>> I take it there's more to it then just asking the computer "hey, will 
>> this break if I put 25 N through it?"
> 
> Not much more

Mmm, OK.

> just specify which parts are fixed (eg mounting brackets) 
> and then tell it which surfaces to apply a force to.  It will then run 
> some sort of iteration algorithm until it has "solved" the structure so 
> that all internal forces are perfectly balanced.  Then you can just ask 
> it for the maximum stress in the structure and see how big it is 
> compared with the yield stress of the material...

I guess like you say, working out whether a component will snap under a 
given load is much easier than determining (say) whether a car engine 
will explode when you ignite petrol vapour inside it.

>> At what point does it become cheaper to just make the thing rather 
>> than simulate it?
> 
> Hmm, the first samples are always quite expensive to make because you 
> are making so few of them, but usually it's more a time thing than 
> money.

Ah, but time *is* money, because...

> of course the customer usually has a deadline for releasing the product

Even so, I'm kinda surprised at the lead-time for a prototype. Mind you, 
I guess if you just want to build a model of something, that's quite 
easy. If you want to build it out of the real materials for strength 
test purposes, that's a tad slower...

> If you know 
> something is critical to get right and you haven't done it much before, 
> then you simulate it.

OK.

> Similar to how your steering wheel is connected to the wheels by a 
> telescopic tube, so in the event of an accident the telescopic bits 
> collapse rather than the rod coming through the front of the steering 
> wheel and into your chest!  Although Jeremy Clarkson would say that 
> would make people drive safer...

LOL! Well, Clarkson is a moron. ;-)

> Well yes, and then you get things inbetween, like the wrist strap for 
> the Wii remote control, where for some reason they didn't realise it 
> would snap if you pulled it hard.  I mean WTF how can you miss that 
> one?  I would have thought they would have got some really strong guy to 
> throw it around while measuring the force, and then made sure it could 
> withstand double or something.  Maybe Japanese "strong guys" are not 
> that strong :-)

Hehehe. Apparently the world record for the loudest human shout is held 
my somebody from Japan or China. They're pretty crazy people, really. 
Not sure how strong though...

>> One thing I always wondered... You see people doing all these crash 
>> tests, right? So... where the hell do you get an endless supply of red 
>> cars to crash into things?!
> 
> A car factory?  They make lots of prototypes too for many different 
> purposes, btw I don't think they have to crash very many to pass the 
> official tests, maybe even only 1 car, IDK.

"And in this test, we're going to crash a new car into an 20-year old". 
Uh, and where did you get that from, exactly?

Also, why do they already have to be painted random colours?

Post a reply to this message