There is a large discrepancy in the open literature about the comparative performance of the existing macro and microscale heat transfer models and correlations when applied to small/micro flow boiling systems. This paper presents a detailed comparison of the flow boiling heat transfer coefficient for R134a in stainless steel micro tubes with 21 macro and microscale correlations and models. The experimental database that was used in the comparison includes the data for 1.1 and 0.52 mm diameter tubes, mass flux range of 100-500 kg/m2 s and system pressure range 6-10 bar obtained in the course of this study. The effect of the evaporator heated length on the comparative performance of the correlations and models was investigated using three different lengths of the 1.1 mm diameter tube (L = 150, 300 and 450 mm). This comparative study demonstrated that none of the assessed models and correlations could predict the experimental data with a reasonable accuracy. Also, the predictability of most correlations becomes worse as the heated length increases. This may contribute in explaining the discrepancy in the comparative performance of the correlations from one study to another. A new correlation is proposed in the present study based on. the superposition model of Chen. The database used in developing the correlation consists of 5152 data points including the current experimental data and data obtained previously with the same test rig, fluid and methodology for tubes of diameter 4.26, 2.88, 2.01 mm. The new correlation predicted 92% of the data within the 30% error bands with a MAE value of 14.3%. (C) 2013 The Authors. Published by Elsevier Ltd. All rights reserved.