3 ms·
My point is that it isn't a good starting point across designs, because the designs vary so much as to make their formula inapplicable in the general case. Now
by chiph 14y ago
My point is that it isn't a good starting point across designs, because the designs vary so much as to make their formula inapplicable in the general case.
Now, if they said that their formula was consistent within a certain class of materials & construction, I'd buy it. But even then with the Plexiglas, their formula predicted 7 cracks, while the experiment produced 8. They need to explain that, and I'm saying it's because of the near-crystalline internal structure of Plexiglas. If they had altered their hypothesis & and prediction, and then retested, and did some analysis to explain it (perhaps I'm wrong - maybe its because of some other factor and not the internal structure) I'd buy it.
- btilly 14y agoIt is a scaling rule for a statistical phenomena, not an exact formula. For different kinds of surfaces, you'll have a different constant. Getting 7 cracks versus 8 in the experiment is actually a pretty good fit. Particularly since repeating the experiment multiple times would be unlikely to give the exact same number of cracks. (The cracking pattern does have a lot of randomness in it.) What is more important is that if the sqrt scaling rule proves reasonably consistent across a wide variety of types of materials, then testing the cracking pattern for different kinds of impacts across different kinds of windshields becomes a much more tractable problem. The obvious followup becomes how angle of impact matters. My hope would be that it is something reasonably simple - such as very little is deposited by friction, so it is only the component of the velocity normal to the surface that matters. Establish those two things reasonably well, and now you're in a good position to measure every kind of windshield a handful of times, and you have some useful information for forensics on all kinds of windshield cracks.