I guess you would get a small cloud of radioactive dust without fission yield. Generally you need to create sufficient neutron flow of neutrons of very specifi

I guess you would get a small cloud of radioactive dust without fission yield. Generally you need to create sufficient neutron flow of neutrons of very specific energy to start fission chain reaction. This happens by bringing more of the fissile material together - more neutrons cause more fissile reactions, leading to more neutrons-boom. An explosive fission will push matter away. If you create an explosion that parts the fissile material before most of the nuclei have split, it’s a dud. To get a really high yield you have to shape the geometry of the fissile material very precisely at a very specific moment so that most of the nuclei get flow of neutrons at exactly correct energy. It does not suffice to bring heat to nuclear material, or just lump it randomly together, you need to be very precise so that there is enough of the fissile at the right place at the right time. The non-nuclear detonators explosives in nukes serve two purposes: push the fissile material together fast enough, and shape the fissile material just-so due to the pressure effect. It’s really, really inmprobable to trigger the nuke by a random explosion other than the trigger one.