A phase diagram optimization has to start with a minimum set of experimental information. Among the different experimental methods used to establish this “minimum set”, some are well known, such as differential thermal analysis, X-ray diffraction at different temperatures, electron microprobe analysis... Some other are not strictly conventional, such as direct reaction calorimetry, or self propagating high temperature synthesis. Even if these techniques are not specifically devoted to phase diagram determination, they can give very interesting details about phase diagrams. Some examples will be given, advantages and drawbacks will be discussed.