The characteristics of a thermal mass airflow sensor are typically reported as response time, or the time delay measured when responding to a step change in airflow rate. However, these measurements do not provide an accurate description of the dynamics that are used to predict system performance. Experimental time response is also difficult to measure and interpret. A number of problems arise in this including difficulty in creating a controlled step change in airflow, effects of flow rate on response time, and non-linear response to mass flow.In order to better explain the dynamic response of the air meter, a frequency domain approach proves useful. A simple model of the meter is a low pass filter with a cut-off frequency determined by the mass airflow rate and the frequency response. When this is combined with a normal static transfer function of mass flow to voltage, the mass airflow sensor is completely characterized. This paper presents the results of a test procedure that is capable of deriving the frequency response of a test sensor and a nonlinear model to produce time domain responses.