Development difficulties
A static test of the C-17 wing in October 1992 resulted in the wing failing at 128% of design limit load, which was below the 150% requirement. Both wings buckled rear to the front and failures occurred in stringers, spars and ribs. Some $100 million was spent to redesign the wing structure; the wing failed at 145% during a second test in September 1993. A careful review of the test data however, showed that the wing was not loaded correctly and did indeed meet the requirement.[19] The C-17 received the "Globemaster III" name in early 1993. In late 1993, the Department of Defense gave the contractor two years to solve production and cost overrun problems or face termination of the contract after the delivery of the 40th aircraft. By accepting the 1993 terms, McDonnell Douglas incurred a loss of nearly US$1.5 billion on the development phase of the program.
In April 1994, the C-17 program remained over budget, and did not meet weight, fuel burn, payload and range specifications. It also failed several key criteria during tests to evaluate its airworthiness. There were technical problems with mission software, landing gear, and so on. In May 1994, a proposal to cut the aircraft's production to as few as 32 aircraft emerged, but was rescinded. A July 1994 GAO document revealed that Air Force and DoD studies from 1986 and 1991 stated the C-17 could use 6,400 more runways outside the U.S. than the C-5; it was later discovered that this study had only considered the runway dimensions, but not their strength or Load Classification Numbers (LCN). The C-5 has a lower LCN, but the USAF classify both in the same broad Load Classification Group (LCG). When considering runway dimensions and their load ratings, the C-17's worldwide runway advantage over the C-5 shrank from 6,400 to 911 airfields.[26] However, the C-17's ability to use lower quality, austere airfields was not considered.
A January 1995 GAO report revealed that while the original C-17 budget was $41.8 billion for 210 aircraft, the 120 aircraft already ordered at that point had already cost $39.5 billion. In March 1994, the U.S. Army had decided it no longer needed the 60,000 lb (27,000 kg) Low Altitude Parachute Extraction System (LAPES) delivery with the C-17, feeling that the 42,000 lb (19,000 kg) capability of the C-130 was sufficient; C-17 testing was limited to this lower weight. Issues with airflow prevented the C-17 from meeting its airdrop requirements. A February 1997 GAO report revealed that a C-17 with a full payload could not land on 3,000 ft (910 m) wet runways; simulations suggested 5,000 ft (1,500 m) was required.
The YC-15 was transferred to AMARC to be made flightworthy again for further flight tests for the C-17 program in March 1997. By the mid-1990s, most of the problems had been resolved. The first C-17 squadron was declared operational by the U.S. Air Force in January 1995. In 1996, DoD ordered another 80 aircraft for a total of 120. In 1997 McDonnell Douglas merged with its former competitor, Boeing. In April 1999, Boeing proposed to cut the price of the C-17 if the Air Force bought 60 more, and in August 2002, the order was increased to 180 aircraft.
