The aim of this study was to determine whether an artificial magnetic field with an amplitude and frequency equivalent to those of geomagnetic pulsations during geomagnetic storms could affect physiology and psychology. Three healthy volunteers wore anambulatory BP monitor and an ECG recorder around the clock for 12 consecutive weekends in Winnipeg, Manitoba, Canada. In a room shielded against ELF and VLF waves, they were exposed for 8 hours per week to either a 50 nT 0.0016 Hz or a sham magnetic field at one of six circadian stages. Real exposure randomly alternated with sham exposure. They provided saliva and recorded mood and reaction time every 4 hours while awake. Systolic (S) and diastolic (D) blood pressure (BP), and heart rate (HR) were recorded every 30 minutes. Spectral analysis of HR variability (HRV) was performed using the maximum entropy method and a complex demodulation method. For these variables, daily means were compared between real and sham exposure, using paired t-tests. Their circadian MESOR, amplitude, and acrophase were analyzed and summarized using single cosinor and population-mean cosinor. Circadian rhythms were demonstrated for HR, SBP, DBP for sham exposure, salivary flow rate, positive affect, vigor, and subjective alertness (p < 0.001, -0.02). One participant showed higher HR, lower LF, HF, and VLF powers, and a steeper power-law slope (p < 0.005, -0.0001) in an early night exposure to the real magnetic field, but not in other circadian stages. There was no significant difference between circadian responses to real and sham exposure in any variable at any circadian stage.