Abstract: A large number of residual plastic films in farmland soil has posed a great threat to soil quality and crop production in recent years. A residual-film packing device can be widely utilized to collect the mulching film fragments in the farmland. However, an improved packaging quality has been highly required for the packing density and rate of residual film, due to the limited packing room space, complex structure, and field environment. In this study, a coreless-roll packing device was designed to test the packing performance of the residual film in the field. The front and back packing belt was arranged in a V shape to replace the roller for the relatively large packing room. The modified device was effectively reduced the wrapping film on the roller, particularly for the much simpler structure than before. A theoretical model was established for the maximum residual film bale diameter (d), which was generated by the packaging device and the length of each row (L). A field test was then carried out to verify and optimize the model and the improved device. The reasonable packing space was fully meet the requirement of d>0.5 m. The forming process of residual film was then evaluated in the packing room. Two stages were then divided into the whole forming period of the residual film. One stage was the small deformation on both sides of the packing belt that was caused by residual film when the residual film entered the packing room at the beginning. Another was the elastic deformation on both sides of the packing belt that was caused by the volume and weight of the residual film, which gradually increased with the increase of the feeding amount of residual film. A force analysis was performed on the residual film in the two stages of molding process. It was found that the improved torque of residual film greatly contributed to reducing the y direction and the angle between the front and back packing belt, as well as the rough packing belt. As such, a better performance was achieved to promote the forming of residual film. An analysis was also made on the motion of residual film bale in the packing room. The results showed that the increasing linear speed of the packing belt was promoted the forming of residual film. A single factor test was carried out with the linear speed of packing belt as the experimental factor. The test results show that the excessive linear speed of the packing belt led to a much more feeding/packing room with the residual film packaging thrown from behind the machine, whereas, the low linear speed of the packing belt caused the limited feeding packing room to fail to pack under the action of before and after moderate density of residual film bale. An optimal linear speed of the packing belt was selected between 0.742-1.167 m/s in this device. An orthogonal experiment was carried out with the density and balling rate of the residual film as the indicators, the anterior horn of packaging, the surface state of the packing belt and the linear speed of the packing belt as the factors. The experimental results show that the optimal operation parameters were achieved as follows. The anterior horn of the packaging was 35°, the surface state of the packing belt was rough, and the linear speed of the packing belt was 1.167 m/s. The balling rate of the residual film was 100%, and the average density of the residual film bale was 121.137 kg/m3, which indicating fully meeting the design requirements. The finding can provide a strong reference for the design and research of residual film packaging devices.