Bridge Aerodynamics Project Ideas

[Cyrus]

Schlieren is used when you have density variations in the flow. For what you are trying to do, you won't get any clean pictures. For flow viz of this type, you would typically use propylene glycol and a wand.

 

[Gordy]

Is that (propylene glycol) a road I can practically move down, or does it get really complex really quickly?

And wouldn't the compression caused by the stream hitting the bridge components affect the air's density? All of the Schlieren imaging I have seen has been heated-based, so of course I'm skeptical.

 

[Cyrus]

Is that a road I can practically move down, or does it get really complex really quickly?

Easy as dirt. They should have a smoke wand for you to use at the tunnel. When you call them up tomorrow just ask.

 

[Gordy]

Fabulous, seriously (Good God, thank you Cyrus). Now, I have three possible directions to move (At least, that's how I see it):

1. Place multiple common types of bridge span in wind tunnel, observe how they interact with flow, and isolate features in each that contribute to stability. (Possibly combine into new span design?)

2. Choose only one span type and observe how different weight distributions/traffic presences affect flow (This might lead to some interesting places involving controlling traffic flow during wind events instead of shutting a bridge down altogether, but that seems doubtful)

3. Work with modifying the actual imaging method (?)

 

[FredGarvin]

Smoke wands and tassels are most likely all you're going to get. Like Cyrus mentioned, low speed stuff will not show up well in other forms of imaging because you don't get very large density differences (plus it is time consuming and pricey to set up). If the tunnel has it, you may ask about high speed video. That can be very helpful, especially when looking at wakes.

The thing that is going to eat up a lot of time is your proof that your model is not going to harm the tunnel at the flows you expect. In some cases I have worked, we had to design for a case that was not even physically possible. Again, this is another point to bring up with the tunnel folks.

Do you have to stick with bridges? That seems like a tough model to make accurately. Can you switch it to something easier like smoke stacks? Then you could really get into shape variations easily and you could see real Von Karman vortex shedding in action.

 

[Gordy]

Do you have to stick with bridges? That seems like a tough model to make accurately. Can you switch it to something easier like smoke stacks? Then you could really get into shape variations easily and you could see real Von Karman vortex shedding in action.

I completely agree, I don't want to overextend myself in terms of just getting the model built; however, it seems like if I were to do something like smoke stacks (i.e. focusing more on the flow phenomena than on the structure itself), that would be creeping into fluid mechanics, an area in which I have absolutely no experience (Remember, high school here, I'm in the most advanced calculus class I can take but we're still just now learning improper rational integration).

What I like about the bridge idea is that I can definitely form some kind of hypothesis (the traffic density/speed across the bridge has x effect on the bridge's flutter/vortex trail/etc.), even at this stage, and precision model construction is an area of particular expertise for me.

The thing that is going to eat up a lot of time is your proof that your model is not going to harm the tunnel at the flows you expect.

Are you saying that I'll have to prove the strength of my model? I'm not sure I know how to do that I mean, I know for certain that I can build a model that is more than strong enough (built a 6-ounce balsa wood bridge that withstood 70 vertical pounds), but I wouldn't know how to theoretically prove it. Again, this isn't like a separate tunnel facility or anything, it's just a big ol' thing built into the engineering department.

 

[Gordy]

Right now, I'm thinking of mapping the airflow (by whatever "mapping" methods I can find) over the cross section of the empty bridge, then progressively imposing traffic (possibly moving, if I can find a way to do that) and observing the effects of the added bluff bodies. Sound decent?

If this is totally bunk, please let me know, I have no time to lose and I would really appreciate it :)

 

[dr dodge]

for the bridge models, could you purchase model train kits and assemble those.
you can get some very nicely detailed plastic kits in scales starting at N scale (1:160) to G scale (1:22.5) this would mean accelerated model completion, based upon actual prototype bridges.

good luck
(and as a fellow procrastinator let me say...ahhh i'll tell you tomorrow)

dr

 

[Cyrus]

for the bridge models, could you purchase model train kits and assemble those.
you can get some very nicely detailed plastic kits in scales starting at N scale (1:160) to G scale (1:22.5) this would mean accelerated model completion, based upon actual prototype bridges.

good luck
(and as a fellow procrastinator let me say...ahhh i'll tell you tomorrow)

dr

How is that going to span across a 6' WT?

 

[dr dodge]

Why would you have to span the whole 6'?,
If you can work in smaller scale (vs 1:1)
6' in 1:22.5 gives you a 135 ft bridge
in 1:160 that's 960 ft.
stryofoam for the "terraformed" approaches, and now you have a complete structure, based upon an actual prototype, so the experiment has meaning, and can be compared with actual results.
here is a link to some actual bridges

dr
http://bridgehunter.com/category/tag/span-length-500-1000-feet/