hey everyone, I just wanted to get your ideas on something.

in my statics/solid mechanics class, we had to build a bridge for homework, and I think my bridge was good. however, I basically just built it in a program calls AnSys, which is commonly used to truss analysis, and then through a trial and error of different designs to figure out what was best. the whole time I was thinking to myself "there has to be a better way" and I even had a few ideas of how I could model it to make a low cost, yet strong bridge, but none of the ways worked.

so basically, wondering if anyone knows how to optimize a bridge design where you have a cost per kg of material (given three different materials to choose from) and a cost per joint. if you need an example, I can explain the homework I did, or just send you the assignment. I already did it, so you don't have to worry about doing my homework for me. however, if you are still weary, just make your explanation broad and general.

I would prefer a mathematical model for the optimized number of members, and their placement. I already know how to optimize the force each one carries.
http://www.siteofthefuture.com/cato.bmp
we had to build a support under it, and the bridge was only attached at the left side, the other was a roller.

You have to be kidding us. You are, I assume keeping up in your class but this is a mystery? Give me a break. :bugeye:

I take offense that you think I am trying to get people to do homework for me. I don't even want anyone to solve the problem, I am just wondering if there is a way to optimize a truss based on the constraints. I already built the damn thing in ansys, its done, I just wanted to know if there was a mathematical model to give what is best. its a sad day when someone jumps all over someone else for wanting to learn beyond what is being taught in school.

how would I know the best number of members to use? how would I find where to attach them? (obviously at the points where forces are applied) can anyone point me in the right direction? I assume there is a way to do it with calculus, but I am having trouble modeling it.

Cato, you've chosen a problem/project with an infinite number of possibilities for the truss members. I don't know if you know much about truss design, but it's extremely complex. Most, if not all, engineers that I know use one of several "standard" truss types and most, if not all, computer programs are based on only those few "standard" designs.

I can't tell from your post if you've used one of those standards or not. Nor could I tell if you have a computer truss design program. And besides, I don't think anyone here would be able to answer a complex truss design problem in words only.

Baron Max

ok, thanks

download a game called 'pontifex 2' it used te be called bridge builder. its a simple to use and technically accurate simulation game. greta for finding cheap ways to build strong bridges

Do a simple truss analysis.

For m members, n pins, and r unknowns, m + r = 2n for a statically determinate system of two-force members on a fixed point at one end and a roller at the other.

Usually, statically determinate systems are most efficient as defined above. We'll just accept this as a valid assumption and using the paradigm you have given above in the first picture (4 point loads), the Warren truss might fit best (m=15, r=3, n=9).

In the math, it doesn't matter what limit you choose for the limit at which a member breaks. Just inspect the ratios and adjust the geometry accordingly, which I believe you may already be aware of.

I take offense that you took offense to my post.

Do you now take offense that I took offense?

If so, I now take offense that you take offense to my having taken offense to your taking offense!

I can only remark that you were certainly defensive to take offense in the first place.

Although, I have to be honest and admit it, I initially took offense, though did not state so, prompting me to make my first post. :mad:

P S : Until I retired, this is a kind of problem I encountered too many times to count every working day. This is a relatively very simple type of problem usually solved by using very simple formulas in the "steel book" and then choosing the best standard steel shape from the tables. Steel shapes are sold by the pound. So, we look for the one that permits no more than the allowed deflection, will carry the calculated load, and be the lightest shape ( per foot ).

want to get really new-wave sciency....try looking into carbon nanotubes for bridge constructions. supposedly, one day you'll be able to build a suspension bridge across the atlantic.

I take offense that you took offense to my post.

Do you now take offense that I took offense?

If so, I now take offense that you take offense to my having taken offense to your taking offense!

I can only remark that you were certainly defensive to take offense in the first place.

Although, I have to be honest and admit it, I initially took offense, though did not state so, prompting me to make my first post.
sorry if I was a bit defensive, I give people a hard time when they ask homework questions here, so I didn't want to look like a hypocrite.

P S : Until I retired, this is a kind of problem I encountered too many times to count every working day. This is a relatively very simple type of problem usually solved by using very simple formulas in the "steel book" and then choosing the best standard steel shape from the tables. Steel shapes are sold by the pound. So, we look for the one that permits no more than the allowed deflection, will carry the calculated load, and be the lightest shape ( per foot ).
can you tell me more about this book? I would like to know just what bridge would be the best, the one I turned in was far from an ordinary truss. however, when I tried to build an ordinary truss, I always got a higher price than my jerry-rigged design. perhaps I simply didn't pick the proper design for the job.