I try to give a slightly simplified account. It should be understandable to an engineer but the technicalities might be difficult for, say, a lawyer.
One or more steam explosions are believed to have occurred during the Chernobyl nuclear accident. In this context a steam explosion can occur when hot molten material, in this case overheated nuclear fuel at ca. 3000 K, comes into near-contact with water. The rapid transfer of heat from the melt to the water produces steam and a local pressure increase. Sometimes there will be a steam gap which partially insulates the water from the melt. However, if the steam gap is distubed the steam production may become so rapid that the surrounding water cannot "get out of the way". Then the pressure increase escalates into a large spike, initiating a shock wave. The shock travels through the region of mixed water and melt and as it does so it disturbs the steam gap, again forcing the melt and water into close contact and perpetuating the rapid steam production.
The situation is somewhat analogous to the detonation of an explosive mixture, but with sensible heat (i.e. high melt temperature) as the energy source instead of chemical energy. When that is delivered to the fluid, in this case the remaining unboiled water, it is converted into mechanical energy. In the case of Chernobyl it destroyed the entire power plant.
The steam explosion was not the initiating fault. The reason why the fuel became so overheated was due to a reactivity insertion when the reactor was being operated in an unstable state. Once the reactor power started to increase it increased more and more, maybe several hundred times in less than a second. A bitter irony about it all is that the operators were running an experiment to try and learn how to improve safety. Basically, they messed up big time.
Melt-water steam explosions do not have to be in nuclear accidents. They can occur naturally when volcanic lava flows into the sea, but without causing much damage.
Finally, I'd better mention that the interactions that determine the magnitude of the explosion, even whether there is an explosion at all, are very complicated and are not completely understood. So far studies show that an explosion is more likely if the space is constricted, if the melt iswell above its melting point temperature, and if the water is well below its boiling point. If all three, then it's fairly likely.