Modified cable bolts are commonly used in underground mines due to their superior performance in preventing bed separation when compared with plain strands. To better test the axial performance of a wide range of cable bolts,a new laboratory short encapsulation pull test(LSEPT) facility was developed. The facility simulates the interaction between cable bolts and surrounding rock mass,using artificial rock cylinders with a diameter of 300 mm in which the cable bolt is grouted. Furthermore,the joint where the load is applied is left unconstrained to allow shear slippage at the cable/grout or grout/rock interface.Based on this apparatus,a series of pull tests were undertaken using the MW9 modified bulb cable bolt.Various parameters including embedment length,test material strength and borehole size were evaluated. It was found that within a limited range of 360 mm,there is a linear relationship between the maximum bearing capacity of the cable bolt and embedment length. Beyond 360 mm,the peak capacity continues to rise but with a much lower slope. When the MW9 cable bolt was grouted in a weak test material,failure always took place along the grout/rock interface. Interestingly,increasing the borehole diameter from 42 to 52 m in weak test material altered the failure mode from grout/rock interface to cable/grout interface and improved the performance in terms of both peak and residual capacity. Modified cable bolts are commonly used in underground mines due to their superior performance in preventing bed separation when compared with plain strands. To better test the axial performance of a wide range of cable bolts, a new laboratory short encapsulation pull test (LSEPT) facility was developed. The facility simulates the interaction between cable bolts and surrounding rock mass, with artificial rock cylinders with a diameter of 300 mm in which the cable bolt is grouted. Furthermore, the joint where the load is applied is left unconstrained to allow shear slippage at the cable / grout or grout / rock interface.Based on this apparatus, a series of pull tests were undertaken using the MW9 modified bulb cable bolt. Verious parameters including embedment length, test material strength and borehole size were all. It was found that within a limited range of 360 mm, there is a linear relationship between the maximum bearing capacity of the cable bolt and embedment length. Beyond 360 mm, the When the MW9 cable bolt was grouted in a weak test material, failure always took place along the grout / rock interface. Interestingly, increasing the borehole diameter from 42 to 52 m in weak test material altered the failure mode from grout / rock interface to cable / grout interface and improved the performance in terms of both peak and residual capacity.