Abstract: In order to accurately measure the path of the incident particles in front of the target at the CSR outer target terminal, multiple sets of small cell multi-wire drift chamber detectors were developed with a drift cell of 5 mm. The sensitive area of this detector is 80 mm×80 mm, each set of detector includes x, x', y and y' 4 detection electrode surfaces, 2 sets of detection electrode surface wires are perpendicular to each other, and the same direction detection electrode surface wires are misaligned and distributed for providing left and right resolution, the misalignment distance of 2.5 mm for half drift cell, each electrode surface leads to 16 anode wire signals. The front-end electronics uses an amplifier based on the SFE16 chip, a 16-way ASIC signal processing developed by Saclay, France, whose functions include charge-sensitive amplification, shaping and signal screening, and outputting the TOT signal. The TOT signal is input to a data acquisition card based on the HPTDC chip developed by CERN, which records the rising and falling edge times of the TOT signal. The T₀ plastic flash detector, placed in front of the drift chamber, provides the onset moment of the incident particles. By subtracting the TOT signal front from T₀, the drift time of the incident particle traces in the multi-wire drift chamber can be obtained. The particles incident to the secondary target are physically concerned as heavy particles, so 2 sets of small cell multi-wire drift chambers were used to test the ¹⁶O beam position at 400 MeV/u. During the test, the operating gas of the detector was Ar(80%)+CO₂(20%), the anode wire voltage was +900 V, and both the field and cathode wires were grounded. After acquiring the test data for data processing, the R-T curves were obtained by simulation with interpolation correction, and the path traces were reconstructed by least squares method. After the data processing, the path residual of the single-layer electrode fitted to obtain the position resolution is 105.9 μm, and the path residual is defined as the difference between the distance from each hitting wire to the fitted straight line and the drift distance measured in each layer. The detection efficiency is 99.3%, and the detection efficiency is defined as the ratio of the count of the multi-wire drift chamber to the count of the plastic flash detector after deducting the background. The above indexes can meet the requirements of most of the nuclear physics experiments of the CSR external target terminal for the localization of the reaction target at this stage.