The eddy covariance (EC) technique is generally regarded as a standard method for crop evapotranspiration measurements. However, the imbalance of energy closure prevails in the EC observations. Evaluating the effect of energy balance closure on EC measurement is critical for improving the accuracy of this method. In this paper, a weighting method (the Lysimeter method) was used as reference to evaluate the effect of energy balance ratio (EBR) on EC evapotranspiration. The results revealed that daytime EBR varied seasonally in the field, where wheat and maize were rotated in the winter and summer, respectively. The EBR was higher in autumn and winter but lower in the spring and summer. For the wheat field, mean daytime EBR varied from 0.26 to 2.84 with an average of 1.15. As for the maize field, EBR varied from 0.19 to 2.59 with a mean value of 0.78. The evapotranspiration (ET) using EC (ET_ec) was clearly lower than derived by Lysimeter (ET_L). The mean ratio of ET_ec to ET_L (ET_ec/ET_L) was 0.61 and 0.50 during the wheat growing season and the maize season, respectively. The ET observed with these two approaches significantly correlated with each other (P < 0.01), with their characteristics of seasonal variation performing in a similar manner. The ET_ec/ET_L was found to be proportional to EBR (P < 0.01) in both the winter wheat field and the summer maize field. Furthermore, the effect of the leaf area index (LAI) on the relationship between ET ratios and EBR was significant in the even crop field during the entire growing season of winter wheat and the maize growing stage when the LAI was higher than 1. However, the effect was insignificant in the uneven maize field when the LAI was less than 1. On the other hand, friction velocity (u^*) exerted a strong impact on EBR and its relationship with ET ratios. EBR was observed to be proportional to u^* in both the winter wheat field and the summer maize field. ET ratios were proportional to EBR when u^* was small. Nonetheless, the correlation between these two parameters showed recession while u^* increasing, especially in summer maize field, ET_ec/ET_L was not significantly correlated with EBR when u^* was bigger than 0.3 m/s. Due to the weak turbulent intensity and the large eddy under the condition of low friction velocity, the EC sensor could not catch enough turbulent energy in a limited observation period (30 minutes), which caused lower measured turbulent fluxes and imbalanced energy closure. The results offer a possible way to correct the underestimated ET_ec using EBR.