Thermo-analytical methods are commonly used for phase transformation temperature identification in a lot of science and industrial applications. Currently, there are most frequently methods based on very small samples analyses like a differential scanning calorimetry (DSC) and/or a differential thermal analysis (DTA) realised for various materials mostly in combination with a thermo-gravimetry (DSC/TG; DTA/TG respectively). The aim of such methods of tiny samples is to fit a transformation process without an influence of the sample mass and to get universal like conclusions. On the other hand, older direct thermal analytical method generally done on tens grams and larger samples are not so popular now excluding few applications regarding to cast iron and/or non-ferrous metals and their modifying process monitoring. However, simultaneous utilisation of both mentioned methods during determination of real steel liquidus (T-L) and solidus (T-s) temperatures leads to suppressing their individual disadvantages and enable to fit the results for such heterogeneous material like steel in the best possible way. The paper is focused on the discussion of T-L and T-s of 18 continuously cast steels analysed in the frame of No. LO1203 "Regional Materials Science and Technology Centre - Feasibility Program". The submitted evaluation shows an importance of the parallel utilization of methods, accuracy and reproducibility of the results and also the deviations of the experimentally determined T-L and T-s and empirically calculated values of the industrial partner and values obtained by thermodynamic calculations.