Elevated Bridge B/C

[andrewwski]

After I couldn't find any websites to help, I asked several physics teachers to help me calculate forces. They were unable to calculate the forces, and interpreted my current bridge in same way I did by looking at which pieces were in tension and compression, how the load transferred, etc. Simply put, calculating truss forces isn't feasible for most teams, especially if one considers the limitations of these truss programs and the countless variations in design this year. Unless you are a physics wizard or have a helpful contact, calculating these forces are exceptionally difficult.

Really? It's not that hard to calculate them all by hand...it's all vectors and moments of inertia. We calculated them many times in my Principles Of Engineering class. Yes, it's a pain in the neck and very time consuming, but it can be done. I'm surprised that your physics teachers couldn't do it.

Designing a decent bridge without analyzing the forces involved is certainly possible, but much more difficult. Without knowing the forces in each member and joint, it makes construction more like guesswork, and analyzing the failure is done with much less certainty. Having an idea of what would happen under ideal conditions at least gives you some understanding of why something happened. Not that it can't be done without that information, but it's a lot more trial and error than needed.

Now, it is important to build, break, and learn from there when rebuilding. I only made one bridge design this year on paper, but by building, breaking, analyzing, and rebuilding, was able to get about 2.5 times the efficiency the original bridge had. But I used the truss calculations extensively in my analyzable. If you were to look at my first and last bridges, they look very similar, however, little things such as a particular member size and stuff can make a huge difference.

I don't think I would have necessarily had a bridge competitive at the national level (my efficiency peaked out at just under 1200) but I certainly was able to take a design and improve on it, and understand why it worked the way it did.

[dragonfly]

I think another thing that was crucial to our success this year was the countless changes that we made to our design. If anyone saw our original bridges, they would have no way of knowing what our nationals bridge looked like. While possibly if we'd been concentrated on one simpler design like Andrewwski, (possibly for one of those original bridges) that may have assisted improving that design, making a better, and essentially different, bridge, required changes that a program could not have told us to make.

[JimY]

We certainly have two distinct camps here. Both are valid ways of working on the problem. Just from a time efficiency viewpoint, designing via programs is much much faster. In a post from last year, I mentioned that both our B and C division bridges were designed in October and not changed after that. If our high school team would have made it to nationals, it would have placed 3rd in the event with no changes past state, so we may have slipped closer to first there with a bit more work. It was a series of 7 interconnected triangles, just like the B division winner. For B, we started the season with efficiencies just over 2000 and ended up at nats at just over 2500. For C, we started the season with efficiencies of about 1600 and ended up at just under 2200.

Designing via a program allows you to make small tweaks in the position of the nodes and see the effect on the stresses of the pieces in just seconds. For example, one node in the B division design was moved around such that one of the chords had a slight amount of tension on it without significantly changing the load on the other chords connected to the same nodes. This allowed a smaller piece to be used than if the piece were in compression. You can get that from trial and error, but at 5-7 hours per build, why spend that much time when seconds will do?

After the design was finalized, the trial and error work began. It was just a matter of how little mass could we get away with and not have it fail too early in the loading. This took something on the order of 10 iterations for the B division winner at nats. We did about 6 iterations for the C division state design above. We tried using significantly more balsa for both divisions, for example, but rejected it for both divisions.

Back in the 2000 season (coaching B only), I was firmly in the trail and error camp for the event. This was because I hadn't really though about using a program to help. Then one of the other parents asked to take one of our designs to his work and put it into his $15000 program that also calculates deflections on pieces (he was part owner in a company that builds roads and bridges). The programming idea stuck, and I found the website given in my previous post on this thread. So, after that season concluded, I decided to put together my own structural design program in Excel (which uses only high school level statics from Physics class) and calculated the stresses on all the trusses that could be calculated on what we built during that season and compared this to where we thought failure occurred. It was a true breakthrough. This allowed our workload in the event to go way down. Despite spending far less time working on the event, we ended up similar results. So, my heart goes out to all the teams that do this event by trial and error because we were there at one point too. Just getting through that paradigm may be very tough, since you are not engineers after all. However, if you manage to get through it, the other side is like a huge breath of fresh air. If you're planning on an engineering career, I strongly suggest adding a truss program to your repertoire for the event. It could be part of your legacy for your team, as it can be used for years to come.

[baker]

JimY, How are you deteriming what a piece of balsa at a given dimension is capable of in compression or tension? Yes a piece of steel is consistant, but no two pieces of wood, even cut from the same parent piece are the same. So I wonder how a program helps with unknowns, or do you do a zillion of tests on different sizes both tension and compression?

[jander14indoor]

Wood selection is a KEY element of this event! This is true whether you are using a trial and error approach or a analytical one. See last years discussion and search using "wood selection" as your key words and you will quickly find several in depth discussions on the topic.

For the analytical approach, yes wood varies, but over a limited range. The analysis tells you what you need, and then you select the wood to meet the requirements. This is NO different than in real life engineering! The end criteria is different, but the approach is the same.

Example, ceiling trusses. These are commonly made of wood and by code have to meet certain load requirements. In order to stay in business, companies use analytical programs to minimize the wood used to save costs, and have to deal with wood variability in doing so.

For the trial and error approach, this is just as important. Unless you select wood for consistent properties, how do you get a design to repeat its performance from one trial to the next? If you can't do that, you are just shooting in the dark and making design changes based on random changes in wood behavior, not truss layout.

Jeff Anderson
Livonia, MI

[JimY]

baker, we use almost all bass and very little balsa. Read my previous posts on this thread. Bass is far more consistent than balsa and reduces trials to very manageable numbers.

So, we have 2 major paradigms in the event. Engineered trusses with interconnected triangles or not, and bass or balsa.

[SLM]

You should probably spend some time on the following website to see some of this for yourself.

http://www.jhu.edu/~virtlab/bridge/truss.htm

Here is a bit more user friendly, and probably easier to use Web-based structural analysis program.

http://www.nexote.net/nexote/Structural ... 0Notebook/

[jander14indoor]

<SNIP>
Here is a bit more user friendly, and probably easier to use Web-based structural analysis program.

http://www.nexote.net/nexote/Structural ... 0Notebook/

I like that one! Thanks for sharing.

Much more complete.
Deals with fixed joints, more representative of SO structures.
Material properties, flexing, etc.
It could analyze those SO structures that aren't fully triangulated and tell you something about what's going on at the joints.
About the only thing its missing is buckling. Unless I didn't dig enough since it did consider member cross section. Hmmm, checking. Nope, no stiffness input so buckling must not be part of the program. But that could be done off line pretty easy.

Jeff Anderson
Livonia, MI

[Cyrus_D]

Jumping in to the conversation. Woo!

Do you guys use sand or water to pour into the bucket, or if something else, what? And also do you guys bring like a measure to the competition to make sure you have the loading bock exactly where you would want it to be? I would also like to have a talk on weather you should use heavier wood or lighter wood? Personally I use lighter wood most of the time but when I build my bridges out of heavier wood the joints don't tend to fail as much and it is more likely that a member would snap. So is there a reason behind this??? On my latest 2 bridges they are exact replicas, except for the fact one uses lighter grade balsa and the other heavier balsa.The lighter one broke at a important joint and it's efficiency suffered greatly. But the heavier one didn't have that problem. I believe that if the lighter one didn't break at the important joint it would greatly have a higher efficiency of the one the heavier one, but because it did it suffered. So opinions? Take the risk of it breaking at that joint, or take no risk but not get as high of efficiency.

Cyrus D

[nejanimb]

I always use sand. It's what competitions have to use, and water just seems like it'd slosh around and such.

I don't bring a measure to competition for the loading block - we line it up really carefully before competition, and make a pencil marking on the bridge for exactly where the loading block is supposed to go (tracing it after we line it up carefully), so that it's quick and easy to line up at competition.

And... as for heavier or lighter, I usually go lighter. But, really, the important thing is to match the density of the wood. That doesn't mean to have it all the same density, but to have it all so it is as heavy as it needs to be for that particular part of the bridge, so that, in theory, the whole bridge would shatter at the same second.