The Hindenburg on Mythbusters
Mythbusters had an episode recently where they tested the hypothesis (not really a "myth") advanced by a NASA scientist that the doping agents in the paint used on the canvas were more a culprit than the hydrogen. The canvas outer shell was doped with iron oxide and powdered aluminum, two constituents of rocket fuel. I'd seen a special on, I think, Nova a few years ago where they explored this hypothesis, and so I was anxious to see what the Mythbusters would find out.
Unfortunately, as I have often found with Mythbusters, I was disappointed. I watched a little when it was first recommended to me, but the methdological errors, while not frequent, always came up in the experiments I was most interested in, so I stopped watching for a long time. I recently started watching again, and things have not changed.
In this particular case, they built three scale models of the Hindenburg: two doped just like the original, and one doped with pure thermite (a combustive mixture of aluminum powder and iron oxide that represented the worst-case, if unrealistic, scenario). They burned one of the ones authentically doped without hydrogen and it took way, way longer to burn than the Hindenburg. Then, they burned the other authentically doped one with hydrogen, and it burned much faster, but still slower than the real Hindenburg (it was on the order of, as I recall, a minute and a half for the model but only 34 seconds for the actual one).
They more or less declared the theory busted at this point, but still, for fun, they burned the model doped with pure thermite with hydrogen. In theory, this should have been much worse than the actual Hindenburg, but, in fact, it still took over a minute for the model, a 1:20 scale, to burn. Then, they declared the theory busted.
But wait a minute! Why couldn't they get a 1:20 scale model to burn up even as fast as the Hindenburg, if not faster? Why did it take so much longer for a model doped in pure thermite, which should have gone up like a torch compared to the actual Hindenburg, to burn?
I don't know. But the fact that none of their tests was able to get a 1:20 scale model to burn even as fast as the full-size dirigible means that there was something different between the real thing and the models that hasn't been accounted for. I don't know what it is, and I don't know why the models didn't burn the way they were expected to. But I don't think the doping theory can really be declared busted when there is a significant discrepancy between the simulation and the actual event. Whatever the factor is that made the Hindenburg burn so much faster than the model, despite being 20 times bigger, could also be the factor that makes the doping theory work.
Or, it might not. But I didn't feel the episode successfully proved anything one way or the other. I know that it's a TV show, so they have limited time, a limited budget, and have to have a show to put on whether they get conclusive results or not. I don't know if they have the option for a judgment other than Busted, Plausible, or Confirmed, but in this case the real verdict should have been 'Inconclusive.' I know an inconclusive result probably doesn't make good TV, but I would have liked at least a nod towards the fact that they realize that there was a flaw in the experiment somewhere, because otherwise I kinda feel like they're trying to put one over on me.