Portfolio Zach Kastner int .pdf
Original filename: Portfolio_Zach_Kastner_int.pdf
Title: Mechanical Engineering Design Portfolio
This PDF 1.5 document has been generated by Acrobat PDFMaker 10.1 for PowerPoint / Adobe PDF Library 10.0, and has been sent on pdf-archive.com on 15/02/2017 at 08:52, from IP address 108.212.x.x.
The current document download page has been viewed 362 times.
File size: 993 KB (11 pages).
Privacy: public file
Download original PDF file
B.E. – Mechanical Engineering
Design of Bipedal Walking Robot
Rapid Prototyping ∙ Modular Design ∙ Servo Control
My senior design project at Stevens was to
design and build a human-proportionate bipedal
walking robot with not mechanical tether and a
budget of $700.
Components used for the design and fabrication
of this robot included:
• Arduino Uno microcontroller
• Servo Motors
• Aluminum U-Channels
• Microrax beams and connectors
Other components used were pieces of scrap
found in the lab and 3D printed parts, allowing
Rapid Prototyping to become useful.
The robot was designed modularly with the
• Hip Joints
• Knee Joints
• Ankle Joints & Feet
Bipedal Walking Robot Design: Pelvis
Adjustable Design ∙ Sturdy ∙ Lightweight
The Pelvis of the robot was made using
lightweight aluminum beams and connectors,
The pelvis was designed to be as lightweight as
possible while being able to house servo motors
and circuitry, including the Arduino board. This
subsystem was also designed with a handle
going from the left servo to the right servo so
that the robot could be picked up and moved
SolidWorks was a useful tool for designing this
subsystem. The Microrax website had CAD files
for all their products, so it benefited the design
process to download necessary filed and
assemble in SolidWorks prior to ordering any
Microrax were also used because their design
allows for mounting of circuit boards.
Bipedal Walking Robot Design: Hip Joints
Pulley System ∙ 3D Printed Parts ∙ Machined Parts
The hip joints of the robot were designed to
allow 1 Degree of Freedom.
In order to maximize the efficiency of the servo
motors, a pulley system with a 1 to 4.5 gear ratio
was designed to put less load onto the servos.
This gear ratio allowed for 40 degrees range-ofmotion in order to mimic a human’s movement.
The pulley system used a timing belt and two 3D
printed gears; one attached to the servo and one
attached to the hinge as a pair of “wings”. The
larger gear was designed as wings because
printing a continuous part would have been too
large for the 3D printer available.
The pair of wings were modeled using
The hinges of the subsystem were machined
from Aluminum U-Channels.
Bipedal Walking Robot Design: Knee Joints
Pulley System ∙ 3D Printed Parts∙ Knee Cap
The design of the knee joints is very similar to
that of the hip joints.
Again, a pulley system, allowing 1 DoF, was
used as a way to put less load onto the servo
motors. The gears were designed in SolidWorks
and then 3D printed, and the hinges machined
from an Aluminum U-channel.
With a gear ratio of 1 to 2.5, this joint allows for
a 36-degree range-of-motion.
The range-of-motion is only 36 degrees instead
of the expected 72, based on the 180 range of a
servo, in order to better simulate a human
kneecap, preventing the joint from
Knee Cap Designed Hinge
Bipedal Walking Robot Design: Ankle Joints & Feet
Rapid Prototyping ∙ Machined Joint ∙ Design for Balance
The ankle of the robot was designed as a passive
joint. Adding springs to the subsystem allowed
the joint to remain level as the foot was
approaching the floor, while allowing the ankle
to rotate upon contact. The stiffness of the
springs was determined through different
iterations of rapid prototyping.
The foot of the robot was made from scrap
pieces of PVC found in the lab that were cut to
size and assembled with screws and nuts.
The shape of the foot was made to be flat and
wide in order to make the robot balanced while
The ideal weight of the foot was determined
through rapid prototyping by adding steel plates
to the bottom of the foot and doing a quick test
for smooth motion.
Redesign of Hockey Training Device
DFMA Results: Quicker Assembly ∙ Fewer Parts ∙ Cost Reduction
For a different project, I redesigned a Hockey
Training Device, namely the xDangler Pro
Stickhandling System, using the principles of
Design for Manufacturing & Assembly
The redesign resulted in:
69.3 seconds quicker assembly time
17 fewer parts
$157.31 cost reduction
Primarily, the system was redesigned for
portability, greater stability, and lower cost of
parts. These aspects of redesign were determined
by gathering the needs from potential customers
and inputting these needs into a House of
Quality table to select the most essential
requirements for the system.
Dividing the system into two modular
subsystems, the base and the top, I followed
DFMA principles and had a more efficient way
of designing the whole system.
Hockey Training Device Redesign: Top
Lighter Structure ∙ Part Reduction ∙ Material & Manufacturing Method Change
The first subsystem, the top, is the part of the
system that houses the electromechanical
components that allow the machine to mimic a
path of motion similar to that of a hockey
defender’s stick. These components were left
untouched in the redesign.
The main focus of redesigning this subsystem
was to reduce its weight. The weight reduction
sacrificed strength, but considering the use cases
for this device, the high strength was
First, the material for most structural parts was
changed to aluminum from stainless steel.
Secondly, the number of parts changed from 62
to 51 parts in total.
Changing the exterior to bent sheet metal instead
of five pieces of welded sheet metal allowed the
exterior to carry the load in this subsystem,
making more parts eligible for removal.
Hockey Training Device Redesign: Base
Disassembly & Reassembly ∙Low Center of Gravity ∙Minimal Deflection
The second subsystem, the base, was redesigned
as a subassembly of five circular aluminum
extrusions and two connectors allowing for
connection with the top subassembly. Before
being bent, the extrusions are all the same size
stock piece (18 inch long, 1.5 inch diameter).
The redesign of the base was chosen based on its
design feature of being able to be disassembled
and reassembled, as well as the results of two
analyses in SolidWorks.
In order to redesign the system to be more
stable, the Center of Gravity had to be lower on
the overall system, making it bottom-heavy. To
manipulate the CoG, I adjusted the dimensions
of the aluminum extrusions until the CoG
The redesign was considered stable after a quick
deflection analysis which showed a maximum
deflection of 2.305∙10-2mm, which can be seen
Center of Gravity