Existing food freezing time formulas have not been tested on low moisture, low freezing point foods or cryogenic temperatures. A systematic numerical experiment was therefore performed to generate freezing times over a wide range of parameters, never before covered in the literature, yet still within industrial practice. The results were used to evaluate four well-known freezing time formulas. Pham's first (three-stage) method has the best theoretical basis and fewest empirical parameters, and agrees best with numerical predictions. By applying simple correction factors, this method agrees with numerical predictions to within +/- 10% while Pham's second (two-stage) method is almost as accurate. For the freezing time of fresh foods (T-f >= -1.5 degrees C), Pham's first method can be used without modification. The corrected formulas remain accurate under cryogenic freezing conditions. (C) 2013 Elsevier Ltd. All rights reserved.