Refining Bridge Design

I've had two good questions about refining your bridge designs that may be of general interest.

Q1 - Why does an "X" type arrangement fail when using the Hopkins "Bridge Designer".

Response:

Probably the best answer is that you have created a "redundant" structure with an "X".  You can test that by removing one leg of the X and probably it will work as in the example above.  If you add the member from the lower left to the center of square so you have a complete "X" Bridge Designer won't calculate it because you have an unnecessary or "redundant" member.

Q2 - Additional Analysis

Throughout the project, I have been replicating the Knex bridge on WPBD as an extra load test. But thinking about what you said today after lab, I realize that's not the best way to test our truss. What at-home methods do you suggest would be best to see if the truss is stable?

Response:

Unfortunately you're running up against the limits of the tools you will understand in your first year.  There are indeed many tools for additional analysis and testing, but to use them effectively you need the knowledge that you'll gain in your more advanced courses.  WPBD and Bridge Designer have to make simplifying assumptions that your real bridges violate.  Let's hope this is your incentive to continue learning.

An Interesting Discussion

I had an exchange with a student who was unhappy about our policy of being timely with the submissions.  It led, I’m happy to say, to a more fruitful discussion about the differences between the WPBD software and the Knex experience.  Below is a portion of that exchange in chronological order with my latest response added via this blog

JEM

I did read your blog entry and appreciated your thoughtfulness about the difference between a Knex and a “real” bridge. I’d add to what you wrote that there’s a significant difference between WPBD and Knex in that WPBD is 2-D only (no torsional effects considered). In that system the gusset plates are perfect connectors as well – which is hardly the case with the Knex

Student

I would like to to thank you for taking the time to read and provide feedback regarding my worthless blog entry. While I did write that the differences are endless, (thankfully in this situation did not waste my time in further elaborating on these due to the fact that it is now resulting in a zero) I did like the specific portions you added and for the most part agree with them. I agree very much with your statement about the gusset plates in WPBD; they can be connected at any desired angle and are also never a point of failure, making them "perfect" connectors. While you are correct in saying that WPBD does not consider torsional effects, I would have to say the same goes for the K'Nex as well. The only reason that bridges in class were experiencing this twisting effect was due to the fact that the applied force was attached to the uppermost portion of the bridge. In WPBD the loads (dynamic in this case, unlike the static type used on the K'Nex bridges) are being applied at the same elevation as the bridges anchor points, not allowing for much torsion on the bridge. In the lab the bridges were tested by having the load applied from a point higher than that of the anchor points, causing the tall flimsy bridges to twist and fail. The use of cross members is another large portion of this, but for the sake of time and possible material for this weeks blog post I will end my thoughts here.

JEM – New Response

In any real bridge the loadings are always uneven.  The torsional effects are thus ones that must be considered and were important in your Knex designs, leading to failures in several cases.  In WPBD the single truck crossing a two-lane bridge would have asymmetrically loaded the bridge and thus torsional effects would have mattered, but were ignored by the software.

Bridge over CSX Tracks–Photos

You can see the photos I took this morning when the bridge was put in place over the CSX tracks.

Picasaweb Photos

Truss Bridge Installation Wednesday Night and Thursday AM May 2nd and 3rd

Here’s a news release from the Schuylkill River Parks Alliance that may be of considerable interest to this class.  It’s tonight and tomorrow morning!

The delivery of the truss portion of the bicycle-pedestrian bridge that will span the CSX tracks between Locust and Spruce streets is supposed to be delivered overnight Wednesday, May 2nd into the early morning of Thursday, May 3rd. The Philadelphia Streets Department will use the Swift Reach  telephone system to notify residents along the route. The contents of the message is below in italics.

Rockport, the contractor, expects that it will take approximately one hour to assemble the truss Thursday morning, then it is going to set the bridge into place between 8:30 AM and 9:00 AM.

The telephone Swift Reach message from the Streets Department:
Philadelphia Departments of Streets and Parks & Recreation is pleased to announce the delivery of the new Truss Bridge that will soon connect the Schuylkill River Park between Spruce and Pine Streets to the Schuylkill River Trail. Truss bridge is scheduled to arrive at the project site on late Wednesday night into early Thursday morning. Bridge will be escorted to the project site by Philadelphia Truck Enforcement unit and transported on two (2) oversized hauling trucks. In order to safely make the move, temporary parking restrictions will be posted along the trucking route, and will be effective from Wednesday, May 2nd at 10pm through Thursday, May 3rd at 7am. No parking signs will be posted at the intersections of Washington Avenue and 22nd Streets and 22nd and Locust Streets, as well as along Locust Street between 22nd and 25th Streets.

Real Bridges – California’s New Bay Bridge

This video from Science Friday isn’t directly about trusses (but notice the use of trusses in the construction), but it’s interesting to see the kinds of things that engineers incorporate to make bridges safe.

As the video indicates, the previous bridge in this location failed during the 1989 Loma Prieta earthquake.

Note the time scale for big projects of this sort – it’s 23 years since the failure and the replacement is just about to open.