I know it’s not like this:
So how?
IANACE. But my first thought is modeling based on material properties and dimensions. Having said that…
About 20 years ago I was involved in a project that measured strain and deflection of bridges. We would instrument a bridge with a couple dozen strain gauges and wire pots, and then slowly drive a big truck over it. All data was acquired with wireless data modules, and then fed to a computer. Cool project.
One of my engineering professors said that in Russia they used to have all the designers stand under the bridge while it was loaded to capacity.
In Soviet Russia, bridge breaks YOU!
And did they repeat the tests with bigger trucks until the bridge fell?
I do not know the answer to the OP but here is a bus that is supposedly 2-3x the weight limit of a bridge it is crossing. I believe this is the Beaver Bridge (aka Little Golden Gate) in Arkansas. I’ll save you the suspense, the bridge holds, but it is scary to watch (turn down sound…someone is honking which is annoying):
Well of course. 
I did find a paper on the testing we did, but it requires access to unlock it. (Professor F*rhey got all the credit, but didn’t really do anything. The wireless data acquisition system was designed and built by me and another engineer.)
There are building codes that you follow that specify how to calculate the load and the strength of the structure. Both are based on physics/mechanics research and testing as well as some consideration as to the consequence of failure.
examples:
- ASCE 7 standard | ASCE
- Current Standards | American Institute of Steel Construction
- Steel Bridge Design Handbook | American Institute of Steel Construction
I’m licensed as a PE in Virginia, but am not currently practicing.
Key to working this out is consistency of materials and how they connect, so you can measure how they will transmit force - both the still and moving weight of the vehicles on bridge deck, but also the dead weight of the bridge. Bridge designers have always had a struggle between making rigid structures that transmit all force well, and having to retain a certain amount of movement to allow for daily natural expansion and contraction.
When they are made of multiple components, like lengths of timber, then it was common practice to massively over-engineer them. These older bridges may be designed for a nominal weight of say 15 tonne (big heavy laden wagon) but can carry considerably more. Unfortunately ‘considerably more’ might mean normal cars and buses but not heavy freight vehicles, and over-size loads increase risk of failure. The huge truck taking a shortcut over a 50 tonne bridge may make it across but make it a deathtrap for the 48 tonne school bus following behind.
When the Sydney Harbour Bridge was constructed they loaded almost a hundred locomotives on it as a weight test. This was partly engineering but also showmanship to assure the public that one of the most massive bridges in the world at the time could carry its own weight plus all possible traffic.
I recall when the Golden Gate Bridge had its 50th anniversary and so many people showed up and walked onto the bridge the natural arch the bridge has started to flatten. Apparently the engineers were not happy about that and, ever since, they do not allow people on the bridge like that.
You can just see it in this photo:
Since retirement, I’ve let my license lapse, and I was more into the other specialties of CE anyhow. That said, there are a couple of points about the Golden Gate Bridge event. People were standing much closer together than, say, cars and trucks and buses traveling along the bridge with normal vehicle spacing and not every vehicle carrying a full load. So sure, there was as much or more ‘load’ than there usually is, pretty evenly distributed for the entire length of the bridge.
The bit to cause public safety nightmares is the thought of what would have happened if there were any cause for panic by anyone on the bridge. Shudder!
I studied this when I was getting my degree in structural engineering. Basically you know how much force a steel girder, cable, weld, or other structural elements can theoretically withstand based on its material, shape, and cross section. You estimate the forces the pieces of the bridge are expected to experience, based on shape, expected traffic, weather, etc. Building codes require you multiply that by a factor of safety.
Ergo, you come up with a theoretical max load the bridge can safely accommodate.
Keep in mind that this will be much less than the actual max load, as the designers will also want to take into account stuff like unexpected forces, shoddy workmanship, poor maintenance, etc.
Every now and then you do get something like the Tacoma Narrows bridge that spectacularly fails in new and unexpected ways because the designers didn’t account for some random thing. In this case, an unusually thin road deck that happened to oscillate at the same frequency as a brisk wind that day.
Exactly. Bridges are designed based on known material and connection properties to a certain weight limit, and over time as materials age and fatigue those limits are reduced. Higher weights may not collapse it, but they’ll do more damage.
A wooden covered bridge near me had a very low weight limit, like 3-tons if I recall, and a large fully-loaded dump truck drove over it. This caused many of the members to crack and bolts/rods to pull out, and the bridge had to be completely rebuilt after being closed for many years.
Another much larger bridge, the John A. Roebling Suspension Bridge in downtown Cincinnati, also has a very low weight limit for a span over the Ohio River at just 11-tons. The bridge was opened in 1867, and in 1896 the deck was rebuilt as a truss with additional suspension cables so it could support streetcars. However, one critical design flaw is that the saddles at each tower on which the cables rest crushed under the weight almost immediately after construction. On a suspension bridge the cables need to slide over those saddles to compensate for uneven loading, but when they’re fused, a heavy weight in the middle of the bridge will pull the tops of the towers towards each other. This causes bending in the towers and voids can open up in the foundation. So strain gauges, crack monitors, recalculation of the loads and stresses, and computer modeling have been used to determine a safe weight. They even have ways to tell when one of the over 10,000 cables that makes up the main cables has snapped, and each one adds another data point to the calculations.
Now you’re just being silly - of course, they stopped one truck before collapse.
Dan
It seems crazy to install a railway line running onto the bridge just so you can test the loading with locomotives.
I hope they also had the people responsible for acquiring the materials and the workers who built it, stand there, too. When buildings fail, is often because some of the components are worse than what the engineers specified, or because the workers cut corners.
When it was built it was both a road and a railway bridge. Seems to still be used for urban transit rail.
Yes, sorry, I was just making a dumb joke.
Whoosh!!! Good one.
Although an awful lot of bridges built before ~1950 originally had rail lines that were later removed. I thought perhaps you knew there were no rail lines now and assumed that had always been the case.
In looking that up to confirm my fuzzy recollection I was surprised to find the Sydney bridge dates from 1932. I’d not recalled it was quite that old. I’d have made it immediately post WW-II if you’d asked me cold.
Once knowing its vintage I was not surprised to find it had at least light rail lines from the git-go. But I was then counter-surprised to find the rail lines are still there, not having been removed to make way for more cars as is common US practice.
So joke or no, I learned a lot from being whooshed. Thank you.
The Sydney Harbour Bridge originally had (from W to E) - walkway, 2 standard rail lines, 6 road lanes, 2 tram lines, walkway - which are now configured as - cycleway, 2 standard rail lines, 8 road lanes, pedestrian walkway, so the tram lanes were removed as part of their general removal throughout Sydney.
Next month is the centenary of the official turning of the first sod leading to its construction.