I6    The Maryland Experiment


A group of researchers from the University of Maryland (USA) carried out the Hafele and Keating experiment in a more scientific manner. Instead of ordinary commercial aircrafts flying imprecise routes, they used an aircraft of the U.S. Navy whose route was very carefully tracked during the whole flight. This plane could complete its circuits in15 hours flying at a height of 35,000 feet (~10,500 m). In addition, during the flight the time of the three atomic clocks on board was continually compared with that of the three clocks of the same type on the ground by exchanging laser pulses with a pulse width of 0.1 ns. The three clocks were carefully shielded against vibration, temperature variations, pressure variations and the influences of magnetic fields. The differences due to construction between the 6 clocks used in the experiment were accurately measured, before, between and after the flights, and their values were corrected according to these measurements.

After several test flights, five main flights, each lasting 15 hours were flown and analyzed. All six (corrected) clocks ran before, between and after these flights at the same speed, however, during the test flights they accumulated a difference, which exactly corresponded to the predictions of STR and GTR. The following plots are from the second of the five flights, which took place on November 22, 1975:


The graph on the right actually shows the same data as the one on the left but with an expanded time scale and also including the calculated influences of the STR and GTR. Since the flight speed (except for the relatively short periods during take-off and landing) was largely held constant (and as small as possible!), the STR ‘velocity effect’ shows itself as practically linear.

Including the influence of the gravitational potential one clearly sees “kinks” in the graph on the right. They arise from the fact that the aircraft had to first fly 5 hours at 25,000 feet before it had used enough fuel to allow it to fly at the next altitude of 30,000 feet. After a further 5 hours, it was light enough to climb to the targeted maximum altitude of 35,000 feet. These ‘altitude levels’ are particularly emphasized in the following graph:



All in all C.O. Alley and his team were able to confirm the predictions of STR and GTR to an accuracy of about 1.6%. A nice summary report of this experiment was written by Alley himself [39]. Our summary and the graphs were taken from that report.