The use of isoconversional, sometimes called model-free, kinetic analysis methods have recently gained favor in the thermal analysis community. Although these methods are very useful and instructive, the conclusion by some that model fitting is a poor approach is largely due to improper use of model fitting, such as fitting a single heating rate or multiple heating rates separately. The current paper shows the ability of model fitting to correlate reaction data over very wide time-temperature regimes for three polymers of interest for formulating high explosives: Estane 5703 P (poly[ester urethane] block copolymer), Viton A (vinylidene-hexafluoropropene copolymer), and Kel-F 800 (vinylidene-chlorotrifluorethene copolymer). The Kel-F required two parallel reactions-one describing an early decomposition process accounting for similar to1% weight loss and a second autocatalytic reaction describing the remainder of pyrolysis. Essentially no residue was obtained. Viton A and Estane also required two parallel reactions for primary pyrolysis. For Viton A, the first reaction is also a minor, early process, but for Estate, it accounts for 42% of the mass loss. In addition, these two polymers yield 2-3% of residue, and the amount depends on the heating rate. This is an example of a competitive reaction between volatilization and char formation, which violates the basic tenet of the isoconversional approach and is an example of why it has limitations. Although more complicated models have been used in the literature for this type of process, we model our data well with a simple addition to the standard model in which the char yield is a function of the logarithm of the heating rate. (C) 2004 Elsevier B.V. All rights reserved.