Bear River Truss

Design Report Cover Sheet

Presented to the Citizens of Bear Valley by:

Team Name:(4) Team Awesome

Member1: Katya Stavisky

Member2: Kelsey Lange
Member3: Griffin Nosenzo
Member4: Brittany Hallawell

Number of Joints: 9 (planar view)

Number of Members: 15
Weight: 10.6 grams

Section Page Number

A. Bridge Design Description
Designs considered during brainstorming 1
Detailed description of final design 2
Design Specification Table 3
Bridge Diagram (label max tension and compression members) 3
B. Analysis and Construction of Bridge
Process Summary and Results 4
Member Force Table (with Factor of Safety) 5
Description of Construction Process 6
C. Testing 9
D. Summary of Project Processes
Description of Work Process and Schedule 11
Description of Individual Contributions 11
Reflections on Process and Results 11
Appendices:
Preliminary Designs 13
Supporting Calculations 23
Test Sheet 28
A. Bridge Design Description
During brainstorming, we considered numerous designs. We quickly decided to take a
strategic approach by choosing to prioritize specific criteria that would earn us bonus points. We
decided that a “creative” bridge design would be the best goal as it does not depend on other
groups’ bridges. We researched other types of bridges, especially focusing on trusses made up of
equilateral triangles, trusses made up of isosceles triangles, trusses with angles that mimicked the
appearance of an arch, and bridges involving lower supports against the table edges (See
Appendix A). With so many variables, we realized we had to make a few basic design decisions
to allow for valuable analysis.
For a “creative” bridge design that was not based on the traditional Howe or Pratt
designs, the Warren bridge, which is a simple design consisting of equilateral triangular trusses,
caught our eye1. It was not one of the four traditional designs, looked simple in terms of
construction, and seemed light-weight, which would help fulfill the minimal material goal. After
we decided to focus on the Warren Bridge for one of our designs, our next question was how
many members to use. We decided to divide and conquer this problem by each doing a hand-
analysis of Warren-like bridges with differing numbers of base triangles at a safety factor of 1.5
and 2.2 (see Figure I). Using a combination of hand calculations and the online truss solver2, we
discovered that the more equilateral triangles we had, the larger the internal loads on the truss
members, and in order to have a safety factor under the maximum of 2.2, we would need to
design a bridge with at least seven equilateral triangles along the bottom (See Appendix A pg 11-
15). At the same time, we considered altering the Warren bridge design to include isosceles
triangles, allowing greater flexibility for the dimensions of the bridge.

Figure I: Initial Warren-inspired Bridge Designs

!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
1!Cridlebaugh,!Bruce!S.!"Guide!to!Bridge!Design."!Bridge'Basics.!N.p.,!3!June!2008.!Web.!20!Feb.!2013.!
2!"Virtual!Laboratory."!Bridge'Designer.!Johns!Hopkins!University,!n.d.!Web.!12!Feb.!2013.!

! 1!
 We also used a combination of hand calculations and online truss solvers to evaluate the
four traditional truss designs: the Pratt Bridge and Roof and Howe Bridge and Roof. Based on
these calculations, we compared the weights in terms of the total length of the members
multiplied by a constant, discovering the Howe Bridge to be the most efficient. Similarly, we
evaluated the various “creative” design options to determine that the relative weights of various
numbers of triangles and heights, and decided that the bridge would be easiest to design and
contain the minimal amount of material if we allowed the triangles to be isosceles--as opposed to
the traditional Warren design with equilateral triangles--so that we could easily adjust the
dimensions to meet the design requirements. We also liked this design because, when evaluated at
a safety factor of 1.6, the angles were very close to 90-45-45 which would be very easy to
implement during the building phase. We decided to use this as our final design.
However, we soon found that there was a slight issue with this design as the calculated
height for a safety factor of 1.6 was 3.2 cm, which was a whole .3 cm under the minimum height
required for the roadbed to fit inside. This proved only a minor inconvenience, however, and we
recalculated the internal loads using a height of 3.5 cm, which still kept our design within the
safety factor requirements at 1.76 (see Appendix B pg 24-25). Throughout the analysis process,
we checked each other’s calculations and independently solved for forces to verify accuracy for
our final designs.
The planar view of our final design consisted of 9 joints, 15 members, spanned 25 cm,
had a width and height of 3.5cm, a weight of only 81.3*c (which is the total length of balsa wood
required multiplied by a constant), and a safety factor of 1.761 (see Table I). It was made up of
four base and three top isosceles triangles as shown in Figure II on the next page. The angles
were still quite close to a 45-45-90 degree triangle for ease of construction. Our team expenditure
came down to $0 dollars as we minimized material use with this design. It fulfilled the non-
standard design preference and had what appeared to be a promising shot at the minimized
material cost specification.

Table I . Bridge design specification table

Design Requirements Our Design Our Built and
Specs Tested Bridge
Number of joints n/a 9 (planar) 9
Number of members n/a 15 15
Span 24 cm 25 cm 25.5 cm
Width 3.5 cm 3.5 cm 4 cm
Height ≤ 10 cm 3.5 cm 4 cm
Cost (team expenditure) ≤ $10 0 0
Weight Minimized 81.3c 10.6 g
Design approach Non-standard preferred Isosceles triangles Isosceles triangles
Safety Factor (SF) Min: 1.5 1.761 1.58
Requirement Max: 2.2

! 2!
 Figure II. Bridge Diagram with Safety Factor of 1.76

! 3!
B Analysis and Construction of Bridge

Description of Analysis Process

When analyzing both the external and internal forces on the bridge, we made several
assumptions. We assumed that the balsa wood is homogeneous, the members are rigid, and that
the mass of the wood and glue is negligible relative to applied loads. When taking calculations,
we assume that gravity is downward and that there are no external forces other than the applied
load of 66 N acting on the bridge, so that the system can be modeled as in static equilibrium. The
equilibrium calculations also assume that the joints behave as pin joints and that the shear force
on the glue will not be the cause of failure, meaning only compressive or tensile loads will cause
failure.
Before calculating any forces by hand we used an online truss solver to determine which
members would experience the most compression and tension. Next, we used the method of joints
to calculate the forces acting on each joint until we had found the force acting on members CD
(maximum compression) and IH/HG (maximum tension) in terms of the angle theta and therefore
the height of the system as shown in Appendix B on page 26. We set these forces equal to the
maximum allowable compression of 69.5 N and tension of 735 N, as defined by the qualities of
the balsa wood. Using a safety factor of 1.5 and then 2.2, we calculated the necessary height of
the bridge. The minimum height to accommodate compression was higher than that of tension, so
the minimum height corresponds to the height required to withstand compressive failure. Next,
we calculated the relative weight of the truss bridge based on the total length of balsa wood
required for that specific height.
Realizing that the required height for the bridge is 3.5 cm, we then analyzed the internal
loads in each of the members with this specific dimension as shown in Appendix B on page
24. This also determined the lengths of the individual members, which would be 4.69cm for the
diagonal and 6.25cm for the horizontal members. Based on the data provided about the balsa
wood, neither length would be at risk for buckling failure. With these final dimensions, we were
able to determine the safety factor of 1.761 based on the expected load of 66 N and the maximum
compression of 39.2 N in member CD. Knowing our final dimensions, we were able to estimate
the design load of 121.5 N and the internal forces as shown in the Figure II and Table II.

! 4!
 Table II Member force table of member loads for applied load of 121.5 N (SF 1.761). Shaded row indicates
member predicted to fail at applied load.

Tension Member Failure Load Capacity

Member Load Compression Length Mode
AB 40.7 Compression 4.69 Yield 735.0 N
AI 27.1 Tension 6.2 Yield 69.5 N
5
BC 54.2 Compression 6.2 Yield 735.0 N
5
BI 40.7 Tension 4.69 Yield 735.0 N
CD 72.3 Compression 6.25 Crushing 69.5 N
CH 13.6 Tension 4.69 Yield 735.0 N
CI 13.6 Compression 4.69 Yield 69.5 N
DE 54.2 Compression 6.25 Yield 69.5 N
DG 13.6 Compression 4.69 Yield 69.5 N
DH 13.6 Tension 4.69 Yield 735.0 N
EG 40.7 Tension 4.69 Yield 735.0 N
EF 40.7 Compression 4.69 Yield 69.5 N
FG 27.1 Tension 6.25 Yield 735.0 N
GH 63.4 Tension 6.25 Yield 735.0 N
HI 63.4 Tension 6.25 Yield 735.0 N

To calculate the internal forces shown in the Member Force Table (Table II), we used the
method of joints and the assumption that the system is in equilibrium. The external forces acting
on the entire bridge as the system could be found by assuming a pin joint at F and a roller at A
(See Figure II). The design load was calculated to be 121.5 N, which would result in 60.7 N
acting on each planar side of the bridge. Because the wood is assumed to be homogeneous, the
load can be approximated at 20.2 N when distributed evenly to each of the three joints on the
bottom edge of each side. The external force acting at joint A and F are then found to be 30.4 N
in the positive y-direction, opposite the 60.7 N downward, using equilibrium equations in the x-
direction, y-direction, and moment about point A. There are no other forces acting in the x-
direction, so the force at pin join F is only vertical.
With the external forces known, the internal forces are found using the method of joints,
starting with joint A. The height of the bridge is 3.5 cm, so the angle between AB and AI is
approximately 48.24 degrees. This angle is the same measurement throughout the bridge because
of the isosceles triangle design. At joint A, there is a diagonal force AB and a horizontal force AI
in addition to the upward force of 30.4 N. Using the sum of the forces in the y-direction, we find
the magnitude of the force in AB to be 40.7 N pointing toward joint A, indicating that the force in
member AB is compressive. With the sum of the forces in the x-direction, the force in member
AI is found to be 27.1 N in tension. Following the same steps of solving the equilibrium equation
in the y-direction and then the x-direction for joint B, we find the force in BI to be 40.7 N in
tension, and the force in BC to be 54.2 N in compression. We can then solve for the unknowns at
joint I, and finally joint C which gives us the compression in member CD to be 72.3 N. This is
the member with the most compressive force, and thus the member that would cause failure at the
design load. This value is slightly higher than the balsa wood’s given maximum compressive
load of 69.5 N; the discrepancy likely resulted from rounding too early in the calculations and the
estimations of the design load. The expected maximum load that the bridge can withstand is
slightly less than the design load of 121.5 N, and has a safety factor slightly below 1.76. The

! 5!
design is symmetrical, so the magnitudes of the individual members can be mirrored on the other
side of the design.

Description of Construction Process

We started the construction process by drawing out a model of our design on a

whiteboard to make sure the whole team was on the same page (see Figure III). We needed to
decide how to align the measurements with the thickness of the balsa wood in order to accurately
represent our measurements. We worked together come up with a design in which the height
remained the required 3.5cm and the simplified analysis design aligned with the centerlines of the
balsa wood in the diagonal members.

The green lines represent the simplified

design, with the red lines illustrating
how the width of the wood was
accommodated.

Figure III Transition to Three Dimensions

We cut the members by mapping the outline of the members of balsa wood at a joint in
Figure IV. The team organized tasks so that one person cut the members, while another glued
using a paintbrush, another finalized design drawings, and the last person calculated design
loads. It quickly became clear that the building process was at most a two person job (cutting and
gluing as shown in figures V and VI on the next page), in which extreme focus was required for
gluing the pieces so that the joints aligned properly.

Figure
Figure IVIV Construction
Joint Reference
Construction !
Sketch
Template

! 6!
 In our bridge, there were two specific member designs: one for the horizontal members
and one for the diagonal members. We had to adjust the alignment of the wood on the template to
account for the width of the pencil lines and so we did our best to use the inner lines for
reference. The angles were difficult to perfectly duplicate, so we had to adjust the alignment on a
joint-by-joint basis using the sandpaper to make fine adjustments to the angles. The ultimate goal
was to make the horizontal members as parallel as possible. We used the straight lines on the
graph paper and rulers for this broader reference point.

Figure V Cutting with Template

Figure VI Gluing with Template

After completing a few of the trusses, we double checked the measurements of the
angles, height, and members with the design plan, as shown in Figure VI. Luckily, the error was
minimal and we were able to continue construction. After the first side was built, we switched to
using its physical members as models with the goal of keeping the two sides as symmetrical as

! 7!
possible. The second side of the bridge ended up being approximately 3.2 mm longer in span
length than the first side, so we did our best to make up for the difference by aligning the center
cross beams first, and adjusting the outer members slightly to ensure that the joints would act as
pins with all members meeting at approximately the same point. Once we had the two side
trusses, we cut some straight members at the required 3.5cm width for the roadbed to hold the
sides together. We glued the entire bridge together using the metal braces, which was a delicate
process with the bridge now in three dimensions. We used binder clips to secure the side trusses
to the metal braces during this final gluing process (see Figures VII and VIII). The whole process
took about six hours, with several breaks to allow for the glue to dry. At the end, we realized we
wanted to put in two more horizontal supports, which we did a few days later.

Figure VII Final Gluing

Figure VIII Gluing With Metal Braces

! 8!
C Testing
On Tuesday, the 19th of February, we tested our bridge. The test consisted of placing the
bridge across a 23 cm span, placing a 3.5 cm x 21 cm x 3.5 cm roadbed inside the bridge, and
hanging weights off of the road bed until failure as shown by Figure IX. Our bridge ultimately
failed at a load of 104 N. The breakage was due to joint failure in several places as illustrated by
Figure X on the following page. Based on how the failure occurred at several of the joints, the
design could have used more glue at the joints. We did not account for this type of failure in our
analysis because we assumed that the failure causes would be limited to compressive or tensile
failure within the wooden members. We also simply did not have a method for calculating the
strength of the glue at the joints. From our calculated safety factor and our analysis of different
failure mechanisms, we expected our bridge to fail due to compression at a load slightly below
121.5 N. Sadly, due to the premature joint failure we cannot know if our bridge would have failed
above or below this compression force. The failure was difficult to predict, because it could have
resulted from various simplifications we used in the calculation process. In particular, some
items that may lead to variation between calculated and actual values would be that the balsa
wood is not homogenous, the members are not massless, the bridge doesn’t have true pin joints,
the constructed bridge is not entirely accurate to our design, etc. There is also a possible
discrepancy in the fact that the weight of the roadbed and the bucket itself may not have been
included in the load measurements. The loads were also applied at increments of 5 N when the
failure occurred, so the determined failure load is an approximation that could have been up to 4
N lower than the actual failure load.

Figure IX Loading Weights During Testing

! 9!
 Figure X Failure at Joints Post-Testing

! 10!
D Summary of Project Processes
We first met on the evening of Monday, February 11th to begin planning our bridge.
This first meeting consisted of some members working on calculations while others looked at
other designs on the internet and truss solving software. By the end of this meeting, we had
settled on one of our designs consisting of equilateral triangles (a Warren Truss), and narrowed
down our other design to a couple possibilities. At this stage, we were primarily planning on
trying to make a unique bridge that did not adhere to one of the four traditional designs. We
divided analysis of the Warren truss amongst the group. With equilateral triangles, the height is
fixed for a given number of triangles and a given length, so we had to calculate several versions
of the bridge with different numbers of triangles to see if any were within our safety factor range.
Katya also calculated a version where instead of equilateral triangles, isosceles triangles were
used so that we could adjust the height and safety factor while keeping the overall length and
number of triangles the same. Our next meeting was on Wednesday night. Here we went over
the analysis of the Warren truss based designs, and decided to have our other design be a Howe
Bridge, as it had the lowest calculated weight of any of the four given designs. We also finished
the calculations for the given safety factors of 1.5 and 2.2 for both bridges.
On Friday, February 15th, we did most of the building for our bridge. We started by
converting our designs with dimensional members to more realistic representations of our final
bridge and ensuring that our dimensions were correct for this new design prior to construction.
At first, Katya took point on construction. Griffin tried to help, but she was much faster at cutting
the members accurately than he was, so it became a one person job for a while. Brittany worked
on the design load analysis, calculating the internal loads based on the maximum external load of
121 N. Kelsey worked on drawing the Bridge Diagram and the joint diagram used for
construction (See Figures II and III). Kelsey and Brittany took over cutting and assembling
members for the second half of the bridge, with Brittany cutting and Kelsey sanding and gluing.
Once the two sides were complete, we all began to glue the bottom cross beams to one side, then
attached the other side. We glued two cross beams into the top of the bridge to keep the sides
parallel. Finally, over the weekend, Brittany and Kelsey added two more cross beams to the top
of the bridge.
Everyone contributed to our project. We all looked at various bridge designs and made
detailed calculations for some of them. Kelsey calculated the Howe Bridge, Griffin calculated the
Howe Roof, Katya did the Pratt Roof, and Brittany did the Pratt Bridge. Then Katya did an
isosceles triangle based bridge calculation, Kelsey and Brittany did different numbers of
equilateral triangles, and Griffin recalculated the Pratt Bridge for the given safety factors of 1.5
and 2.2. During building, initially, we all helped figure out how to make our three dimensional
design most closely match our calculated values. Then, at first, Katya started to do most of the
cutting and gluing. Eventually, she switched off with Kelsey and Brittany, until the two sides of
our bridge were complete. Then, Kelsey, Brittany, and Griffin attached the two sides together.
Over the weekend, Kelsey and Brittany added the two additional cross-beams. Then, we divided
the sections of the report among everyone, and we each drafted our sections. Finally, we all
worked on editing and formatting the final draft of the report

Reflections

In retrospect, there are a few things our team could have done to ensure a more successful
bridge. First and foremost, during the construction phase we should have used more glue at the
joints. Our bridge’s early failure was most likely due to this misstep.
There are several things, however, that our team did well. The design and analysis phase
was a particular strength of ours. We each chose a slightly different bridge to analyze after our

! 11!
first meeting, and later came together and compared results. This ensured a thorough exploration
of possible designs and led to an efficient bridge in terms of weight and load requirements. There
are many variables that we could investigate, for instance we could have investigated the
isosceles triangle design with more members, and yet we did a good job of limiting our options to
allow for precise analysis and comparisons. We also conducted independent calculations of the
same designs to corroborate our results which gave us confidence in our final design--it is easy to
make a small error during analysis that would lead to a large structural problem later on during
building or testing. Further, we made sure all members of our team were on the same page by
using a shared Google Doc, thorough email and texting communications, and many drawings on
paper and whiteboard that we could all reference.
We did our best to equally share in the work, but it became especially difficult during the
construction process. We realized that the work was more easily and accurately done if one
person focused on cutting the pieces and another glued the joints together, leaving the other two
members of the team without much to do. To counter that, we switched roles a couple times and
allotted more of the report writing to the members of the team that were unable to help with
bridge construction. This worked well, and each member drafted assigned sections of the report,
with all of us working together to edit and format the final version.

! 12!
Appendix A Preliminary Designs

! 13!
! 14!
! 15!
! 16!
! 17!
! 18!
 Howe Bridge Design

! 19!
! 20!
! 21!
! 22!
Appendix B Calculations

! 23!
! 24!
! 25!
! 26!
! 27!
Appendix C Test Sheet

! 28!