To clarify the correlation between vibration and thrombus formation in a centrifugal blood pump, a preliminary simulated thrombus study was conducted for possible detection of thrombus formation inside a pump. Additional in vitro thrombogenesis studies were performed to confirm the results of the preliminary study. The primary data acquisition equipment included an accelerometer (Isotron PE accelerometer, Endevco, San Juan Capistrano, CA, U.S.A.), digitizing oscilloscope (TDS 420, Tektronic, Inc., MA, U.S.A.), and pivot bearing centrifugal pumps. The accelerometer was mounted to the top of the pump casing to sense radial and axial accelerations. For the preliminary study, a piece of Silastic was adhered to each of the 3 common areas of thrombus formation inside the pump. The results provided baseline information to speculate on the possibility of detecting thrombus formation by vibration signal changes. For the next studies, fresh bovine blood was harvested under sterile conditions and with strict avoidance of air contact, adding 1.0 U/ml of heparin. The sterilized test circuit consisted of 3/8 inch tubing (Tygon) and a soft reservoir. During the operating time, the activated clotting time (ACT) was maintained between 150 to 300 s using protamin. A restrictor on the outflow tube maintained the flow rates at about 4.5 L/min. The pumps ran continuously for 6 h. Possible blood clot formation inside the pump was monitored by observing the vibration signal from the device for 6 h. These studies revealed that it was possible to distinguish between an impeller that did not form thrombus and ones that formed fibrogenous thrombus using vibration signal assessment. Vibration assessment is worthwhile as a thrombus monitoring tool for a centrifugal blood pump.