BACKGROUND: Resource planning is essential for successful transport of the mechanically ventilated patient. Mechanically ventilated patients require adequate oxygen supplies to ensure transport is completed without incident. The LTV-1000 portable ventilator utilizes a program to calculate oxygen cylinder duration, based on cylinder size, fraction of inspired oxygen (FIO2), and current minute ventilation. We evaluated the accuracy of the cylinder-duration algorithm in a laboratory setting. METHODS: The LTV-1000 was attached to a test lung. Lung compliance was set at 0.04 L/cm H2O, and airway resistance was 5.0 cm H2O/L/s. We tested 7 different combinations of ventilator settings a minimum of 2 times each. With each setting, minute ventilation was kept at 10 L/min. Breath type, positive end-expiratory pressure, and FIO2 were varied to evaluate the accuracy of the algorithm across a range of clinical scenarios. The cylinder-duration calculation from the ventilator program and manual calculation was determined at each setting and compared to the actual cylinder duration. RESULTS: The ventilator algorithm and the manual calculation underestimated the actual cylinder duration by 12 ± 3% with each test. The range of differences between calculated and actual cylinder duration was 2‐26 min across the 7 conditions. CONCLUSION: Actual cylinder duration averaged 12% longer than the cylinder duration estimated by the algorithm of the LTV-1000. One explanation is that the E cylinders may contain more liters of oxygen than indicated by the sticker on the side of the tank. Additionally, the bias flow during expiration is affected by inspiratory-expiratory ratio and respiratory rate. Clinicians should be aware of these differences when planning for patient transport.