Abstract: The C6D6 detector has the advantages of low neutron sensitivity, fast response and simple structure. It is often used to measure neutron radiation capture cross section on white neutron source. The key to measure the neutron radiation capture cross section with C6D6 detector is to obtain the weighting function with pulse height weighting technique, which makes the weighting efficiency be proportional to the excitation energy of the composite nucleus. At present, the main method to calculate the weighting function is to fit the deposition energy spectra of the C6D6 detector, which are obtained by simulating the scattering and absorption of gamma rays in the sample and the detector with MonteCarlo method. Therefore, the material and thickness of the sample are very important in the calculation. This work is based on the C6D6 detector system of the white neutron beam line (Backn) of the China Spallation Neutron Source (CSNS). The spallation neutron source is widely used in the field of neutron nuclear reaction data measurement because of its wide energy spectrum, high flux and short pulse. CSNS is one of the four largest pulsed spallation neutron sources in the world and is a key scientific device in China. The Backn on CSNS has a high neutron fluence rate and good energy resolution, and can provide a continuous neutron beam from thermal neutron to fast region for neutron radiation capture cross section measurement in the resonance region. The important nuclide 209Bi was taken as the research objects, the MonteCarlo simulation was performed and the pointwise weighting function of C6D6 detector was developed. The pointwise weighting function method takes the function as the form of matrix, solves the equations of gamma ray energy and weighting function and detector response matrix through LU decomposition and Tikhonov regularization, and finally obtains the weighting function. According to the simulation results of this work, the systematic uncertainty of weighting function of C6D6 detector system on Back-n increased rapidly with the increase of experimental sample thickness. This is harmful to the measurement of neutron radiation capture cross section, while the systematic uncertainty of pointwise weighting function remains less than 0.5% when the sample thickness reaches 6 mm, which improves the data accuracy of thick samples. This work provides a technical basis for the measurement of neutron radiation capture cross section of high atomic number and small cross section nuclides, such as 209Bi, on Backn in the future.