I was going to get to that, if our newcomer was interested.
From other large, non-mining style explosions, such as the
Port Chicago disaster, we find that about 1/1000th of the energy of explosion shows up as seismic activity. This is rough and probably varies by a factor of 10 at least, but let's go with it. That's also why the 1993 bomb was not detected. Too small.
We also know the magnitude of seismicity from the collapses, and the total energy of each from calculations of the Tower mass and height. These reveal that in the Tower collapses, about 2% and 5% of the energy, respectively, shows up as ground motion. This larger coupling should be intuitive, given that it's caused by things actually hitting the ground, not exploded away from the ground.
So we conclude that explosives used for demolition are roughly 20 to 50 times less efficient at creating ground shaking. Hence, to get a comparable signal on a seismograph, we'd need 20 to 50 times as much of them.
The Tower collapses each were about 100 tons TNT equivalent. This means to get the same shaking, i.e. to cause the seismographic traces that our newcomer insists were explosives and not from the collapses, our "Demolition Explosives" would be about 2 to 5
kilotons.
That's not a demolition explosive. That's a tac nuke.
Didn't happen.