This paper reports on an experiment conducted with airline pilots on a fixed-based cockpit simulator. The objective was to assess the use of a speed managed mode for airborne spacing, compared to a speed selected mode. The availability of a speed managed mode reinforced the pilot acceptability. All pilots found airborne spacing feasible and compatible with their usual flying task, from cruise until automatic disengagement at 2000 feet. The speed managed mode was found as the most appropriate during the initial descent but the speed selected mode was preferred by some pilots during final approach. The spacing was well maintained below the 5 seconds tolerance margins with an average of-0.1 second and a 95% containment within ±2.5 seconds. The cost induced is in the order of 60 knots additional speed changes for the complete descent phase, with a maximum but also a minimum cost when using the speed selected mode. This raises the issue of trade-off between required performance (spacing accuracy) and cost induced (speed changes).