SAMPE Journal - November/December 2016 - 26

Feature Article
Accuracy and Thermal Stability
To assess accuracy and stability of FDM composite
tools, multiple tools were evaluated both before and
after thermal cycling. Three different tool designs were
produced and build construction (shell vs. sparse build)
and sizes were varied for a total of five variants. The
tools were sent to an external inspection facility for
precision 3D scanning. A Platinum FaroArm (from FARO®
Technologies) with an SLP 300 laser head (from Laser
Design, Inc.) was used. The scan data was compared
to the CAD model for each variant using PolyWorks®
View™ 3D metrology software (from Innovmetric).
All composite tools used for this evaluation were
post-processed (abraded) prior to the initial 3D imagery.
This configuration was selected since nearly all FDM
composite tooling will undergo such preparation prior
to use, making the accuracy of a post-processed tool
the most relevant data. Although there is likely some
variability in post-processing between operators, the
overall amount of material removed during abrasion was
found to be quite small (using standard "best practice"
procedures) and does not represent a significant portion
of overall dimensional variation.
As stated, tools were scanned before exposure to
elevated temperatures and then sent for thermal cycling.
For cycling, the tools were vacuum bagged (envelope
bagging scheme), heated to 180°C, held at temperature
for two hours (minimum) under full vacuum, and then
ramped back down to below 65°C between cycles for a
total of 10 consecutive oven cycles.
Moisture Exposure
Many polymeric materials absorb moisture to some
extent over time at various rates. Per the manufacturer
(Sabic), ULTEM™ 1010 will absorb 0.7% when saturated
(23°C / 50% RH). Since moisture can be detrimental to
composite laminate quality, relatively rudimentary testing
was performed to ensure that such adverse effects can
be prevented with basic precautions.
To ensure saturation and a "worst case" exposure
scenario, four tools (two each shell style and sparse

build constructions) were placed in a humidity chamber
at 60°C / 90% RH for two weeks. After conditioning, two
were subsequently dried for 4 hours at 125°C. Eightply quasi-isotropic carbon/epoxy laminates were then
produced on each tool. The laminates were visually
inspected after cure and then sectioned for microscopy
to inspect for porosity, delamination, blistering, and
other indications of moisture-induced effects. The
primary objective was to demonstrate that even in the
most severe climates, if moisture absorption becomes a
concern, oven drying of tools prior to use is sufficient to
prevent adverse effects on cured parts. In reality, most
tools in a state of regular use are likely to be stored in
environments far less harsh than those tested.
Tool Life
A thorough understanding of the potential useful life of
a non-metallic tool is critical, particularly for production
tooling consideration (or for any substantial part volumes
beyond prototyping). It is also challenging information
to obtain experimentally due to the time and resources
involved. In working toward a preliminary baseline, both
practical (empirical) and analytical data was gathered.
For empirical testing, the basic approach outlined for
the accuracy and thermal stability testing described
previously was followed, but extended to higher
numbers of thermal cycles. A single tool geometry (UAV
fan blade), built in the two primary build constructions
(i.e. shell and sparse build styles) was tested (the tools
are as shown in Figure 2). Tools were cycled for 30, 60,
and 90 cycles (180°C, full vacuum, oven only), followed
by evaluation (inspection and 3D scanning) and laminate
fabrication (eight-ply, quasi-isotropic carbon/epoxy) with
subsequent inspection and dimensional evaluation.
For the analytical portion, dynamic mechanical analysis
(DMA) was used to evaluate creep in flexural specimens
(3-point bend configuration). Isothermal testing was
performed with a 0.7 MPa (100 psi) loading condition at
multiple elevated temperatures (i.e. 180°C, 195°C, and
205°C) and then time-temperature superposition (TTS)
principles were used to form an understanding of long-

Figure 4. 3D scan data for a UAV fan blade (shell
style) tool with color map comparison to the
original CAD model (no thermal cycling).
Dimensions are in inches.

SAMPE Journal, Volume 52, No. 6, November/December 2016

Table of Contents for the Digital Edition of SAMPE Journal - November/December 2016