Tuesday, June 5, 2012

Week 9 term review

In our week 9 lab we tested our 3' bridge. It ended up holding 35 pounds before one of the bottom members seperated from the a gusset plate. We are content with this weight as it seems that we were pretty average with the rest of the class. In the next week we will all work to complete our A4 final report, which we will present in week 10. Overall I believe that our group did well in this class. Our 2' bridge held the most weight, and our 3' bridge did satisfactory. I feel that I have learned a lot about bridge design and the type of work that goes into designing a good real-life bridge. I have learned that the west point bridge designer program does a good job at designing a good bridge structure, but it lacks the ability to take into consideration the lateral forces that a real bridge will encounter. Knex were a choice for the materials for this project because they are similar to real bridge pieces but they lack the freedom to attach members at any requested angle. If I could add something to this course, it would be more opportunities to break bridge designs. I say this because it is the best way to learn what exactly is failing in the design, making it easier to improve the desing.

Week 9: Term Review

This past week the group tested the finalized 3' bridge in class. It held a total of 35.2lbs and costed $418,000.  The cost/load ratio turned out to be 11875 $/lb.  I believe this is probably an average ratio.  The group will be working on the A4 report to turn in for next class and it is coming along well.

I believe the group met the goals for this course and in the process learned a lot about bridge design and truss analysis.  Not only did the group learn about these things but they also learned about the importance of efficiency in engineering especially when dealing with the cost of a project.  I think the most beneficial part of this course was the experiment and calculations the group performed.  It really helped us get a hands on experience in designing and revising a plan for problem.  The least beneficial to me was the lack of things to do in the lab at times.  This was also a problem for the group because we would rather get out of the classroom so it was not necessarily due to the lack of work.  For future classes it would be important to make sure all designs make the specifications exactly as they are laid out.  It would also make things a lot more interesting to have the best groups make larger models of their bridges, like a 5' bridge, to see how the strength decreases for each bridge and how the cost/load ratios are affected.

A4- Fitzpatrick, Miller, Parker

  1. Background 
    The course was set up in way that was conducive to learning, There were not many restraints put on the students and there were clear checkpoint tasks that were used to gauge our progress and pace the course. This was geared toward the civil engineering majors giving them a taste in bridge design and insight into some of the tasks they might face in the work force, in addition it provided a basic introduction to concepts that civil engineers are applying to designs on a daily basis. There are various types of bridges that are used around the world; most of which can be seen in the greater Philadelphia area however in this course we only focused on truss bridges Through various tasks throughout the term we were able to exceed the course goals. We were introduced to the design process in which there are many ways of solving a problem; we were given the task of finding the best solution for the problem given a set of constraints.
  2. Design Process
    1. What goals you set at the beginning
      1. In the beginning of this project we wished to design a bridge that would adequately satisfy all of the necessary contraints and perform well enough to be considered a "safe" bridge  
    2. How they changed during the project
      1. Though we learned different key things about bridge design while constructing our bridges, our goals stayed the same. 
    3. The role of individual projects with WPBD, Truss Analysis, Individual Knex Bridge Designs 
      1. WPBD
        1. West Point Bridge Designer was used as our introduction and crash course in bridge design. Many aspects of this program are realistic while there are still parts of the program that are not so realistic; The blog post: Week 3: WPBD goes further in depth as to the strengths and weaknesses of WPBD. In working with WPBD one of the keys to success in lowering the cost of the bridge was getting the tension and compression forces in each member as close to 1 with out going over 1 as they could be. While we all had very similar designs the best design was Miller's. In the end the designs by Parker and Fitzpatrick were fairly close with all of the forces maximized. The goal of WPBD was to create a bridge with the lowest cost, when all of the bridges of the class were lined up our three bridges were the top three in the class.
      2. Individual Knex Bridge
        1. The individual knex bridge provided the stepping stones used to build up to the 36" span. For individual designs the constraints are as indicated in the link however the main concern for our designs was the 2' span. Each member of the group came up with a different design based upon the same general shape, this is one iteration: A2-Fitzpatrick . Each member of the group followed the same general rectangular frame with a different web design.
      3. Truss Analysis
        1. Truss analysis is used in order to determine the tension and compression forces acting upon the different members in the bridge. As can be seen in A3-Parker the truss analysis that we have completed closely relates to the truss analysis provided when using bridge designer for a simple truss bridge. When using truss analysis and bridge designer for a more complicated truss design such as our knex design the analysis was more complicated and it became more and more difficult to interpret the results as the complexity of the truss increased.
    4. How the final design was chosen 
      1. The final design was chosen based upon the results of our bridge from the individual bridge design . This design was a major part of our finished design which held the most in the class with 49.2 pounds. Moving up to the three foot bridge we did expect a decrease in the load that the bridge could handle due to a longer span without supports and the requirement for a 2" x 3" opening through the bridge which eliminated the cross bracing.
    5. How it was modified during construction - if at all 
      1. Modifications during construction include the addition of cross members on the top and the bottom of the bridge to help prevent the bridge from racking and to spread the load more evenly across the entire span.
    6. Predicted load at failure 
      1. 30 pounds, this prediction was based upon testing designs and modifying them along with the use of bridge designer software to show where the point of failure would most likely occur.
3. Description of Your Final Bridge
  1. Plan View
  1. Elevation View
  2. Bill of Materials
  3. Cost 

    1. The cost of the bridge as calculated in the Bill of Materials above was $418,000.00
  4. Photographs of the bridge



4. Testing Results
  1. Load at failure
    1. The bridge failed with a 35.2 pound load.
  2. Describe the failure mode of the bridge 
    1. The bridge failed at one of the lower gusset plates, the knex bars pulled out under a load of 35.2 pounds.

5. Conclusions About Your Bridge Design
  1. Did it behave as predicted in terms of load and failure mode
    1. Our bridge was able to hold more weight than we had expected, The design was expected to hold 30 pounds and during the testing our design held 35.2 pounds exceeding our expectations by 5.2 pounds. The mode of failure was separation of the gusset plate on the lower edge of where the load was applied. The location of the failure occurred in the location that we were expecting failure. It failed graciously, it was not an explosive failure it was a failure that would allow vehicular traffic to safely vacate the bridge.

6. Future Work
  1. How would you modify your bridge if you were to design another version 
    1.  If we were to modify the bridge to design another version with the same goal of lowest cost to weight ratio in mind then the bridge would take a completely new form with added internal cross supports to help spread the load.
 
Failure of the 24" Bridge Span

 
Failure of the 36" Bridge Span

Wednesday, May 30, 2012

Week 9 Term Review


This past week we completed our final three foot bridge span and tested it. In the coming week we will work on A4 and submit it before class in week 10. The only major accomplishment this week was that our bridge was able to hold 5 more pounds that we had anticipated. We are facing no major issues as a team.

I think that each goal was met and exceeded through the course, if I were to pick one goal that I wish we had more time to explore or would have gone more in depth with I would say it would be forensic analysis. The least beneficial goal was the planning, I understand the need to plan and using the tools we were given at the end of the course I can see why planning is useful however going through the course I thought that planning was not beneficial and most of our discoveries were obtained by trial and error. Once we had these discoveries and were given the mathematical tools our discoveries were confirmed and we were able to understand why. I also feel that bridge designer was good for the simple bridge but it did not meet my expectations when it came to the group knex design. The most beneficial parts of the course were the physical modeling and teamwork aspects because we were able to test a variety of ideas and then collaborate to make what we felt was the best design. In the future I would give a better introduction to WPBD also I would have a block of wood to resemble the car to make sure that all designs meet the given specs.

Tuesday, May 29, 2012

Week 8: Bridge Process

This past week the group built a bridge that spans 3' and has a hollow tube of 3"x2" running through the center of the bridge to meet the criteria for the second KNEX bridge design.  Using what was learned from the truss analysis the group put together a bridge to use for the test.  The group did some practice tests on the design and made changes to better the bridge.  This next week final adjustments will be made to the bridge and it will be tested using the same set up as the 2' bridge.

Each week this term was a learning experience.  I have had some experiences with trusses and how they work through building buildings but this course certainly taught me many things about trusses and bridge designs.  I realized how important a efficient cost ratio is for engineering projects.  WPBD helped me to get started with this concept, but the hands on experience with the KNEX certainly showed how many ways a goal can be met and how different the cost ratios can be.  The truss analysis helped me to better understand how trusses displace a force.  This will be helpful for me moving forward, and I feel that it gives me an advantage over other students who have not had this opportunity yet.  Through the movies and background history provided in class I realize the importance of engineering and safety.  One little mistake can cost hundreds of people their lives.  It certainly makes me double check my work when I am creating designs and ideas.  After learning all this it makes me wonder how the bridges that are in existence now will hold up with the weathering and increased loads applied to them by traffic and things done in an effort to help preserve them.  There is a trade off in safety measures and added weight and cost.

Week 8: Bridge Process

In the past week our group has been working on building the best 3' span bridge possible. We have experimented with many different design options and we think we have found one that will perform  satisfactorily. In our week 9 lab we will test this bridge to see the results of our efforts. Through designinf these bridges I have learned that the bridge design process is not as simple as throwing together a structure and seeing if it holds up. I have learned about the different loads that a bridge faces over the course of its life such as compression forces and tension forces resulting from the load, and also lateral forces caused by elements such as wind. Some of the main things that i have learned though is that the one smallest issue with a bridge can cause catastropic failure. Bridge designers learn from these failures though. While hopefully these failures do not occur on a full scale active bridge, the use of breaking scaled down bridges is the best way to learn about how to improve them.

Wednesday, May 23, 2012

Week 8 Bridge Process


This past week we have worked to complete a bridge designed to span a 36” opening, and worked to determine ways that our bridge could be improved using the analysis tools we have at our disposal. This coming week we will work to further improve our design to have the best possible bridge ready to test in week 9. Overall I would say that we had no major accomplishments this past week. As a team the only issues that we faced were disagreements on how to improve our bridge and prevent it from racking.
Throughout the term, each week brought something new to the table and something that I had not seen before. We started out using modeling programs that helped us to learn to maximize the tension and compression forces that a gusset plate feels. In maximizing the forces that the gusset plate feels you are able to use the smallest size members possible, in many cases this lowers the total cost of the design.  In looking back at the videos the bridge failures shown in class it is clear that a bridge designer plays an important role in keeping people safe, if even one small aspect of the bridge has a flaw it can lead to a catastrophic failure and significant loss of life. I have also learned that when designing a bridge that is most efficient, that the bridge should be able to handle a load that will change over time; just as we have seem with local bridges the weight of paint is added on as the years go by and our vehicles continue to increase in size as a result increasing the load that the bridge must carry. Bridges are subject to more forces than meet the eye, there is the typical gravitational force but what many people don’t realize is the tension and compression in each member as well as the overall impact of the elements and external factors on the bridge.

Tuesday, May 22, 2012

A3: Fitzpatrick

1) Calculations of forces in the truss members using the Method of Joints

Truss bridge being used in calculations




The figure above shows the calculations neccessary to obtain all of the forces on the truss. They are place into an Excel doc for easy accessibility 


2) Truss analysis results


Note that the forces are equally and evenly spread across the bridge


 3) Truss analysis via Bridge Designer


4) In order for the forces on Bridge Designer to match the forces from the method of joints, one must correctly scale the bridges to the same size. If either bridge is longer or shorter than the pther in any way, the results are not going to be the same, because they are simply different bridges

5)


Once again, if this bridge in Bridge Designer is scaled exactly the same as out Knex bridge, the forces on each of the members should be near equal. If the design has been sone correctly, the forces from one side should be symmetrically equal to the forces on the opposite side. While this is usually true, something must have happened when creating this bridge in Bridge Designer, because it continues to show a force of zero on one section of the bridge, which cannot be possible while under a load from the center of the bridge.


6) The method of joints analysis is a good way to maually determine an idea for a load capacity for a bridge. breaking the desgin down into single specific pieces gives one a better chance at not making a mistake. One of the most vital things we can do to assure a better bridge is to use gusset plates with at least 3 joints on them. These joints are stronger than all the others when it comes to the load testing. While these are more epensive than other joints, they are well worth it in the long run. This program is  help when it comes to doing this.

A3: Miller

1) Calculations of forces for the truss members using the Method of Joints.

Truss Bridge used for the Method of Joints
This shot shows the calculations done in excel to determine the forces on each member.

2) Results of Analysis
Forces Acting on Each Member
Members symmetrical to one another have the tension and compression force acting upon them.

3) Replication of Bridge Analysis done in Bridge Designer

4) In order for the hand analysis to show the same results as the Bridge Designer calculations the bridge must be scaled to the same size.  In Bridge Designer each individual square is 2"x2".  The drawing that I did in the Bridge Designer does not have the same lengths as the hand drawn one I did the calculations for.  It does however have very similar results.

5)
The forces in the KNEX bridge will be fairly similar to the forces calculated by the bridge designer analysis.  The member lengths are a little different and there are less sections than the real bridge.  The forces felt by the 2 middle sections should be relatively realistic when compared to the KNEX bridge.  The 2 outside sections are showing unrealistic numbers.  While this is not the full bridge formation it should give us the general idea we need to help improve our bridge design.

6) Using the method of joints analysis allows the determination of tension and compression on individual members of a bridge for a given load.  Using this information it is possible to apply the KNEX connection strenghts to determine a failing load for the bridge.  The connectors with 3 members attached to them have the highest pull out force at an average of 35.6lbs.  Connectors with only 2 members attached suffer almost a 10lbs. decrease in strength which leaves them at 26.5lbs.  The weakest attachment to a connector is 1 member.  The pull out force for this is 20.7lbs.  With this data it is obvious that connectors should be loaded with at least 3 members to assure the highest pull out force.  There is also a possibility to determine pieces that are not giving the bridge any structural support by analyzing the results of calculations and using the pull out test information.  These pieces can be removed to decrease cost.

A3- Parker



1) Calculations of forces in  the truss members using the Method of Joints.
Truss Bridge That was used in Method of Joints
This image shows the calculations that were made to determine the forces on each member, these formulas were derived by hand when solving the truss and were imputed into Excel so that hand calculations are not necessary as the bridge constraints change.

 2) Results of the Analysis above

As you can see above members symmetrical to one another have the tension or compression force acting upon them


3)  Replication of Bridge Analysis using Bridge Designer




















4) In order for the hand analysis to correspond to the analysis compiled by Bridge Designer one must scale their bridge so that the forces correspond with the values calculated by Bridge Designer. In the drawing above, each block represents 2" in length, and the nodes correspond to the connecters on the hand drawn bridge so the bridge in the image corresponds to the bridge that the analysis was preformed on, not surprisingly the results of the two analysis's are similar.

5)

The forces in the knex bridge should be similar to the forces in the bridge designer analysis. The member lengths are similar therefor the forces felt should the similar for the center two sections. the forces shown in the diagram for the sections on the outer sides of the diagram are showing non-realistic forces. The bridge design above while not a full bridge, provides a glimpse at the final product, a bridge constructed using the same sections as shown above held 49.4 lbs, this analysis is showing forces in that are far beyond the tolerances of knex with a smaller load. Also the load that I used for the knex bridge was 15 lbs, when creating an analysis multiple times I was shown a picture that did not seem right, tension and compression of zero on one section and on the symmetrical section the forces are to strong to be possible.


6) Using the method of joints analysis one is able to determine a number for the tension and the compression of members of the bridge for a given load. When viewing the information about the strength of knex connectors one is able to use the analysis to determine under what load the bridge is expect to fail and at which connector. Looking at the tensile pull out force chart it is clear that connectors with three members attached to them have the highest pull out force (avg 35.6lbs). Connectors with two members attached to them have an average pull out force of 26.5 lbs this is almost a 10 pound decrease in strength from the connectors with three members attached to them. Lastly the weakest connections are the ones that only contain one member; these have an average pull out strength of 20.7 lbs. Using this information the goal of the bridge design is to maximize the number of connectors that have three members attached while reducing the number of connectors with only one member attached. Using the analysis provides the number of pounds of force a given member has acting on it and paired with the information from the table it should now be possible to determine what members in bridge are adding cost and no structural advantage.

Week 7: Analysis Process

During the past week we experimented with some 3' bridge designs, in an attempt to come up with the best one possible. in addidtion to this search, each of us are also working on a truss analysis to help us better understand the loads and forces that a bridge goes through. By doing this, we will be better prepared and more knowledgeable when creating our 3' bridges. In the nextr week we plan to revise our plans and come up with a satisfactory final design that will hopefully perform well in the load test.

The method of joints system is a good way to better visualize and understand the physics behind a bridge design. It helps point out potential weak spots in a design and then show why it is weak. One of the biggest factors in a good bridge is its ability to resist outside force, such as wind, in addition to the loads traveling across it. This is something that the method of joints system lacks. Though this system may help one develop a structurally sound bridge, it cannot help in the question of how it will withstand outside force. If a bridge cannot handle all of these outside forces the design is no good.

I would like to see the effects of a bridge in the outside world. It is the only sure fire way to know if a bridge is going to make the cut. Obviously this is not possible to do in a full scale experiment, but it would be very intriguing to be able to run some sort of proportional small scale test. And even so, this is not a fully fail proof experimentation because it will still be hard to test the lifespan of a bridge. Not only does a bridge design have to hold up to brutal conditions, but it also has to last a long time. Finding a way to do this would be the only way to sucessfully perform a full test on a design.

Week 7: Analysis Process

This past week the group worked on the 3' KNEX truss bridge.  In this process each member is also working on a truss analysis for a sample bridge and for our own KNEX truss system.  This should help us analyze the truss and make adjustments to strengthen it.  This next week the group will look deeper into the analysis and come up with a final design for the groups 3' KNEX bridge.

The method of joints is a great way to get a general idea of how a truss system works.  I feel that it would not be a great way to analyze a large scale bridge.  Nature plays a large role in affecting how a bridge holds up.  The method of joints does not take into account the lateral movement perpendicular to the joints.  In a real world situation the wind would cause a large difference in these calculations.  On a small scale and to give a person a general idea of how a truss works the method of joints is a great tool.

I would like to be able to analyze how a moving load affects a bridge.  It would also be useful to be able to see how the material of a bridge decays over time.  Bridges should be built to last a long time and be safe.  It is important to see how the bridge would deteriorate over time.  Including wind forces would also be a large benefit.  In a real world bridge this would have to be calculated.  It would be useful to see what needs to be added to our bridges and ideas to make them actual models of a real life possibility.

Wednesday, May 16, 2012

Week 7: Analysis Process


This past week we have continued to work on our final knex bridge design, we have also been working to complete a truss analysis of a sample bridge and apply the same logic to our knex bridge to improve our design. In the coming week we will continue on the truss analysis and design of our knex bridge. This week we had no major accomplishes. The only issue that I see is related to post A3, I don’t believe that Bridge Designer will allow us to submit what Dr. Mitchell is asking of us.

I believe that the method of joints is sufficient for small scale or low load bridges such as foot bridges across a wet area. I do feel that while it provides good insight about what should be changed in a bridge design it does not take into account all of the forces acting upon the bridge. The method of joints should work great in a situation where effects related to nature are negligible; wind is a major factor in bridge design. The bridge needs to not only stop the wind from pushing it over but it also needs to be able to support the load of cars while wind is acting upon the bridge.
I would like to analyze how a bridge preforms over time with an increasing load on the bridge. The load of the bridge changes each time the bridge is repainted or the bridge deck is surfaced, the load also changes as the vehicles we drive change. A bridge does no good if it only lasts 5-10 years because there is no tolerance for maintenance to be performed.  As  the vehicles we drive change so does the load on the bridge and how the bridge handles on windy days. A bridge of cars is going to handle different on a windy day than a bridge full of tractor trailers since the trailer are going to be more impacted by the wind. I feel that we need a method that incorporates weather related stresses as well as stresses ove time.