Char made from narrow size fractions of lignite in a fluidized bed was briquetted at 140 MPa with various binder tars produced from the lignite. The raw briquettes were cured in air at 200 °C and then carbonized, usually at 925 °C. The tensile strength of the resulting formed coke could be as much as three times that of the best metallurgical coke, or as low as 2–3 MPa, depending on the type of binder used, or even a factor of ten lower than this if the curing stage was omitted. Curing is thus a crucial stage. Distinct optima of binder percentage and char particle size were found, giving maximum briquette strengths (≈ 13 MPa). A corresponding ‘strength surface’ suggested four factors associated with weakness of a briquette. Binders produced by air-blowing rather than nitrogen-blowing gave stronger briquettes. Fluidized bed tar was greatly ‘improved’ in this way, but Gray-King tar only slightly so, probably because the former contained more generally reactive species. Cured-only briquettes were usually surprisingly strong and carbonization increased this strength ≈ 1.5 times. Curing is considered to ‘fix’ the binder; carbonization changes the chemical nature of the ‘cement’ between the char surfaces but does not appreciably alter its spatial distribution. The carbonized briquette thus has substantially the same structural flaws as were present in the cured state and so has a strength closely correlated with that achieved merely by curing.