Why do the Strongest Parts Look Like This?

@Paul Jones the all fan

F

Oh come on ok nobody said to make a series of lectures for an engineering degree but you surely could enrich this video with more details and trivia and make it a 20 to 30 minute one... in 9 minutes I didnt even finish my cereal ! :P

Just kidding. Great video!

*topological

rofl mao

Weird, I keep the study bibles and concordance on the bottom.

why no love for diff. eqs?: The power of rates of change at your fingertips!!! I think Thanos' cavity search handwear had a related pebble. /s

Hahaha that’s a heck in good joke

@Mike Clarke this is accounted for in structural margins

One problem today seen in more modern structural systems, is before would over engineer as could only be within a certain tolerance using primitive calculation methods and knowledge. AS we know more can can stretch things to more limits, we end up stretching things to "The" Limit and sometimes we need more tolerance to absorb the extra stresses times when things may go beyond limits, aircraft, rockets, etc when the unexpected is encountered. Sometimes less is not good.

@carlos vargas Not a big cost for the aerospace industry.

@Mattsoup and if we'd start printing on the microscopic/quantum level? I know I'm thinking in the future and metal tend to become uniform in terms of their properties of existing when pressure (heat or force) Like forging is like smashing stuff together until the protons find their equilibrium (with metals at least) and just but what if we place the puzzle pieces where they need to be and after you heat treat it to solidify to further strengthen it. Or is brute force just better?

@ba doem additively manufactured parts still need machined at the interfaces. Huge printers for making large parts are extremely expensive. Forged/rolled metal stock is stronger than additively manufactured material, even when heat treated.

A cheetah is optimized for running, but want it to swim and it will die. A fish is optimized for swimming, but it can't run. A beaver can do both but neither that well.

"Given unlimited materials and budget, even a two year old can build a bridge across the Atlantic that could carry dinosaurs. Anyone can make a bridge or a building that stands. But it takes a structural engineer to make one that _barely_ stands." Before people knew how to engineer structures, the best solution they had was "overbuild the hell out of it." They'd guesstimate how much "building" or "bridge" they need to do the job, then fucktuple it. That's why older buildings are so "heavy" compared to modern ones. When modern framed houses first came into use, traditional builders laughed at the idea of such a "lightweight" building, and joked that they'd blow away in a strong enough wind.

I think people don't get that about engineering. Engineering is usually about finding the weakest possible design that is still sufficiently strong for the use while including after accounting for an appropriate margin for safety. A lot of the largest and most impressive feats of engineering wouldn't have been possible if they had to aim any higher with the technology and methods that were available at the time.

@Stef Hirsch nice. By any chance do you have photos?

@Saying N*igro Makes TRvid cry it was also made with care and quality workmanship, the dropouts, linkage and such are machined with lots of detail… thus light and strong enough to withstand 21 years of use while being reasonably light. It also uses 26” wheels which allow the combination of the fork and front wheel to be shorter - which is effectively a lever. The chain/seat stays are also allow a cross brace which is also inherently more structurally more efficient. It has a 120mm fox vanilla rl on the bike.

edinBERG 😂

@Ben Sinor no simulation will ever be able to totally capture all of the variables of the real world. The one that they "thought would be the worst", they didn't expect it to be the worst because of the simulation results (which should be obvious, since it was among the 10 best of thousands), but rather because the one that performed well was so different from what they were used to seeing in the past, which were all developed from less-sophisticated means of simulation.

I hope they then changed their simulation parameters and retested the thousands.... if your real world results don't confirm the simulations then you did it wrong. There could very well be a design filtered out initially that is even better.

Could it be a super corona ring? I’ve done powder my whole life and that seems like a good description of it. Look it up and tell us if that’s it cus I’m also curious now lol

I'd like to see it too

@Nick Leland I rarely have time to watch TRvid videos, much less to write comments. But after watching this video and looking through a few hundred comments, I was gravely disappointed by the amount of abject stupidity found in so many of them. So my response to you wasn't personal, nor did I intend for it to be rude. It just reflected all of the idiotic comments that I read, aggregated. Case in point, there is an absolutely moronic comment from someone claiming that optimized parts "aren't particularly strong" because material is removed to the point where they become weaker for "unexpected loads". Even worse, that comment has almost 400 likes! That is a patently absurd statement that no real engineer would make. That's a comment from (what's called today) a "maker". Someone that has access to 3D modeling tools, including optimization software, but doesn't understand the fundamentals of engineering. And that is dangerous. Computer software is a tool, not a replacement for knowledge and experience. If you don't understand the underlying fundamentals of what you're doing, then you're going to get specious results at best. Actual engineers consider ALL load cases while designing a component and / or assembly of components. Optimized components / assemblies are literally optimized for ALL of those cases. That's the point! And of course, for most engineers (those who work for someone else), there are numerous checks along the way in case they miss something (we're all human). They don't operate in a vacuum, they have a team of other engineers and dedicated analysts for stress / FEM / FEA, CFD, Multi-Physics, etc. Further, real engineers, especially those who are working on their own (inventing, consulting, etc), never rely solely on hand calculations and / or computer based analysis. They physically test prototypes and / or test articles to corroborate their various analyses. As a mechanical and structural engineer with over 30 years of experience, it's frustrating and sad to see so many comments on here that are completely divorced from reality. I suspect most of them are not made by actual engineers (people who earned a BSc in a field of engineering with a GPA of at least 3.0+), but some might be, and that is scary. It also reinforces that "makers" i.e. - fabricators, welders, machinists, computer operators / CAD technicians, etc. are NOT engineers!

@GoogleEqualsEvilFair enough but no reason to be rude about it. Rewatched the video and don’t know how I missed it 10 months ago. I work mainly with CFD components but I was working on a project regarding FEA and wanted to do some extra research. Appreciate the late reply.

@Nick Leland 1) The fact that you're asking that question makes me seriously doubt you'll be able to understand Michell's work. 2) It's a singular paper, not papers. 3) Watch the video again...if you still haven't figured it out, It's at 5:28 where he talks about A.G.M. Michell's paper that was published in 1904 in the "London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science". 4) Michell's paper provided the foundation for modern topology based optimization techniques.

Just curious, what 100+ year papers are you referring to? Would love to take a look at them for more information

Your boss lets you watch TRvid?! Now that's a new idea! :)

@First Last That would be so cool if motorcycle frames would have some insane topology optimization. Even less weight and it would look awesome. But it better be strong :D

@Steve Lowe I suppose if you want the frame to still be made of steel tubes just optimised and not going the crazy carbon fiber route like discussed in this thread, you could use the topology optimisation by determining the cross sectional area of the struts generated and with a good understanding of the forces on that beam and how that area would translate into the desired cross-sectional shape (tube), you might be able to optimise the individual struts of the frame when it comes to diameter and wall thickness. also the topology optimisation would show where to place the tubes. of course this would then require validation by means of of other calculation methods as things like buckling are very different for different cross sections This is just the idea that pops into my head. I might be wrong here and am open to discussion as I'd like to learn.

@ttttt Only most commercial aeroplanes...

@ttttt because it easier for mass production But still good regidity to weight Having front and rear subframe structures as cf makes it very difficult to fix also in case of a crash or any accident and much more expensive even when could be fixed This is as close as they can do it production wise... maybe in the future

@First Last yes but car bodies aren't true monocoque structures because the pillars and subframe rails are all box beams and take the majority of the loads. The rest of the skin just fills in the gaps and has very little strength or rigidity.

Fun fact. Königsberg means as much as Kings mountain .

@Wesley Kagan By the way, Edinburgh is pronounced 'Ed-in-bruh'. Really interesting video - thanks!

Very true. I felt it was good to keep Konigsberg for simplicity, although I don't do well with pronunciation.

pause

I appreciate it, thanks!

Thanks! I agree, as it becomes more viable for local solutions, i.e cloud based or not needing a week to run a study, I feel it will become a basic design step.

It's when I try to think of the countless hours of programming and testing that went into it that my brain really fries.

It's pretty impressive.

Thanks!

@AAA AAA Stimmt, wenn du andere Sprache lernst. Dann wird es leichter neue dinge zu erinneren.

@XDSDDLord It's a German name using German spelling.

@ElvianEmpire It's an eternal struggle to figure out which names and words need to be kept original and which ones can be adopted with the local language's pronunciation. And sometimes, the spelling is just changed so that the two are closer together than they would seem.

@ElvianEmpire as a non-native english speaker, I need to mention that english does not care about proper pronounciation either. We just get used to this mess

Vaguely threatening lol

Thanks! I don't think they would hire me, though hah

Thanks for watching! Sounds like a cool project. There’s so much to these systems that it’s hard to get a handle on all of them

Upgrade your grey matter, because someday it may matter.

Haha, it's a fantastic album! One of my favorites.

Yes- exactly. It wasn't a pleasant thing to read. But, it worked.

@BambiTrout too generic. I want my daughters to know that I named them after a male mathematician with a gammy eye and some of the most incredible insights the world has ever known.

His full name was Leonhard Euler, so you could have got away with Leona!

Didn’t want to make that her middle name? First name olive

@Ellis Jackson I have accepted the compromise that we will name our next dog Euler.

She didn't like Eulerina?

@Wesley Kagan Your welcome from Saudiarabia

Thank you! I appreciate it!

Noted! Thanks for watching!

High school doesn't have any "complex math". Not every college has it.

All of it is designed with a car project in mind, just sometimes it doesn't quite get all the way there.

Many AI actually made one, look up AI designed Wheel and if you find the orange one, that's the one.

I suspect the result would be a disc with no holes (no spokes, a zero spoke wheel), with a thickness gradient from the hub to the tire. This would also be very aerodynamic, at the cost of insufficient brake cooling. In any case, I would be very interested in a stress analysis of both this design and the multi-spoke designs mentioned by others. My idea would have eliminated stress raisers, possibly at the cost of more material.

@Tiit Saul nice try soldier. Try again.

@Dennis auto correction. Meant to write boldie

@Tiit Saul bolder wheels? What are you talking about? Im talking about the bolide's 3d printed calipers e.g.

Thank you!

@Wesley Kagan Superb! Leaving smarter than I came in. I guess Nature has been doing this sort of thing for aeons. Makes sense to sort of crib her blueprints for the some of the starting points. Hope the move goes well. Bigger castle, bigger booms.

Essentially- yes! Generative design uses similar structure to biological design in formation like plant structure

I completely agree! The conversation in the comments is one of my favorite parts of having a channel.

Glad you enjoyed it!

Yeah, Altair is doing some VERY cool stuff right now- I mean, it's fantastic stuff but also I have a feeling that I couldn't afford to walk in the door of their building, let alone explore their suite of design and AI programs... Although if they are listening want to send it over, I promise I won't be responsible with it.

It’s a great book!

That's good to hear- I worry I error on the side of too complicated sometimes, I'm glad that it can be understood!

Love mine! This is 928#4 for me haha