Belz suggested to someone that they build a miniature of the tower and test their theory. Somewhere else, someone suggested firing a bullet into the ground to simulate what happened to UA93 in the field in PA.
Anyway, would either of those things work? A miniature of the tower doesn't have the mass. Ditto with the bullet.
So, CAN you simulate something that had such a huge mass (building and plane) relative to the miniature with which you're simulating?
Not really, no. There are ways to build relevant models but it's nowhere near as simple as just scaling things down. Despite your admission of having a non-technical background, your intuition that it's not so easy is dead on -- that puts you leaps and bounds beyond the Loosers!
Material properties, fluid flow, and heating are alll things that really don't scale well. Let's say you were to build a perfect miniature of the WTC towers, all scaled down by a constant factor S. Then a steel column, for instance, would be S times shorter and S times narrower. If you made this mini-column out of the same structural steel, you have a problem. The
strength of the column will decrease by a factor of S
2, but the amount of gravitational potential in this column decreases by S
4 -- the mini-column is S
3 times lighter, being of the same density but S times smaller in each of three dimensions, and it's S times closer to the ground since the whole building is smaller.
So scaling the problem means you've created a mismatch between strain and gravitational energy. There is no reason to suspect the scaled model would behave the same as the original article. This should be obvious if you take it to an extreme, for example the milk-carton getting hit by a pencil igniting a thimble-full of kerosene argument I've seen in conspiracist discussions. Bizarrely, the conspiracists seem not to realize the absurdity is because it's a bad model, not that what really happened in WTC 1 and 2 is impossible.
Sometimes, if you're careful, you can compensate for these scaling errors by changing the model. If all we cared about was yield strength and gravity, we could compensate by changing materials. If we found a new material that had a different ratio of
strength to weight -- say, replaced all of the columns in our model with lead -- then we could scale it some and have consistent results. For example, have you ever heard of a
water tunnel? It's the same thing as a wind tunnel, but with water instead of air. Sometimes you can scale down an aircraft problem and get correct results by changing the operating fluid to water.
In our WTC example, though, there are so many things going on at once that trying to balance them all will be impossible. Our mini-tower made of lead will now behave incorrectly with respect to the initial airplane impact, and have improper heating characteristics. The air and fuel flows governing the fires will be wrong. Wind load on the building will be different, as will coupling to the bedrock underneath. And this assumes we've done a meticulous job recreating it. Something so crude as a solid-core bullet simulating Flight 93 is obviously not going to be close.
The only way to make decent models are to take pieces of the structures -- say recreate a single column of the WTC, and apply a load and heat -- or to use a computer model. These approaches can be accurate, if you're careful. Many such experiments have been done, notably in the NIST report. This is part of the reason why the scientific community is satisfied with the official explanation.