If they really wanted to test what happened, they would've had to put the test article and the foam cannon in a wind tunnel. The wind speed and atmospheric conditions would've had to replicate what the Shuttle was experiencing approximately 82 seconds into the launch. The foam would've had to be cooled to the best estimated temperature, too. NASA didn't do any of that.

Such a test would've generated more realistic results, but that may not be necessary. There were some skeptics who didn't believe a foam stike would seriously damage the wing. They know better now. While this test may not provide the definitive results that a wind tunnel test would have, it was sufficient to show that a large foam strike was the most likely cause of the Columbia accident.

As for the candle experiment, I'm not familiar with it. I have seen firsthand what even a small object moving at high velocity can do. Back in 1974, a bunch of tornadoes hit my home town. A good friend's home was heavily damaged (he and his family were inside at the time). I found a small piece of asphalt shingle no bigger than a quarter that had punched all the way through their wall and into the home. It had to go through the wood siding, the insulation, and the drywall. Believe me, it was the impact of the piece hitting their home and not the air carrying it that made that particular hole. I've also seen photos of pieces of straw imbedded into telephone poles by tornadoes. Remember the equation for kinetic energy (e = 1/2mv^2). Even a small mass, if accelerated to high velocity, can have a lot of kinetic energy.