer-review under responsibility of the scientific committee of the 26th CIRP Life Cycle Engineering (LCE) Conference.
doi:10.1016/j.procir.2017.04.009
26th CIRP Life Cycle Engineering (LCE) Conference
Causes of Desktop FDM Fabrication Failures in an Open Studio
Environment
Ruoyu Song, Cassandra Telenko*
George W. Woodruff School of Mechanical Engineering, Georgia Insitute of Technology, Atlanta GA 30332, USA
* Corresponding author. Tel.: +1-404-385-3801. E-mail address: cassandra@gatech.edu
Abstract
The usage phase often dominates the lifetime environmental impact of energy and material consuming products. Improper human behaviors
could increase the environmental impacts during the use stage of the product’s life cycle. The use phase of desktop fused deposition modeling
(FDM) printers involves large energy and material consumption. In university makerspaces, many novices choose to use desktop FDM printers
because of its easy operation, but novices often make improper design and operation decisions. The human errors increase the fabrication
failures leading to higher material and energy consumption. Therefore, human behaviors should be investigated in order to identify the causes
of fabrication failures in an open studio environment. Three types of failure causes are studied, which are designer error, operator error and
machine error. In this research, computer-aided design (CAD) models, printing settings, user experience information and printing results were
tracked and analyzed. In addition, the energy and material consumption were recorded from the printings. These data are analyzed to identify
the key factors affecting printing failure rates and to improve the sustainability and efficiency of desktop FDM. From the collected data, a
failure rate of 41.1% was observed. The failures caused by human error accounted for 26.3% of the total prints, which shows that human
behaviors could influence the environmental impacts of FDM. For the factors impacting failure rates, user’s experience level and printing
parameters were analyzed. We found that that experience did not result in higher expertise or lower failure rates.
© 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Peer-review under responsibility of the scientific committee of the 26th CIRP Life Cycle Engineering (LCE) Conference.
Keywords: Additive manufacturing; fused deposition modeling; failure analysis, makerspaces, building cost
1. Introduction
Nowadays, people can engage in product design and
manufacturing easily because of the development of additive
manufacturing (AM). Compared to traditional manufacturing
processes, AM is more easily implemented in a home
environment and can minimize material waste, reduce
shipping costs and make parts more efficiently in a small
batch [1,2]. Therefore, AM machines have become popular in
makerspaces in universities and schools. However, the
environmental impacts of AM could increase dramatically due
to inexperienced users and malfunctions of inexpensive
machines.
Fused deposition modeling (FDM) is the most widespread
AM technologies in university makerspaces. Barrett et al.
found that desktop FDM machines such as MakerBots, are the
most common piece of equipment by studying 40
makerspaces that were identified from 127 top undergraduate
institutes in United States [3]. The FDM machine produces
parts by extruding molten material to form layers as the
material hardens. Desktop FDM printers are popular because
of compact sizes, affordable prices and low maintenance
costs. Desktop FDM printers are also novice-friendly because
of its easy operation. With free open-source slicer software
tools such as Cura, Slic3r and Repetier, the users can upload
their 3D files, set the printing parameters and print parts
easily.
Ruoyu Song et al. / Procedia CIRP 80 (2019) 494–499 495
2 Author name / Procedia CIRP 00 (2019) 000–000
However, the environmental impact of FDM could
significantly increase due to human and machine errors.
Novice users, the major users of desktop FDM printers, are
often inexperienced in design and operation and can make
improper decisions leading to fabrication failures. In addition,
the inexpensive desktop FDM printers are more prone to
malfunctions which can also result into fabrication failures.
Our prior studies in two different university makerspaces
found that the material waste rates are from 35% to 45% [4].
These failures can increase the life cycle energy costs by 50%
or more [5].
Such failures may be useful in education because it can
help students to better understand the structures and
constraints of problems [6]. Embracing failure has also been
identified as one of the three guiding principles for an
educational makerspace [7]. However, youth can experience
the failures of makings as demoralizing [8]. In addition, since
there are always limited number of printers in a makerspace,
failures can result in inefficiency of the makerspace operation.
Many of the printing failures could be caused by user
behaviors. User behaviors could result into uncertainty and
variability when estimating environmental impacts of FDM
printers. For example, a makerspace that does not require any
training before using the printers could lead to higher
environmental impacts [4]. Investigating user behaviors to
reduce environmental impacts of FDM printers is especially
important because FDM is expected to make AM a tool for
everyday household life [9]. According to Wohlers Report
[10], more than 278,000 desktop 3D printers were sold
worldwide in 2015. The market of desktop 3D printers further
grew by 49.4% worldwide in 2016 [11]. Therefore, FDM
printers will consume large quantities of energy and material
if its adoption reaches that of inkjet and laser printers. For
inkjet and laser printers, Kawamoto et al. estimated that the
stock of laser printers was 28 million and the stock of inkjet
printers was 74 million at the end of 1999 consuming 6.23
TWh/year and 2.88 TWh/year, respectively [12].
Few studies look into the causes of fabrication failures in
makerspaces. User experience and expertise level may
influence the possibility of fabrication failures. Cerdas et al.
found that experienced users could better select printing
parameters that minimized waste [13]. This study aims to
investigate how failure rates change with user experience and
expertise level in university makerspaces. The printing
failures and daily users of various levels of experience were
studied in an open-access university makerspace. Specifically,
user experience level, computer-aided design (CAD) models,
printing parameters and results were tracked and analyzed.
2. Literature Review
This section reviews the background of the research in two
fields, which are FDM in makerspaces and fabrication failures
of FDM.
2.1. FDM in Makerspaces
FDM printers are the most common equipment in
university makerspaces [3]. Since AM can produce parts in a
small batch efficiently [1], students in the makerspaces can
use the FDM printers to build physical prototypes easily.
Integrating AM and makerspaces enables the democratization
of design and manufacturing [14].
Combining AM and makerspaces can also assist students
learning design and manufacturing. Practicing physical
prototyping can help students gain deeper understanding of
design requirements and features [15]. Hand-on experiences
in the makerspaces can help students realize the limitations of
FDM. From a survey in Georgia Tech Invention studio, more
than 80% of students self-reported a positive impact on their
manufacturing skills [16].
However, the environmental impacts of FDM increase
significantly in this type of open studio environment due to
fabrication failures. A life cycle inventory combining material
waste and energy consumption showed that the energy
consumption of FDM in an open-access makerspaces may be
50% more than under controlled experiments [5]. With the
development of AM, people may be able to fabricate products
by themselves at home, and FDM is expected to become a
tool for everyday household life [9]. FDM printers will
consume large quantities of material and energy. Therefore,
practices should be suggested to decrease the fabrication
failures and environmental impacts of FDM in open studio
environments.
2.2. Fabrication Failures of FDM
When evaluating the environmental impacts of AM, few
studies investigated the effects of fabrication failures. Most
studies focused on energy usage in successful and controlled
printing scenarios [1,17,18] and how to determine the optimal
printing orientations [19–21]. Our previous work classified
nine types of AM failures by collecting failed AM prints from
a university makerspace, and categorized them into designer
error, operator error and machine error [22].
In addition, the relationships between part geometries,
printing parameters and manufacturing failures are explained
by a few studies. Seepersad et al. created a designer’s guide
for dimensioning and tolerancing selective laser sintering
(SLS) parts [23]. This research determined the limiting feature
sizes of various types of features including slits, holes, letters,
mating gears and shafts for SLS through a series of
experiments. Many other design for AM guidelines were
developed [24–27]. However, Booth et al. argued that some
guidelines are more specific than most novice users need or
understand, several even tend to be out of the control of many
novices [28]. They developed a design for AM worksheet for
novice and intermittent users, which can decrease the failure
rate by 81.7%.
3. Methodology
For this research, one hypothesis was tested: failure rates
decrease if users have more experience and higher education
level. We examined the type and number of FDM failures and
user demographics in a free-access university makerspace, the
Invention Studio at the Georgia Institute of Technology. The
makerspace is run by trained volunteers who are
496 Ruoyu Song et al. / Procedia CIRP 80 (2019) 494–499
Author name / Procedia CIRP 00 (2019) 000–000 3
undergraduate and graduate students. All students can use the
printers without training requirements. This study was
approved under protocol H17008 by Georgia Tech IRB.
The Invention Studio has a 3D printing makerspace with 9
Ultimaker 2+ printers in Fall 2017 semester, and 10 Ultimaker
2+ printers in Summer 2018 semester. The Ultimaker 2+
printer is a FDM printer with build volume of 223 x 223 x 305
mm and resolution of 0.02 to 0.60 mm. The printers used PLA
as the raw material.
From our previous research in this makerspace [5], we
collected failure rates every two weeks in one semester and
found that the failure rate for one time period was
significantly lower than other time periods. That time period
was spring break that the makerspace was closed to public
and only trained staff could access it and use the printers. In
addition, we found that the makerspace requires training
before using the printers lead to lower failure rates than the
makerspace does not require any training. Therefore, we
assume that user experience level may influence failure rates.
To test the hypothesis, we collected and analyzed CAD
files, failed parts and energy consumption within the
makerspace. For each CAD file, we recorded the source of the
part, if it was created by the user or downloaded from online
websites such as GrabCAD.com and Thingiverse.com.
Additionally, we observed individual students using the
printers and the decisions that they made. Users’ experience
information including education level and number of times
using CAD software and 3D printers was collected in order to
test the relationship between failure rates and users’ design
experience level, operation experience level and education
level. The following multiple-choice questions were used in
the survey:
1. For how many projects have you used CAD software?
a. <3
b. 3-5
c. 6-10
d. >10
2. How many different parts have you printed on a 3D
printer?
a. This is the first time
b. <5
c. 6-10
d. More than 10
3. How often do you use 3D printers in the Invention
Studio?
a. <4 times/semester
b. 1-3 times/month
c. 1-3 times/week
d. >3 times/week
4. Which year of study are you in?
a. Freshman
b. Sophomore
c. Junior
d. Senior
e. Masters
f. MS/PhD
g. PhD
The printing parameters set in the slicer software were
recorded including layer height, infill density, print speed and
support material settings. The slicer software used in the
makerspace is Ultimaker Cura. In addition, the users were not
allowed to alter the temperature of the extruder and the
printing bed.
For each printing job, we recorded if the print was a
success or a failure. We asked the users if they considered the
print a success or a failure and checked for physical defects. If
the print failed, the failure reason was recorded.
To measure the energy consumed by failed prints, HOBO
UX120-018 Plug Load Data Loggers were connected to 6
Ultimaker 2+ printers in the makerspace. The data logger
recorded the power in Watts at a sampling rate of 0.1 Hz. The
data from the loggers were exported and saved regularly.
Based on the collected information, the energy consumption
data for failed prints were extracted and recorded. The mass
of the failed prints was measured using a scale, accurate to 2g.
The collected data were analyzed to show the causes of
failures. Three sets of analysis of variance (ANOVA) were
done for the user’s experience level, the source of part and
printing settings.
4. Results
In total 95 sets of individual observations were recorded
with 39 failed prints. The overall failure rate was 41.1%. The
average energy consumption per part was 3.0 MJ. The
average mass of the failed parts was 28 g. The printing energy
intensity is 107.1 MJ/kg. Table 1 summarizes the failure
category, primary cause, detailed cause and number of failed
prints with respect to each failure cause. More explanation of
detailed causes can be found in our previous work [5].
Table 1. Failure reasons and number of prints
Category
Primary Cause
Detailed Cause
# of Prints
Designer
Support Material
Support cannot be removed
2
9
Feature Size
Complex features
3
Cannot assemble
2
Wrong part size
1
Wrong hole size
1
Operator
Printing Settings
Did not generate support
2
16
Printed out of area
1
Printer Operation
Loose calibration
8
Platform was moved
1
Printed wrong file
1
Out of filament
2
Filament tangled
1
Machine Machine
Malfunction
Skip layers
1
14
Nozzle clogged
10
Layer shift
3
Among the 39 failures, nine failures were caused by
designer errors. Two prints failed because the designers did
not consider support material removing process. Therefore,
the support material could not be removed and ruined the
Ruoyu Song et al. / Procedia CIRP 80 (2019) 494–499 497
4 Author name / Procedia CIRP 00 (2019) 000–000
surface finish. Since support material is not needed for
traditional manufacturing processes, the designers may not
consider it. Nine prints failed because of improper feature
sizes. If the designers did not consider the resolutions of the
FDM printers, they may design too complex and small
features which cannot be printed. If the designers did not
consider the tolerances of the FDM printers, the mating parts
could not be assembled. In addition, two parts failed because
the designers specified wrong feature sizes for printing.
Sixteen failures were caused by operator errors including
improper printing settings and operations. Three prints failed
because the operators did not choose to generate support
material for parts with overhang structures. From the
observations, the operators did not click the option to generate
support material in the slicer software because they did not
know what the function of support material was. One part
failed because the operator placed the part out of the printable
area of the printer in the slicer software. Eight prints failed
because the printers were not calibrated properly. Therefore,
the prints could not adhere to the printing platform and
warped; the layers of the prints could not adhere to each other
either. One print failed because the printing platforms were
moved accidentally during the printing process. One print
failed because the operator uploaded or selected a wrong file
to print. Three parts failed because the operators did not check
the status of the filaments before printing. Among these three
failures, two prints failed because the remaining filament was
not enough for the parts. One print failed because the printer
was not able to extrude the tangled filament.
Fourteen failures were caused by machine errors. The
primary cause was nozzle clogging (71.4% of failures).
Nozzle clogging could be caused by incorrect temperature for
extruding, poor quality filament, tight calibration and printer
aging. To decrease the environmental impacts of failures
caused by nozzle clogging, printer should be stopped and
repaired timely. From the observations, sometimes the
printers ran without extruding any material to the end of the
programmed printing process, which wasted a large amount of
energy.
We have analyzed the material and energy loss due to each
failure cause and its uncertainty and variability in our
previous work [4,5,22]. In this paper, we analyzed the impacts
of user’s experience level and printing settings on fabrication
failures of FDM.
4.1. Impacts of User’s Experience Level
We expected that a user with higher experience level is less
likely to make failed prints. Therefore, data were analyzed to
show the relationship between user’s experience level and
failure rates. Table 2 shows the summary of the collected data,
which includes the failure rates caused by designer error,
operator error, overall failure rates and number of users for
each experience level respectively.
For designer errors, failures rates decreased as more CAD
projects have been done. The failure rates of designer errors
also decreased as more parts printed. Higher print frequency
increased the failures rates of designer errors. For the year of
study, the failure rate of Junior were four times as much as
that of senior and higher. Therefore, design experiences
gained in CAD projects, previous printed parts and knowledge
learnt from class could decrease the fabrication failures
caused by designer errors. Without such experience and
knowledge, higher print frequency cannot benefit the
designers.
Table 2. Summary of user’s experience level vs. failure rates
Category Experience Level
Failure Rates (%)
# of
Users
Designer
Error
Operator
Error Overall
Number
of CAD
projects
<3
9.5
23.8
52.4
21
3-5
9.1
13.6
36.4
22
6-10
27.3
9.1
45.5
11
>10
4.9
17.1
36.6
41
Number
of parts
printed
0
10.0
0
30.0
10
<5
14.3
14.3
32.1
28
6-10
5.9
11.8
41.2
17
>10
7.5
25.0
50.0
40
Print
frequency
<4/semester
6.4
10.8
27.0
37
1-3/month
12.5
31.3
68.8
16
1-3/week
4.5
9.1
18.2
22
>3/week
15.0
25.0
70.0
20
Year of
study
Freshman
50.0
50.0
100
2
Sophomore
0
25.0
50.0
4
Junior
43.5
21.7
34.8
23
Senior and higher
10.6
13.6
40.9
66
Not all parts printed were created by students themselves.
Some users downloaded parts designed by experts from
websites such as GrabCAD.com and Thingiverse.com. we
expected that failure rates for parts designed by experts are
lower than parts designed by novice designers. Among the 95
prints, 26 parts were CAD file downloaded from internet. The
other 69 parts were created by students themselves. We
consider the students as novice designers since they do not
have enough amount of design experiences. From calculation,
the failure rate for expert parts was 44.9%, and for novice
parts was 26.9%. Therefore, parts created by users with higher
level of design experience are less likely to fail. An ANOVA
was done for the source of part. However, the p-value is 0.22
which shows no statistical significance.
Improper part geometries could lead to fabrication failures.
When designing the parts, the designers should have ideas of
the printer specifications including resolutions and tolerances.
If the designed feature sizes are too small based on the given
printer resolutions, the features cannot be printed. If the
designers do not consider the tolerances of the printers, they
may create mating parts with same size and have risks that the
parts cannot be assembled. In addition, the design should
avoid large and flat area since they tend to warp. From the
observations, there were four failures caused by loose
calibration, but could also resulted from part geometry issues,
since all four parts had large and flat areas.
To reduce the fabrication failures caused by part geometry
issues, designers should know the printer specifications. If
498 Ruoyu Song et al. / Procedia CIRP 80 (2019) 494–499
Author name / Procedia CIRP 00 (2019) 000–000 5
possible, test parts with different geometries and feature sizes
could be printed in order to have a deeper understanding of
the printer’s capacities.
For operator errors, the failure rates did not change
significantly with the number of CAD projects done. The
number of CAD projects relates to users’ design experience,
which should not impact the operator experience. The failure
rates decreased as higher year of study. In addition, the failure
rates increased with more parts printed. The failure rates did
not change significantly with higher print frequency. These
two observations were in contrast with our expectations.
An ANOVA was done for the statistical analysis of the
experience variables impacting the failures rates. However, no
statistical significance was shown for the results. The p-value
for CAD experience is 0.20, for parts printed is 0.10, for
printing frequency is 0.28. The result for year of study is not a
full rank (rank deficiency) which means the right observations
to fit the model are not in the data.
4.2. Impacts of Printing Settings
We expected that there should be a set of optimal printing
settings when using Ultimaker 2+, which can minimize the
failure rate. To figure out the optimal settings, the layer
height, infill density, infill pattern, print speed, support
material settings and build plate adhere type were
investigated.
Fig. 1 shows the failure rates versus four different printing
parameters: layer height, infill density, print speed and
support overhang angle. The support overhang angle is the
maximum angle of overhang structure for which support
material is added. The smaller the angle is, the more the
support material is added. From the diagram, the failure rates
increased with larger layer height. The infill density did not
show obvious relationship with the failure rates. When the
print speed was at 50 mm/s, the failure rates were at the
lowest point. The failure rates increased with higher support
overhang angle.
Fig. 1. Failure Rates vs. Printing Parameters
For the infill pattern, the failure rates for Cubic was 28.6%,
for Lines was 50.0%, for Grid was 50.0%, and for Triangles
was 50.0%. Therefore, the infill pattern did not have
significant influence of the failure rates. For the build plate
adhere type, the failure rates for Brim was 21.8%, for Raft
was 100%, for Skirt was 76.9%. Therefore, to decrease the
failure rate, Brim could be chosen as the build plate adhere
type.
An ANOVA was done for the impacts of printing settings
on the failure rate. However, no statistical significance was
shown for the results. The p-value for the layer height is 0.30,
and for the support overhang angle is 0.65. The results for
infill ratio and printing speed are not full ranks which means
the right observations to fit the model are not in the data.
5. Discussion
From the three sets of ANOVA, no statistical significances
were shown. We tested the hypothesis that failure rates
decrease if users’ amount of experience increases. From our
observations, it does not seem that experience results in
effective expertise, thus alternative hypotheses are:
• Increased affordances in shops can reduce failures.
• Dedicated training for operating FDM can reduce failures.
• Design for FDM education can reduce failures.
Based on the observations in the makerspaces, users with
less operation experience tended to seek assistance from
trained staff. On the contrary, users with more printing
experiences tended to work independently. The assistance
provided by staff is a type of noise to the measurement. In the
future, the assistance should be measured in order to quantify
its impact to the failure rate.
Although the results do not show any statistical
significance when analyzing the experience and printing
factors influencing failure rates, the results do show that
human behaviors can affect the environmental impacts of
FDM. The fabrication failures caused by human errors
accounted for 26.3% of the total number of prints, which
increased the environmental impacts by around 35%. The
calculation methodology for the environmental impact is
presented in our previous work [5]. Therefore, solutions
should be provided to decrease the failures caused by human
errors. Education, training and assistance provided by
software tools could be solutions [29].
6. Conclusions
This study investigated three types of failure causes of
FDM, which are the designer error, operator error and
machine error. 95 sets of data were collected with a failure
rate of 41.1% observed. The average energy consumption per
failed part was 3.0 MJ. The average mass per part was 28 g.
The printing energy intensity is 107.1 MJ/kg. For the 39 failed
prints, nine were caused by designer errors, sixteen were
caused by operator errors, and fourteen were caused by
machine errors. The detailed failure causes are reported.
The impacts of user’s experience level and printing settings
on fabrication failures are investigated. Parts created by users
with higher level of design experience had lower rates of
failure but were not statistically significant or statistically
>
Ruoyu Song et al. / Procedia CIRP 80 (2019) 494–499 499
6 Author name / Procedia CIRP 00 (2019) 000–000
different from the general population’s failure rate. We thus
distinguish between experience and expertise. Students can
gain design expertise through CAD projects, designed parts
and knowledge learnt in class, but must increase their skill
deliberately and with adequate supporting information. For
operators without training, the failure rates did not decrease
with printing experiences. For the printing settings of
Ultimaker 2+ printers, a small layer height, a small support
overhang angle and a print speed at 50 mm/s should be
adopted to reduce failure rates.
ANOVAs were done to test the influence of user’s
experience level and printing parameters to failures rates of
FDM. However, no statistical significances were found from
the results. The types of experience we measured are not
sufficient to explain the failure rates. Therefore, more work
should be done to understand actual variables affecting the
failures rates of FDM and human behaviors. Nevertheless, the
results still demonstrate that accounting for human behaviors
is critical when estimating the environmental impact of a
products.
Acknowledgements
The authors would like to thank the users who participated
in the research and staff who permitted experiments using
their machines and space in Georgia Tech’s Invention Studio.
References
[1] Telenko C, Seepersad C. A comparison of the energy efficiency of
selective laser sintering and injection molding of nylon parts. Rapid
Prototyp J 2012;18:472–81.
[2] Huang SH, Liu P, Mokasdar A, Hou L. Additive manufacturing and
its societal impact: a literature review. Int J Adv Manuf Technol
2013;67:1191–203.
[3] Barrett TW, Pizzico MC, Levy B, Nagel RL. A Review of
University Maker Spaces. 122nd ASEE Annu. Conf. Expo., Seattle,
WA: 2015.
[4] Song R, Clemon L, Telenko C. Uncertainty and Variability of
Energy and Material Use by Fused Deposition Modeling Printers in
Makerspaces. J Ind Ecol 2018.
[5] Song R, Telenko C. Material and energy loss due to human and
machine error in commercial FDM printers. J Clean Prod
2017;148:895–904.
[6] Kapur M. Productive Failure. Cogn Instr 2008;26:379 –424.
[7] Kurti RS, Kurti DL, Fleming L. The Philosophy of Educational
Makerspaces Part 1 of Making an Educational Makerspace. Teach
Libr 2014;41:8–11.
[8] Soep E. Participatory politics: Next-generation tactics to remake
public spheres. MIT Press; 2014.
[9] Xu T, Chen Z, Li J, Yan X. Automatic tool path generation from
structuralized machining process integrated with CAD/CAPP/CAM
system. Int J Adv Manuf Technol 2015;80:1097–111.
[10] Wohlers T. Wohlers report 2016. Wohlers Associates, Inc; 2016.
[11] Wohlers T. Wohlers Report 2017. 2017.
[12] Kawamoto K, Koomey J G, Bru ce N, Brown RE, Piette MA, Ting
M, et al. Electricity used by office equipment and network
equipment in the US. Energy 2002;27:255–69.
[13] Cerdas F, Juraschek M, Thiede S, Herrmann C. Life Cycle
Assessment of 3D Printed Products in a Distributed Manufacturing
System. J Ind Ecol 2017.
[14] Huang Y, Leu MC, Mazumder J, Donmez A. Additive
Manufacturing: Current State, Future Potential, Gaps and Needs,
and Recommendations. J Manuf Sci Eng 2015;137:014001.
[15] Kim MJ, Maher M Lou. The Impact of Tangible User Interfaces on
Designers’ Spatial Cognition. Human–Computer Interact
2008;23:101–37.
[16] Forest CR, Moore RA, Jariwala AS, Fasse BB, Linsey J, Newstetter
W, et al. Advances in Engineering Education The Invention Studio:
A University Maker Space and Culture. Adv Eng Educ 2014;4:32.
[17] Baumers M, Tuck C, Wildman R, Ashcroft I, Rosamond E, Hague
R. Transparency Built-in: Energy Consumption and Cost
Estimation for Additive Manufacturing Baumers et al. Energy and
Cost Estimation for Additive Manufacturing. J Ind Ecol
2013;17:418–31.
[18] Kreiger M, Pearce JM. Environmental life cycle analysis of
distributed three-dimensional printing and conventional
manufacturing of polymer products. ACS Sustain Chem Eng
2013;1:1511–9.
[19] Cheng W, Fuh JYH, Nee AYC, Wong YS, Loh HT, Miyazawa T.
Optimization of Part- Building Orientation in Stereolithography.
Rapid Prototyp J 1995;1:12–23.
[20] Das P, Chandran R, Samant R, Anand S. Optimum Part Build
Orientation in Additive Manufacturing for Minimizing Part Errors
and Support Structures. Procedia Manuf 2015;1:343–54.
[21] Karim KF, Hazry D, Zulkifli AH, Ahmed SF, Razlan ZM, Wan K,
et al. Feature Extraction and Optimum Part Deposition Orientation
for FDM. Appl Mech Mater 2015;793:642–6.
[22] Song R, Telenko C. Material Waste of Commercial FDM Printers
under Realstic Conditions. Proc. 27th Annu. Int. Solid Free. Fabr.
Symp., Austin, Texas, USA: Laboratory for Freeform Fabrication
and University of Texas at Austin; 2016, p. 1217–29.
[23] Seepersad CC, Govett T, Kim K, Lundin M, Pinero D. A
Designer’s Guide for Dimensioning and Tolerancing SLS parts.
23rd Annu. Int. Solid Free. Fabr. Symp., Austin, TX: 2012, p. 921–
31.
[24] Becker R, Grzesiak A, Henning A. Rethink assembly design.
Assem Autom 2005;25:262–6.
[25] Atzeni E, Iuliano L, Minetola P, Salmi A. Redesign and cost
estimation of rapid manufactured plastic parts. Rapid Prototyp J
2010;16:308–17.
[26] Adam GAO, Zimmer D. Design for Additive Manufacturing—
Element transitions and aggregated structures. CIRP J Manuf Sci
Technol 2014;7:20–8.
[27] Lieneke T, Denzer V, Adam GAO, Zi mmer D. Dimensional
Tolerances for Additive Manufacturing: Experimental Investigation
for Fused Deposition Modeling. Procedia CIRP 2016;43:286–91.
[28] Booth JW, Alperovich J, Chawla P, Ma J, Reid TN, Ramani K. The
Design for Additive Manufacturing Worksheet. J Mech Des
2017;139.
[29] Song R, Telenko C. Towards Sustainable Additive Manufacturing
in University Makerspaces. Proc. Constr. 2018 Conf., Atlanta,
Georgia, USA: 2018.
... The expired FFF patent has made this process even more accessible and is still being developed and improved to this day. Despite these improvements, there is a 41.1% probability of problems that can arise during printing, 26.3% of which are human caused [4]. This can lead to an irreparable print which hast to be cancelled. ...
... Another differentiation of the causes was proposed by [4], in which a distinction was made between human and machine errors. In this study, the human error was also divided into the designer and operator error. ...
Computer Vision Based Analysis for Fused Filament Fabrication Using a G-Code Visualization Comparison
Chapter
  • Jan 2023
The first open-source and low-cost fused filament fabrication 3D printer of 2004 represented a new opportunity to manufacture mechanical components easily and affordably. 3D printers have been continuously developed since the first one entered the market and are equipped with various sensors that can monitor the printing process. Despite these improvements, a probability of 41.1% remains that the printed part will have errors. This can lead to an irreparable print that has to be canceled, but is often not noticed by common sensors, while costly time, electricity and filament continue to be consumed. This investigation provides an account of a camera based monitoring system developed to detect complex problems that are not easily recognizable with sensors commonly used for fused filament fabrication. Image segmentation was used to remove the background of the printed part and the result was compared to a visualization of the G-Code. By using an exclusive-or method it was possible to determine differences, which can indicate defects. Depending on the similarity, the printing process can be canceled promptly. Tests have demonstrated that this method works reliably even under changing lighting conditions in most cases but can lead to poor segmentation due to shadows being cast in the infill. The application is also able to recognize differences when printed parts detach or layers have shifted if they are not covered by lower layers. The use of a light source on top of the 3D printer and additional cameras, beside the build plate, could solve both problems in the future.KeywordsAdditive manufacturingFused filament fabricationImage processingProcess monitoring
View
... The stl file errors must be corrected, and the topology of the CAD model must be recreated. Ruoyu Song [12], et al., conducted a survey on the causes of desktop FDM fabrication failures in an open studio environment where the prints failed due to designer and operator error. Designer errors are improper design, wrong selection of machine for printing, tolerance and so forth. ...
... These design errors were very common in previous applications, particularly in the biomedical, 3D scanning, and complex geometry designing areas. As mentioned in previous citations [7][8][9][10][11][12], the stl file is filtered and stl errors are detected, but none of the citations mention the root cause of the error, which type of error is affecting the print, and how the errors are classified. Here we are attempting to explain the root cause of errors in different fields and classify the design errors. ...
Effect of naked and non-manifold errors on polymer printing process
Article
Full-text available
3D printing technology is rapidly evolving from small scale to large-scale industries because of the ease of achieving complex geometries with minimal time. Because none of the manufacturing processes is 100% efficient in the same way, the 3D printing process has a few drawbacks between the printing methodologies. In this study we focus on the effects of design errors on the final object. As we know majorly, we have two types of errors naked edges and non-manifold edges. These errors can occur in various ways during the design process. These design errors directly affect the printing layers, surface quality, and infill density of components. Naked edges directly affect the printing layers whereas non-manifold edges affect the infill and surface quality of the component. Because we have different slicing software’s each software has its own limits to read the design errors, but the effect of naked edges is the same in all slicing software’s and the effect of non-manifold edges is different for different software’s. Based on the type of design error, size of the error and position of the error we developed permanent and temporary solutions.
View
... A hurdle in the further adoption of desktop FFF 3D printers is their failure rate of more than 20% [4]. When a defect develops, the printer will often continue manufacturing an unusable part, wasting time and material, unless manually stopped. ...
... Section 3.2 presents the results of these tests before conclusions are drawn in Sect. 4 showing the effectiveness of the MTouch system, and future work is discussed. ...
MTouch: an automatic fault detection system for desktop FFF 3D printers using a contact sensor
Article
Full-text available
Desktop fused filament fabrication (FFF) 3D printers have been growing in popularity among hobbyist and professional users as a prototyping and low-volume manufacturing tool. One issue these printers face is the inability to determine when a defect has occurred rendering the print unusable. Several techniques have been proposed to detect such defects, but many of these approaches are tailored to one specific fault, use expensive hardware, and/or use machine learning algorithms which can be sensitive to ambient conditions making them unreliable. This paper proposes a novel, low-cost system, named MTouch, to accurately detect millimeter-scale defects that tend to make prints unusable. MTouch uses an actuated contact probe designed with a low-power solenoid, magnet, and hall effect sensor. This sensor is used to check for the presence, or absence, of the printed object at specific locations. The MTouch probe demonstrated 100% accurate readings, which was significantly higher than the 74% achieved using a repurposed commercially available bed leveling touch probe (the BLTouch). Additionally, algorithms were developed to detect common print failures such as layer shifting, bed separation, and filament runout using the MTouch probe. In head-to-head testing against a commercially available print defect detection system (The Spaghetti Detective), the MTouch was able to detect faults 44% faster on average while only increasing the print time by 8.49%. In addition, MTouch was able to detect faults The Spaghetti Detective was unable to identify such as layer shifting and filament runout/jam.
View
... It was decided to opt for this search based on recent studies that show that it is important to analyze the behavior of the operator or user by material consumable, (Song & Telenko, 2019). That is why it is important to carry out a deeper investigation of human error within the additive area and it was from various articles and from my own experience that we wanted to find out more about the opportunity of the field of study that we have. ...
... This study aimed to investigate the failure rate based on user experience and the level of experience in university creation spaces. Print failures and daily users of various experience levels were studied at an open access university (Song and Telenko, 2019). ...
STUDY OF HUMAN ERROR IN PRODUCT DESIGN FOR ADDITIVE MANUFACTURING: A SYSTEMATIC LITERATURE REVIEW
Chapter
  • Dec 2021
Currently, the industrial sector is constantly growing in response to the demand for product design processes for additive manufacturing (AM). Developments of a product design is considered important in the management of AM projects, as various actions are carried out by designers, software design managers and equipment operators. However, poor practices in the process contribute to human errors. These can generate major problems such as material and energy waste, economic losses among other effects (Song and Telenko, 2017). Cognitive Ergonomics is the branch of Ergonomics that studies the processes by which human beings process and analyze information about all aspects of human work, having as essential components decision-making, perceptual aspect, information analysis and motor skills (Cañas, 2001). Its purpose is to study the activities and mental processes on human performance in routine situations, cognitive tasks, human-machine interactions, among others, and has methods to study these processes (Benito, 2008). Among them are the methods of analysis, identification and evaluation of human error (Marchito, 2011). Thus, human error is one of the most important aspects to study within Cognitive Ergonomics. It is conceptually defined as "any set of human actions or activities that exceed some limit of acceptability, i.e., an out-of-tolerance action in which performance limits are defined by the system" (Di Pasquale et al., 2015).
View
... Of the DQ limitations, poor DA is especially regarded as a significant boundary for full acceptance [9,10]. For example, one study reported that ME AM fabrication has a failure rate of 41.1%, whereby~2 5% of total failures were due to poor DQ as a consequence of processing difficulties [11]. It is widely acknowledged throughout existing literature that dimensional quality errors are significantly influenced via printing parameter selection [12,13,14]. ...
How generalisable are material extrusion additive manufacturing parameter optimisation studies? A systematic review
Article
Full-text available
  • Nov 2022
Goal Material extrusion additive manufacturing, is a relatively inexpensive and popular manufacturing technique that can be used to fabricate complex 3D geometries at low cost. However, parts produced by this process are often characterised by poor quality, particularly with regards to dimensional and geometrical accuracy. This review provides a comprehensive analysis of experimental studies conducted over the past 25 years that have aimed to improve these quality variables via printing parameter optimisation. Methods An initial non systematic scoping study coupled with a subsequent scientific systematic literature review protocol to identify experimental studies on dimensional quality in material extrusion additive manufacturing was conducted. 127 individual studies are identified and analysed. Results The authors critically analysed the relevant and salient studies (127) by evaluating which machines; materials; sample sizes; artefact designs; and most importantly what printing parameters have been used in the experimental investigations. A total of (79) machine variations were used; ABS and PLA made up (43%) and (36%) of materials investigated respectively (84%) of studies had sample sizes of less than (40); and artefact dimensions ranged from (10–270 mm) (1–240 mm), and (3.5–220 mm) in the X, Y, and Z axes respectively. In many cases, the relationships between printing parameters (independent variables) and dimensional qualities (dependent variables) were found to be uncertain or even contradictory between studies. Conclusions A wide range of studies have sought to optimise parameters (e.g., Nozzle gap height, print head velocity, filament volumetric velocity) to address dimensional quality issues in the ME AM. However, the authors have demonstrated that a lack of agreement among studies limits the generalisability of these parameter optimisation findings. More recent studies have considered the local dimensional variance of deposited single strands. This offers greater potential to understand the underlying causes of component defects and inaccuracy.
View
... Material extrusion is the most common additive manufacturing (AM) technology due to its ease of use, low cost, and the range and combinations of materials it can print [1]. However, manufacturing errors are a frequent occurrence in extrusion AM and can limit the application of material extrusion for end-use, and particularly safetycritical, products [2][3][4][5]. Warping is one of the most prevalent errors in extrusion AM and is caused by the contraction of extruded material. Typically this occurs because the extruded thermoplastic polymer experiences a large temperature gradient that results in residual thermal stresses (see Fig. 1A) [6][7][8]. ...
Automated recognition and correction of warp deformation in extrusion additive manufacturing
Article
Full-text available
  • May 2022
Warp deformation is a common error encountered in additive manufacturing. It is typically caused by residual internal stresses in the manufactured part that arise as material cools. These errors are challenging to prevent or correct as they build over time and thus are only visible long after the actions that caused them. As a result, existing work in extrusion additive manufacturing has attempted warp detection but not correction or prevention. We report a hybrid approach combining deep learning, computer vision, and expert heuristics to correct or prevent warp. We train a deep convolutional neural network using diverse labelled images to recognise warp in real-time. We compute five metrics from detection candidates to predict the severity of warp deformation and proportionately update print settings. This enables the first demonstration of automated warp detection and correction both during printing and for future prints.
View
... A hurdle in the further adoption of desktop FFF 3D printers is their failure rate of more than 20% [3]. When a defect develops, the printer will often continue manufacturing an unusable part, wasting time and material, unless manually stopped. ...
MTouch: An Automatic Fault Detection System for Desktop FFF 3D Printers Using a Contact Sensor
Preprint
  • Dec 2021
Desktop fused filament fabrication (FFF) 3D printers have been growing in popularity among hobbyist and professional users as a prototyping and low-volume manufacturing tool. One issue these printers face is the inability to determine when a defect has occurred rendering the print unusable. Several techniques have been proposed to detect such defects but many of these approaches are tailored to one specific fault (e.g., filament runout/jam), use expensive hardware such as laser distance sensors, and/or use machine vision algorithms which are sensitive to ambient conditions, and hence can be unreliable. This paper proposes a versatile, reliable, and low-cost system, named MTouch, to detect millimeter-scale defects that tend to make prints unusable. At the core of MTouch is an actuated contact probe designed using a low-power solenoid, magnet, and hall effect sensor. This sensor is used to check for the presence, or absence, of the printed object at specific locations. The MTouch probe demonstrated 100% reliability, which was significantly higher than the 74% reliability achieved using a commercially available contact probe (the BLTouch). Additionally, an algorithm was developed to automatically detect common print failures such as layer shifting, bed separation, and filament runout using the MTouch probe. The algorithm was implemented on a Raspberry Pi mini-computer via an Octoprint plug-in. In head-to-head testing against a commercially available print defect detection system (The Spaghetti Detective), the MTouch was able to detect faults 44% faster on average while only increasing the print time by 8.49%. In addition, MTouch was able to detect faults The Spaghetti Detective was unable to identify such as layer shifting and filament runout/jam.
View
... A few industries believe AM can reduce the effect of user behavior (Schrank & Stanhope 2011;Valtas & Sun 2016) but the actual resource waste caused by machine or user behavior would be larger than expectations (Song & Telenko 2016). Notably, users whose lack experience in AM induce part failure, and environmental impact can increase by 26.3%, when caused by material and power waste (Song & Telenko 2019). They suggested that human and organizational behavior may alter the failure rate. ...
Performance of wearables and the effect of user behavior in additive manufacturing process
Article
Full-text available
  • Dec 2021
Additive manufacturing (AM) which can be a suitable technology to personalize wearables is ideal for adjusting the range of part performance such as mechanical properties if high performance is not required. However, the AM process parameter can impact overall durability and reliability of the part. In this instance, user behavior can play an essential role in performance of wearables through the settings of AM process parameter. This review discusses parameters of AM processes influenced by user behavior with respect to performance required to fabricate AM wearables. Many studies on AM are performed regardless of the process parameters or are limited to certain parameters. Therefore, it is necessary to examine how the main parameters considered in the AM process affect performance of wearables. The overall aims of this review are to achieve a greater understanding of each AM process parameter affecting performance of AM wearables and to provide requisites for the desired performance including the practice of sustainable user behavior in AM fabrication. It is discussed that AM wearables with various performance are fabricated when the user sets the parameters. In particular, we emphasize that it is necessary to develop a qualified procedure and to build a database of each AM machine about part performance to minimize the effect of user behavior.
View
Development of a Quality Gate Reference Model for FDM Processes
Chapter
Full-text available
  • Apr 2022
Additive manufacturing (AM) enables industries to accomplish mass customization by creating complex products in small batches. For this purpose, fused deposition modeling (FDM) is widely used in 3D printing where the material is applied layer-by-layer from a digital model to form a three-dimensional object. There still exist problems in FDM processes regarding the failure rate of printed parts. Failures vary from deformed geometry, clogged nozzles, and dimensional inaccuracies to small parts not being printed that may be attributed to various process steps (e.g., poor quality CAD models, converting issues, overheating, poor quality filament, etc.). The majority of these defects are preventable and are caused by imprudent try-and-error print processes and troubleshooting quality control. The aim of this chapter is to propose a quality gate reference process with defined requirement criteria to prevent the occurrence of defects. The framework shall be applied in quality control and in-situ process monitoring to enhance overall manufacturing quality.
View
A fuzzy Bayesian network-based approach for modeling and analyzing factors causing process variability
Article
Purpose In order to minimize the impact of variability on performance of the process, proper understanding of factors interdependencies and their impact on process variability (PV) is required. However, with insufficient/incomplete numerical data, it is not possible to carry out in-depth process analysis. This paper presents a qualitative framework for analyzing factors causing PV and estimating their influence on overall performance of the process. Design/methodology/approach Fuzzy analytic hierarchy process is used to evaluate the weight of each factor and Bayesian network (BN) is utilized to address the uncertainty and conditional dependencies among factors in each step of the process. The weighted values are fed into the BN for evaluating the impact of each factor to the process. A three axiom-based approach is utilized to partially validate the proposed model. Findings A case study is conducted on fused filament fabrication (FFF) process in order to demonstrate the applicability of the proposed technique. The result analysis indicates that the proposed model can determine the contribution of each factor and identify the critical factor causing variability in the FFF process. It can also helps in estimating the sigma level, one of the crucial performance measures of a process. Research limitations/implications The proposed methodology is aimed to predict the process quality qualitatively due to limited historical quantitative data. Hence, the quality metric is required to be updated with the help of empirical/field data of PV over a period of operational time. Since the proposed method is based on qualitative analysis framework, the subjectivities of judgments in estimating factor weights are involved. Though a fuzzy-based approach has been used in this paper to minimize such subjectivity, however more advanced MCDM techniques can be developed for factor weight evaluation. Practical implications As the proposed methodology uses qualitative data for analysis, it is extremely beneficial while dealing with the issue of scarcity of experimental data. Social implications The prediction of the process quality and understanding of difference between product target and achieved reliability before the product fabrication will help the process designer in correcting/modifying the processes in advance hence preventing substantial amount of losses that may happen due to rework and scraping of the products at a later stage. Originality/value This qualitative analysis will deal with the issue of data unavailability across the industry. It will help the process designer in identifying root cause of the PV problem and improving performance of the process.
View
Uncertainty and Variability of Energy and Material Use by Fused Deposition Modeling Printers in Makerspaces
Article
Full-text available
Desktop‐grade fused deposition modeling (FDM) printers are popular because of compact sizes and affordable prices. If we are moving toward a future where desktop FDM printers are in every school and office, like conventional printers, then these machines will consume a large amount of energy and material. However, it is very difficult to evaluate the environmental impacts of FDM printers since there are so many different brands and types of printers using different raw materials under different scenarios. This study uses data from two different printing sites to evaluate the scenario and parameter uncertainty and variability in energy and material balances for FDM printers. Data from the two makerspaces provide insight into the material and energy consumption data using polylactic acid and acrylonitrile butadiene styrene (ABS) with four types of printers. The use of actual performance data allowed for the additional study of scrap ratio. Regressions provide insight into predictive factors for energy and material consumption. Monte Carlo simulations show the range of energy life cycle inventory values for the desktop‐grade FDM printers. From the regressions, Type A Pro was the most energy‐intensive machine. For material waste, an open‐access makerspace using ABS was associated with higher scrap ratio. Regression analysis indicates that the rate of material usage is not a strong predictor of waste rates. The amount of waste generated across both sites indicates that more ubiquitous access to FDM printing may create a significant addition to the waste stream.
View
The Design for Additive Manufacturing Worksheet
Article
Full-text available
  • Jul 2017
Additive manufacturing (AM) technologies have become integral to modern prototyping and manufacturing. Therefore, guidelines for using AM are necessary to help users new to the technology. Many others have proposed useful guidelines, but these are rarely written in a way that is accessible to novice users. Most guidelines (1) assume the user has extensive prior knowledge of the process, (2) apply to only a few AM technologies or a very specific application, or (3) describe benefits of the technology that novices already know. In this paper, we present a one-page, visual design for additive manufacturing worksheet for novice and intermittent users which addresses common mistakes as identified by various expert machinists and additive manufacturing facilities who have worked extensively with novices. The worksheet helps designers assess the potential quality of a part made using most AM processes and indirectly suggests ways to redesign it. The immediate benefit of the worksheet is to filter out bad designs before they are printed, thus saving time on manufacturing and redesign. We implemented this as a go-no-go test for a high-volume AM facility where users are predominantly novices, and we observed an 81% decrease in the rate of poorly designed parts. We also tested the worksheet in a classroom, but found no difference between the control and the experimental groups. This result highlights the importance of motivation since the cost of using AM in this context was dramatically lower than real-world costs. This second result highlights the limitations of the worksheet.
View
Life Cycle Assessment of 3D Printed Products in a Distributed Manufacturing System: LCA-Distributed Manufacturing System
Article
Full-text available
Motivated by the rising costs of doing business overseas and the rise and implementation of digital technologies in production, new strategies are being explored to bring production and demand closer. While concepts like cloud computing, internet of things, and digital manufacturing increasingly gain relevance within the production activities of manufacturing companies, significant advances in three-dimensional (3D) printing technologies offer the possibility for companies to accelerate product development and to consider new supply chain models. Under this production scheme, material supply chains are redefined and energy consumption hotspots are relocated throughout the life cycle of a product. This implies a diversification of energy mixes and raw material sources that poses a risk of shifting problems between life cycle phases and areas of protection. This study compares a conventional mass scale centralized manufacturing system against a 3D printing-supported distributed manufacturing system on the basis of the production of one frame for eyeglasses using the life cycle assessment methodology. The study indicates clearly that the optimization potential is concentrated mainly in the energy consumption at the unit process level and exposes a close link to the printing material employed.
View
Dimensional Tolerances for Additive Manufacturing: Experimental Investigation for Fused Deposition Modeling
Article
Full-text available
  • Dec 2016
Additive manufacturing creates parts in layers without using formative tools. Compared to established manufacturing processes, additive manufacturing offers many advantages. However, only a few research institutions and technology-leading companies use additive manufacturing for end-use part production because relevant challenges have not been sufficiently researched yet. Missing restrictions become apparent in the available geometrical accuracy. The objective of this investigation was the experimental determination of dimensional tolerances using standard parameters. To this end, a methodical procedure was set up. Based on experimentally determined deviations, dimensional tolerances were derived.
View
Optimum Part Build Orientation in Additive Manufacturing for Minimizing Part Errors and Support Structures
Article
Full-text available
  • Dec 2015
Additive Manufacturing (AM) is the process of part building by stacking layers of material on top of each other. Various challenges for a metal powder based process include reducing the staircase effect which leads to poor surface finish of the part, and minimal use of support structures for regions with overhangs or internal hollow volumes. Part build orientation is a crucial process parameter which affects part quality, in particular, Geometric Dimensioning & Tolerancing (GD&T) errors on the part, the energy expended and the extent of support structures required. This paper provides an approach to identify an optimal build orientation which will minimize the volume of support structures while meeting the specified GD&T criteria of the part for a DMLS based process. Siemens PLM NX API is used to extract the GD&T callouts and associated geometric information of the CAD model. The regions requiring support structures are identified and a Quadtree decomposition is used to find the volume of support structures. The mathematical relationships between build orientation and GD&T are developed as part of a combined optimization model to identify best build orientations for minimizing support structures while meeting the design tolerances. The feasible build orientations along with the corresponding support structures are depicted using a visual model.
View
View
View
Material and energy loss due to human and machine error in commercial FDM printers
Article
Additive manufacturing is thought to have environmental benefits. However, material waste and energy consumption could be larger than expected due to human or printer error. In general, fused deposition modeling (FDM) printers using ABS plastic have three stages for parts fabrication: the standby period, preheating process and the printing process. In practice, the quantity of support material is influenced by the part orientation and settings of the printing. Material waste and energy consumption for commercial FDM printers using ABS material in a heavily utilized open shop were analyzed. The failed prints were classified into 9 different categories and analyzed. The data indicated that about 34% of the plastic used in the open studio was wasted. Only considering the failed prints as the extra amount of material consumed under realistic conditions, the mass of material lost to failed builds was about 2.22 times what might be estimated in a controlled process study. For energy consumption, the standby period and preheating time vary for every job, which results in variability of energy consumption. The printing time is based on the geometry of parts being built, the part printing orientation and setting of that print. From collected data, the preheating energy consumption is 0.835 MJ/kg, the printing is 21.5 MJ/kg, and the standby is 9.5 MJ/kg. Moreover, suggestions to reduce the building cost for each failure type are given. A life cycle inventory (LCI) combining the material waste and energy consumption data of FDM reveals that actual energy consumption may be 50% more than under ideal conditions.
View
View
Participatory politics: Next-generation tactics to remake public spheres
Article
  • Jan 2014
Although they may disavow politics as such, civic-minded young people use every means and media at their disposal to carry out the basic tasks of citizenship. Through a mix of face-to-face and digital methods, they deliberate on important issues and debate with peers and powerbrokers, redefining some key dynamics that govern civic life in the process. In Participatory Politics, Elisabeth Soep examines the specific tactics used by young people as they experiment with civic engagement. Drawing on her scholarly research and on her work as a media producer and educator, Soep identifies five tactics that are part of effective, equitable participatory politics among young people: Pivot Your Public (mobilizing civic capacity within popular culture engagements); Create Content Worlds (using inventive and interactive storytelling that sparks sharing); Forage for Information in public data archives; Code Up (using computational thinking to design tools, platforms, and spaces for public good); and Hide and Seek (protecting privacy and information sources). After describing these tactics as they manifest themselves in a range of youth-driven activities - from the runaway spread of the video Kony 2012 to community hackathons - Soep discusses concrete ideas for cultivating the new literacies that will enable young people to participate in public life. She goes on to consider some risks associated with these participatory tactics, including simplification and sensationalism, and ways to avoid them, and concludes with implications for future research and practice.
View