In order to solve the problems of unknown bandsaw blade wear state and the deterioration of sawing quality in sawing process， the sawing behavior change of FTCUT bimetallic bandsaw blade during sawing of 42CrMoA alloy structure steel was studied and the relationship between acoustic emission signal and the single-blade sawing process， saw teeth wear and sawing surface quality change was discussed. The results indicate that with the change of sawing depth， the number of saw teeth increases first and then decreases and the constraints of the bandsaw change constantly， which affects the acoustic emission signal amplitude and sawing surface quality. Based on the time domain analysis， the saw teeth wear process can be divided into three stages： run-in wear， fast wear and stable wear. Based on the FFT frequency domain analysis， it is found that when there are multiple dominant frequencies in the spectrum diagram and the amplitude of the first dominant frequency decreases， the acoustic emission signal sources are scattered， and the bandsaw blade failure can be determined. Based on the analysis of wavelet packet frequency domain， it is found that the 7^th frequency band signal has the highest proportion and can reflect the change of workpiece surface quality.A certain relationship can be established between the fluctuation and micro grain and ripple of the sawing surface. Therefore， the saw teeth wear and sawing surface deterioration can be analyzed and judged based on acoustic emission signal in time domain and frequency domain.A certain idea for online monitoring of saw teeth wear and sawing surface quality is provided.