Measurements of acoustic vector quantities in the near field of a solid-fuel rocket motor are useful for enhancing prediction models related to rocket noise. However, the probes used to measure these quantities traditionally have low-frequency bandwidth limitations (e.g., below 45 Hz) thereby excluding the lowest, and in some cases, the loudest frequencies generated by large rocket motors. At these low frequencies, the phase and magnitude mismatch between microphones become greater and the acoustic phase separation between any two microphones becomes smaller, resulting in more error in estimating the pressure gradient between microphones. To investigate the low-frequency response of an acoustic intensity probe, a turntable is used to rotate a four-microphone probe with variable microphone spacing in a low-frequency noise field and an experimental assessment of the bandwidth is given for both magnitude and directional response. Also discussed is the effectiveness of a microphone interchange calibration technique to remove amplitude and phase mismatch and increase the usable bandwidth of the probe.