Step 1: Recall the chemistry behind thermit welding.
Thermit welding relies on an exothermic reaction between iron oxide and aluminium powder, which produces molten iron along with a large amount of heat, and this molten iron is what actually fills and fuses the joint.
Step 2: Separate the genuine features of the process from its shortcomings.
Getting heat from a chemical reaction, needing no separate flux because the aluminium oxide slag floats up and protects the weld on its own, these are simply how the process works and are actually conveniences, not drawbacks. A narrow, tightly focused energy beam causing minimal distortion is also not a valid description of thermit welding at all, since the reaction releases heat over a broad area rather than a concentrated beam, that description fits processes like laser or electron beam welding instead.
Step 3: Focus on where the process genuinely falls short.
Because the reaction is designed around producing molten steel, and it releases a large, hard to control burst of heat, the process only really suits joining thick, heavy sections of ferrous components such as railway rails or large shafts, it cannot be scaled down for thin sections or easily adapted to non ferrous metals. This restricted range of suitable applications is the real limitation of thermit welding.
\[ \boxed{\text{applicable only to ferrous metal parts of heavy sections}} \]