this post was submitted on 27 Nov 2023
908 points (100.0% liked)
196
16503 readers
2156 users here now
Be sure to follow the rule before you head out.
Rule: You must post before you leave.
founded 1 year ago
MODERATORS
My wife and I agreed to this years ago before we had children to always explain to our kids in detail. I explain engineering and technology, she explains medical science and history.
I thought we would raise super smart kids but they'd just ask their question, go "oh neat thanks" and then play video games again.
Haha yeah my son is tired of my shit. You can see the instant regret sometimes as I "over explain" how something works. I'm trying to find a good balance so he doesn't stop asking questions.
A trick I found with my little cousin was to pretend to think about it for a minute. I'd say "ummm .... " and furrow my brow and be quiet for a few seconds. It creates this little moment of suspense which makes the answer seem more desirable to her.
This is good. Thank you!
I used to do this to my nieces and nephews but get scolded for "making fun of the children." Like it's my fault that they are universally idiots until somewhere around the third semester of grad school.
How to determine the elastic tensor without any FEA tools?
Experiments and theory.
Build a bridge and see if it collapses!
...and then rebuild it.
I could swear there was a comic strip about this...
That's a valid approach, yet today, experiments are usually on a smaller scale.
Actually, a bridge like this is generally considered simple enough not to require FEA. Calculating it theoretically using trusses should be enough to make it. I'm sure FEA is still done but only because it's easy and cheap.
Magic & Voodoo
No, that's the FEA part.
If you have an actual engineering degree then you've done basic forms of these calculations by hand at some point, and have likely written a basic numerical solver in Matlab.
I used to play bridge builder too! If this were Facebook that would make me an expert. Sadly.
I can say I have a degree in everything and an expert of all matter but too bad this isn't reddit
You joke but there are legit a bunch of YouTube "engineers" who got their degrees from Kerbal Space Program and Polybridge as far as I can tell.
Look, if you took some classes in 1995 and then never practiced in that area and now make YouTube videos for a living, you are not an engineer. You are a hobbyist.
I used to watch the Real Civil Engineer channel a lot and he said he was mostly working on fluid drainage, and that he worked on the project to build a (planned) tunnel for the road that goes near Stonehenge.
I haven't heard any other youtubers say that they are engineers.
Fun fact about bridge load limits: school buses are legally allowed to ignore them. While this might seem insane given that the point of a school bus is to transport children safely, the posted load limit on a bridge isn't the weight which will cause the bridge to collapse - it's a weight which, if traffic heavier than that were to regularly use the bridge, would cause abnormally high maintenance and repair issues over the long term. Bridges can bear much greater loads than the posted limits without instantly collapsing.
Source: school bus driver who got a question about this on the test for his CDL endorsement and looked it up after going WTF?
what i've learned is that any piece of architecture should have like at least a 10x margin of error, you want it to tolerate astronomical degrees of fucking up.
You've also got dynamic loading to comsider, eg if a truck full of sand is driving across bridge and slams on the brakes, thats a nuch different impulse than just driving across normally.
I'd go for a food grade liquid tanker in this example. Muuuch more weight transfer as the liquid sloshes around, and food grade means no baffles to slow it at all.
Edit: spelling
Yep, I typically throw more steel at structures than required. If it 'looks' strong enough it is probably twice as strong as it needs to be. There are certain things like walkways that can meet design codes, but it would be bouncy and unnerving to walk on. Beefing up the structure also mitigates the 200lb ape factor i.e. more likely to stay standing if someone drives a forklift into it.
Any old schmuck can make a structure with a 10x safety factor. The thing is making the safety factor as low as reasonably possible to minimize costs. If there's a regulation that says 3x minimum, you're probably aiming for 3x. Which is why those regulations are important, I guess.
Source: I write code for a living, don't listen to anything I say
https://safetyculture.com/topics/factor-of-safety/#typical-overall-factors-of-safety-1
Well yes, but also whatever your maximum is, divide by 10.
Nah, we don't need any of your pansy "safety factors".
that already happened by assuming π = 10
Divide by 10. Pi = .314
There's a nugget of truth in the dad's explanation. Empirical data, some which are critical failures, are used to predict the limits of infrastructure. They don't destroy a copy of every single bridge because it's expensive and not even that useful for getting new data, but things in general get built and destroyed to test new materials or designs.
Careful analysis of the stress tensor of a simple geometric model with parameters given by empirical testing of simple materials can give you a reasonable ballpark estimate. I wouldn't be the first one to drive on a ballpark estimate though.
Haha, I understood this meme. Haha, hilarious.
