The full scale fire resistance test forms the basis of acceptance of structural integrity in fire. However, such tests are too expensive to provide more than a limited amount of test data of parametric significance as a means of fully exploring a width of structural responses, required for research purposes as underpinning to the formulation of design guidance. In this paper a methodology for the fire testing of reduced-scale structural models, previously outlined in other fire engineering journals, is presented for dissemination to the wider testing community. The principles of scale modeling applied to general structural testing and the specific requirements for scaled thermal modeling are detailed. Their application to the formulation of a new model fire curve for reduced-scale modeling is defined. An initial normalized temperature distribution condition in a test specimen requires the imposition of an inverse scale incident heat flux rule and a scale squared time scale rule. The increased flux requirement can be accommodated by an enhanced temperature axis formulated by consideration of convective and radiative surface heat transfer processes. An alternative suggestion to control model test furnaces by heat flux instrumentation is also postulated. Finally, experiences with the application of the methodology to some wide-ranging research studies on brickwork compartment walls, reinforced concrete floor slabs, and steel columns in standard fire tests are presented to illustrate different facets of the application of the methodology. Further applications to ''natural'' fire tests performed within the Cardington Frame program are also referenced.
