IEEE Robotics & Automation Magazine - March 2017 - 73

(Maxon Motors and Faulhaber are examples of suppliers with
sourceable, coreless dc motors) so there is no cogging torque
and so the manipulandum feels smooth to operate. For lowcost devices such as the Novint Falcon, anticogging software
is used, where motors are calibrated and cogging torques are
subtracted by an additional current feed-forward term. This
results in a humming sensation through the manipulandum,
yet it also reduces the perceived cogging torques [16].
Material
Plastics such as acrylic are as easy to cut as plywood and
come in different colors for the designer to experiment with,
but they can be brittle. They also tend to be heavier and
have to be supported with more motor torque for gravity
compensation. Increased stiffness can be achieved by using
other materials, such as aluminum (which can be cut into
sheets using a waterjet rather than a laser cutter). This metal
can result in a stiffer and in fact lighter device. (The density
of 6061 aluminum is 2,700 kg/m 3 as compared to Baltic
birch plywood at approximately 3,500 kg/m 3 .) Its strength
also allows increasing porosity in the structures without losing structural integrity. The disadvantage is the higher material cost, requiring all screw holes to be tapped separately,
and tighter tolerances for the press-fits of bearings and finger joints. Choices of solid and composite woods can provide different stiffness and weight tradeoffs. Physical
stiffness, the inertia of the device, and even visual appeal can
be explored by using different materials. Figure 9 shows a
variant where one part is hand-fabricated from solid wood
using a lathe.
Add-Ons
A user may add buttons, sensors, or even vibrotactile
actuators on the manipulandum, which can further improve
the perception of textures [17]. Different grips or end
attachments that interface with the user can be explored.
Community Reception
Our previous studies [3] showed that the user experience of
the final device, in terms of perceived stiffness and related
characteristics, was rated as most similar to that of the
Phantom Desktop, the most expensive device in the test. It
was also shown how it was possible for a nonengineer to
assemble the device, under supervision, using a limited set of
tools in an ordinary office environment [3]. The assembly
(Figure 5) took 11 h stretched over a few sessions. The kit was
provided to users in its physical form and facilitated by one of
the authors.
In this section, we present a case study of the early community reception. Since WoodenHaptics' initial public
release in January 2015, about 30 people have made online
inquiries about purchasing the kit, and one external group
has successfully built the device themselves. Except for that
group, too few have yet constructed the device to validate its
community impact as a whole or whether the kit works for a
particular user goal for users in general. Nevertheless, we can

Figure 10. Some different motors that have been used for haptic
devices: (from left) Maxon RE40, Maxon RE30, Maxon RE25
(Phantom), Mabuchi RS-555PH (Falcon), Mabuchi RS-455PA
(Omni).

learn more about how current and potential users appropriate the kit, what value they think it brings, and what limitations they see through an in-depth case study. Borrowing
from the wide use of such studies in HCI, a case study is
defined as "an in-depth study of a specific instance (or a
small number of instances) within a specific real-life context"
[18] and is used for several reasons, including 1) exploration,
that is, to understand novel problems or situations, 2)
description, that is, documenting a context of technology
use, or 3) demonstration, namely, showing how a new tool
was successfully used.
Our study is demonstrating one case where the kit was
successfully used by a large company to create a public demo.
The study also shines a light on various perspectives of opensource robotics use, for example, for supporting learning
goals in education, including for both teachers and learners in
engineering as well as for technological novices.
Practically, the study comprises in-depth semistructured
interviews with seven subjects (male, aged 23-48, the average
being 32, representing four different nationalities). Two interviews were done in person and three by video link, of which
one was a group interview with three subjects. All participants
consented to their involvement and to being recording for
confidential analysis by the researchers. A semistructured
interview protocol was used, and each interview lasted
30-60 min. Notes taken during the interviews were partly
transcribed and analyzed and sorted into themes.
Two of the respondents, both professional engineers but
with different backgrounds, had built the device, one using
plywood and the other acrylic. A third was a professor who
had shown interest in using the device primarily for teaching
mechatronics and controls. A fourth was a master's student in
HCI with a nonengineering background, and the last three
were mechatronics engineering students. Apart from the
professor, none of the respondents had prior formal haptics or
robotics training, but all had some experience building
physical things.
Results
The results of the interviews can be sorted in terms of
1) motivation, including the motivation of those who built the
march 2017

IEEE ROBOTICS & AUTOMATION MAGAZINE

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - March 2017