In this project I laser cut and assembled a bridge. I then replicated different compression and tension components of the bridge so that I could test and calculate their load. I then predicted what the bridge would hold and tested it. My partner for this project was Darby Collins.
Project Files
Links to instructions:
Here's a link to the PDF for the U.S. Military Academy "Designing and Building File-Folder Bridges" manual that provided the design drawings and whose instructions we followed for assembling the bridge: https://drive.google.com/file/d/11RKlRrYNdPL36KQxqHdWlv9gMUraVS8x/view
Here's a link to the PDF for the U.S. Military Academy manual for testing the bridge: https://drive.google.com/file/d/1cNY5hX-5Yuvxe-fHNdjc8orUe1pY9JAH/view
Design graphic JPG + labeled SVG:
Here's the JPG graphic design of the laser cutter bridge parts. The picture on the left is from Ginny Foster's design which I borrowed for the beam supporters and tension strips. The picture on the right is my Cuttle design exported into CorelDRAW for laser cutting.
These images are from the SVG file made with Cuttle.xyz then exported into CorelDRAW. Each part that was used of the SVG design is labeled below. I did not label each individual size for the compression beamsas you can see that in the "Required Parts Table" section below.
Workflows for lasers
Here are the workflows for the small, medium, and large laser cutters in the lab that I used for this project:
Small Laser Cutter:
https://docs.google.com/document/d/11z-DqA5AEVWgzxgJst_3dKyjn7n3JrUvnbmQJtXeA4M/edit (written by Griffin Orsinger)
Medium Laser Cutter:
https://docs.google.com/document/d/1GOyWMEj3ViTaLE_5XUlAyPPT1OH4zOMdRdKFBexndLw/edit
Large Laser Cutter:
https://teddywarner.org/Machine-Profiles/FusionPro48/#machine-workflow (from teddywarner.org)
Laser cutter workflow (design, loading to computer, settings, instructions for file & laser setup), differences in result, and kerf
Here is a diagram of the Gusset Plates. I used CorelDRAW to modify an initial re-creation of the technical drawing that Mr. Dubick provided. The Gusset Plates hold different compression beams and tensions strips together to help support the bridge.
1] DESIGN: This is the technical drawing for the different compression beams. I used Cuttle to parametrically re-create the design efficiently for laser cutting. Here's a link to the design: https://cuttle.xyz/@AdamStone/Civil-Engineering-Compression-Tension-Test-iLaUIBwfLLBO
See the design in the "Tubes together" component! Here's a picture:
2] LOADING DESIGN TO COPMUTER: I uploaded the SVG file to the "engproj" google drive and then downloaded it onto the computer next to the large laser cutter. I then opened the design in CorelDRAW.
3] SETTINGS: We used chipboard material for our bridge. It took a long time to get the score settings correct when using the medium-sized laser cutter. We found that ~80 speed with ~20 power worked well, with the recommended frequency for vectoring cardstock. On the large laser cutter we found that the cardstock settings worked well.
4] INSTURCTIONS FOR FILE AND LASER SETUP: On the big laser cutter, it is not required to move the design onto a specific part of the page in CorelDRAW, however it is vital to make sure the page reflects the size of the material and move the design to the top-left of the page for the medium and small laser cutters. All vector lines must be hairline thickness and lines with different settings must be different colors. Then it is necessary to manually focus the lasers (the auto-focus was not functional on the large laser cutter). For these processes, refer to the workflows linked earlier.
DIFFERENCES IN DESIGN VS RESULT: We noticed that the score lines were much stronger (and fell apart more easily) when the line being cut was vertical as opposed to horizontal - we found this result very puzzling. To solve this, we simply cut all of the pieces horizontally. Here are videos and images of the laser-cutting process:
KERF: Kerf is defined as "A portion of material that the laser burns away when it cuts through" (see source here). Since sub-milimeter-precision isn't necessary for these parts, and for some of the beams we will need to trim them with scissors anyways, it is not necessary to take kerf into account. However, to be extra precise, we could have offset the design by 1/2 the kerf of the laser cutter (which was calculated and explained in detail during Fab Academy here).
Part Processing
Cardboard template setup and use
Our cardboard template was taped to a large piece of cardboard. As we laser cut pieces, we verified that they were in the right size by comparing them to the outlines on the template printed out on the cardboard, and after verifying we layed them down in the right positions.. However, we first took a large piece of cardboard with a cutout of the design and put a plastic sheet over everything. We first layed the gusset plates down and taped them to the plastic. Then we layed the tension strips and compression beams and glued them to the gusset plates. This process was fairly smooth although time consuming. Putting tape around our fingertips was helpful for holding down pieces without covering our fingers in glue.
"Processing Pieces"
We folded each of the compression beams along the score lines and used Super Glue and held it in place as it tried. Then we glued the tension and compression pieces to the gusset plates. The part processing was a fairly smooth procedure and most of the difficulties came from getting glue on my hands/under my fingernails. However, with the help of a partner, this process was fairly expedient. We verifying measurements by comparing them to 1] the design specification tables (see below) 2] the cardboard template. We repeated this same process when "processing" our bridge testing pieces.
Required Parts Table and Table Explanation
Below on the left is a table of the required truss members to build the bridge (the different compression beams and tension strips). The schedule of gusset plates on the right details all of the required gusset plates and their connections (the letters in the Connections correspond to the diagram below on the right of the bridge. The very bottom left displays the technical drawing for each of the compression pieces in the table on the right (these were the technical drawings from which I created the SVG in Cuttle, detailed above).
Problems encountered
I encountered one major problem while creating, cutting, and folding individual components of the bridge. The score lines that were cut vertically on the medium laser cutter cut almost all the way through, so when we tried folding the pieces, they broke! This led us to recut them, horizontally oriented, which worked better.
Component Screenshots
Beloware screenshots of all 5+ different kinds of components that went ito the bridge: the different types of pieces are explained above with included excerpts of table requirements from the textbook from West Point.
Constructing the Bridge
Image of Final Construction
Here's a picture of the final construction of the bridge:
Attaching bridge components
The attached the bridge components we mainly used Super Glue. All of the joint are a gusset plate with several members (tension strips and compression beams) super-glued to it. On occasion, if there was too much Super Glue, it wouldn't dry, so we used painter's tape (see the picture above on the right) to hold it in place then took it off after it dried.
Shortcomings of the final build & potential future improvements to address shortcomings
The shortcomings of the final build include its aesthetic appeal. There are some tiny pieces of tape that were not able to be removed without damaing the bridge on several spots (as you can see above). The bridge has a strong structural integrity and our work process prioritized function over form (which is always a necessity in real life Civil Engineering: bridges must be safe and rigorously tested).
In the future to address these problems, we could pre-emptively use a smaller amount of glue so that we were never required to use painter's tape to hold components in place as an excess of glue took a long time to dry.
Testing the Bridge Components
Processes and tools used in testing bridge components
To test the bridge components, we cut replicas of the different tension strips and compression beams so that we could use the bridge testing machine to measure the mass that each piece would support. We measured the distance from the fulcrum at which we placed all of the pieces (and the distance from the fulcrum to the weigh was constant). We slowly filled a bucket with sand then measured the mass of the sand and the bucket together. Sometimes the very thin tensions pieces even broke from the weight of only the bucket!
Then we cut replicas of the different tension strips and compression beams so that we could use the bridge testing machine to measure the mass that each piece would support. We measured the distance from the fulcrum at which we placed all of the pieces (and the distance from the fulcrum to the weigh was constant). We slowly filled a bucket with sand then measured the mass of the sand and the bucket together. Sometimes the very thin tensions pieces even broke from the weight of only the bucket!
Videos and diagrams of testing and machine used
Here's a diagram of the Bridge Testing Machine!
Equation for force measurement of Tension and Compression
Here is an except from the bridge-building manual for this project. Here's an explanation of the equations: T represents the load on the tension strips, W represents the weight, L1 and L2 represent the lengths from the fulcrum to the piece being tested and from the fulcrum to the weight. C represents the load on the compression beams tested. The graph on the bottom represents the strength of a member (load is can bear without failure) versus it's width. It reveals that wider widths can hold a higher load, which is intuitive.
Table of measurements for compression and tension tolerance of each component
We used Google Sheets to record all of the data for the different dimensions of tensions strips and compression beams, as well as the distances of the different pieces from the fulcrum (pivot point). Here's a picture of our sheet below. Here's a link to the spreadsheet: https://docs.google.com/spreadsheets/d/103hxnowsHfISYreoW6abfrvlt9Iabtiu1iwFdqz6nxY/edit?usp=sharing
Analysis of table and potential reasons for observations
Here were our takeaways from the data:
A smaller length of a compression beam corresponds with a higher strength
Reasoning: on a very long compression beam, there are lots of spots at which the beam could bend and fail, so shorter beams are less likely to fail than longer beams
Thinner/non-square compression beams have lower strength than thicker/more-square beams
Reasoning: a thick, square compression beam distributes the compression force along a much larger area/volume than a not-square, thin beam, explaning why thicker, square beams should have a higher strength
Thicker tension strips correspond with a higher strength
Reasoning: a thicker tension strip has a larger area, meaning that the load can be distributed throughout a longer surface, allowing it to bear a higher load than a thinner tension strip
Analysis and graphic of bridge force distributions
GRAPHIC: This from COSMOL is a clear graphic of the force distribution of a Patt Truss Bridge (which means "a bridge with a single tension diagonal in each panel and a compression vertical with parallel chords and an inclined end post" according to STRUCTURE Magazine).
ANALYSIS: the thing, tensile strips away from the center of the bridge experience the largest load. this is likely because the compression beams near the center of the bridge directly take on the vertical compression forces exerted by a weight on the top of the bridge. The load on these central, top joints is directly transmitted vertically to the bottom of the bridge, so we see that the tension strips near the center of the bridge are grey, indicating a small load. However, the tension strips near the edges of the bridge have a higher load as the bottom of the bridge pulls on them significantly. Because they are attached to the blue-colored compression beams, these compression beams also experience a high load transmitted through these tensile connections.
This analytic graphic from West Point University also shows the distribution of forces on the bridge. To calculate the on any member, you multiply the total load (mass * gravity) by the decimal on any of the joints. If the number is negative, it indicates compression instead of tension. For example, for a 100g load the very center middle vertical beam would be 1kg * 9.81m/s^2 * -0.167W = -1.64N so a compression force of 1.64N.
Analysis of potential weak and strong points
Based on the analysis and graphic above, the members that experience the highest load are likely to fail. Independent of whether the material is ductile or brittle, whichever members experience the highest load are likely to fail. These are compression beams on the top of the bridge and near the center and tension strips away from the center of the bridge. Based on observations of testing the bridge components (videos above), the compression beams tended to fail by folding near the center and tension strips tended to fail by ripping near the edges. Since the Bridge Testing Machine simulated the same compression and tension forces that the members will experience in the actual bridge, I expect that in the actual bridge these members will fail in the same manner. It's also possible the gluing will not hold and the bridge will fail due to detachment of the compression and tensile members from the gusset plates.
Testing the Bridge
Process and tools used to test the bridge.
We used two tables, a scale, six legos, and four large books to test the bridge. The process was as follows:
We moved two tables far enough apart to where only the four corners at the edge of the bridge were touching the table, as if the bridge was spanning a canyon.
We took six 2x3 lego pieces and put them on the six legos on the middle, top six joints so that the force would only be applied to those joints (and our calculations could use the method of joints).
We measured the books and split into groups of 3500g and 1503g. We first put the 3000g group of books on the bridge and counted to "3 Mississippi" then did the same with the 2003g group on top of the other group.
After 10 seconds of holding the 2000g group, we heard a crack noise, but there was no visible breakage.
Here's a video of the testing process (thank you Griffin Orsinger for helping!):
Analysis of failure points:
There were no failure points for our bridge.
Based off of the MOJ calculations (from the analytical graphics from West Point, also explained above), we noticed that the one of the members shouldn't have any load. If you see the end of the video above, you'll see us moving around that tensile member, and there was no load on it! Our calculations were successful!
See the "0" load vertical member on each side of this bridge. Independent of the load, this will always be zero in a perfect model.
Explanation of failure and future improvements:
Although our bridge did not fail, there was one structural issue. Before a load was applied, the bridge wouldn't sit flat on all four bottom points. This is because of a slight rotation during assemble. You can see this wobbling property at the beginning of the video above. However, once the load was applied, this problem was resolved as elastic deformation of the members ensured all of the points were on the table.
To solve this in the future, we could glue a smaller piece of cardboard to the bottom of one of the corners so that they are flush against the table. However, this concern is not urgent.
Final Conclusion
Summary of tools and processes used.
Tools/supplies (see sections above for specific lists for different processes):
super glue
large, medium, small laser cutter
gloves/tape for protecting hands during gluing process
cardboard outline/printout
CorelDRAW
books
mass scale
tables
legos
bridge testing machine
Processes overview (see sections above for specific workflows)
Design process of formatting SVG files for laser cutting
Laser cutting process
Assembly process
Member testing + analysis/calculations using Bridge Testing Machine
Bridge testing
Short summary of what was made, and what was tested and how.
Based off of a technical drawing, we made the SVG design files for all of the members of a bridge and laser cut the pieces. We then assembled the pieces according to the design. On top of making the actual bridge, we made three replicas of all of the types of members to test their theoretical maximum strength. We tested these using the Bridge Testing Machine, and used analysis based off of MOJ caluclations. We tested the actual bridge using books and a mass scale.
See the details for all of these processes in the sections outlined above.
What did you learn about civil engineering from this project?
Why is it important to account for both compression and tension when designing a bridge?
I learned the importance of theoretical models to predict a FOS for a bridge, as well as an intuitive understanding of factors that influence the compressive and tensile strength of members. The real world implications of a lack of foresight and preparation in a civil engineering project can be devastating and even have a death toll. In our bridge, we did not calculated a FOS before testing the bridge since we only found out the weight of the materials the day of the test. However, the extensive testing process with calculating theoretical load taught me very much about the properties of compressive and tensile materials, and I intuitively understood how cross-sectional area, width, material, and length can affect the compressive and tensile properties of a member of a bridge.
It's important to account for both compression and tension when designing a bridge because static forces require a net vertical and horizontal force of zero at every joint. If we only accounted for one type of force, it would be far more difficult to ensure that any joints, especially those at the ends of the bridge, would have zero net force. Furthermore, triangles are a very strong shape, as we've seen in a cardboard bridge able to hold 5kg. Trianges utilize both compressive and tensile forces to strengthen a bridge and handle a load. Balanced forces require compression and tension.
What improvements could be made to the design to allow it to hold more weight without collapsing?
We have yet to test the bridge until failure, however thicker, shorter compression members would allow the bridge to hold more weight without collapsing. From both the bridge-building manual and our Bridge Testing Machine experiment, we determined that shorter beams directly correlates with a great maximum load. The same is true for a greater cross-sectional area of the compressive members. Although shortening the compression members would require shorter tension members, we found that tension length did not impact the maximum theoretical load. We should also increase the width of the tension members, as our experiments revealed this will increase their theoretical load. Finally, if we build our bridge out of a different material, like steel, would also increase its strength, although this is unrealistic for such a project.