Abstract: Abstract: Water repellency is a widespread phenomenon in soils. Almost every soil shows water repellency to some degree. It may reduce the infiltration in soil and is easy to form finger flow. Therefore, the soil holding capacity of moisture and the yield of crops are reduced. It's very important to study the soil water characteristic curve in repellency soils. In this study, we aimed to explore the impact of repellent levels on soil hydraulic characteristics, clay loam of 7 repellent levels (treatment L0 to L6) and sand of 6 repellent levels (treatment S0 to S5) made by adding octadecyl primary amine collected from Yangling, Shaanxi. The experiment was conducted in the Key Laboratory of Agricultural Soil and Water Engineering, Ministry of Education, at Northwest A&F University in August 2016. Each of the 13 treatments had 4 replicates. Soils were mixed with octadecyl primary amine which varied from 0 as control and 0.2, 0.3, 0.4, 0.5, 0.6, 0.8 g/kg respectively for clay loam and 0.06, 0.08, 0.09, 0.115, 0.12 g/kg respectively for sand. The repellency level of treatment L0 and S0 was wettable, L1, L2 and S1 was slight repellent, L3, L4 and S2 were moderate, L5 and S3 were severe, and L6, S4 and S5 were extreme. The soil moisture characteristic curves were measured with a high speed centrifuge. The van Genuchten (VG) model was fitted to obtain the hydraulic parameters with RETC software. The results showed that under the same suction, the soil moisture of repellent clay loam was higher than that of repellent sand and the soil moisture decreased with the increase of repellent levels. The residual moisture and the shape parameter decreased and the reciprocal of air entry value increased rapidly with increasing the repellent levels for repellent clay loam. There was no difference in residual moisture of treatment of S0 to S3 for sand, but the residual moisture of treatment S4 and S5 was lower than the other treatments for repellent sand. The reciprocal of air entry value increased with increasing the repellent levels for repellent sand. The air entry value decreased with increasing the repellent levels for both repellent clay loam and sand. There was a significant negative linear correlation between logarithm of air entry value and the amount of octadecyl primary amine, and the determination coefficient was 0.92 and 0.93 for repellent clay loam and sand, respectively. There was no difference in wilting coefficient between the treatments for repellent clay loam and sand respectively. The field capacity, effective water and easily available water decreased with the increase of the repellent levels for both repellent clay loam and sand. Meanwhile, the gravity water showed a decreasing trend. The results also showed that there was no difference in the proportion of micropore and macropore while the micro-porosity and porosity decreased with the increase of water repellency, with increasing the repellent levels for repellent clay loam. The soil voids increased sharply with the increase of repellent level for repellent clay loam. The micropore did not exist in repellent sand. There was no difference in the proportion of micropores, macropores and soil voids with the repellent level for repellent sand. The proportion of small pore of S0 was significantly higher than the other treatments but there's no difference between S1 to S5. The proportion of medium porosity increased obviously with the increase of repellent levels for sand. Therefore with the repellency increase, field capacity, easily available water and micropore of sand decreased, and soil porosity of clay loam increased. It led the water that crop can absorb reduced. Repellency was unconducive to crops growth. We should try to avoid the increase in soil water repellency level. This study can provide valuable information for the infiltration, evaporation and numerical simulation of repellent soil.