Structural tests of the AIAA Wright 'Flyer'
To give an impression as to how delicate the structure of the 1903 Flyer is,
here is a picture of our first airplane with the fabric removed.
Prior to being allowed to test in the wind tunnel, NASA required the Wright Flyer Project to insure that nothing peculiar to our replica would cause damage to their facilities. Typically, this requirement is met with a thorough engineering analysis of the structure and ours was prepared by the late Carl Friend, a structural engineer at Lockheed Aircraft in Burbank. However, since wood and fabric can have variable mechanical properties and because our replica is so different from wind tunnel models normally tested by NASA, it was decided to perform a static proof test to corroborate the analysis.
The weights of the individual sandbags clearly marked on the plastic: 4.7 lbs near the leading edge, 2.9 lbs in the second row of bags, 1.6 lbs in the third row, etc.
Proof testing is not unusual. As part of the certification process, all airplanes must be tested to destruction to prove that the new design is as strong and as safe as the engineers claim it to be. These tests can be quite impressive, with the airplane installed in a huge truss structure surrounded by hundreds of hydraulic jacks pushing and pulling the wings up and down, applying forces to mimic what the airloads are around this airplane. Years ago, these tests were much simpler. Instead of hydraulic jacks pushing up on the wings, the airplane was flipped upside down and sandbags were stacked on the wings, thus using the weight of the sand to simulate the lift forces the wing would produce. Since our airplane is light and the loads we needed to apply were relatively small, we opted for the old-fashioned approach to test our Flyer. Fred Erb designed a rig to support the Flyer by hanging it from two steel wires attached to the steel ribs we had embedded within the substructure in order to connect to the large supports in the wind tunnel, and it took nearly a year to build.
The airplane mounted inverted (note the bend to the wings - the wingtips actually bend down) in the test rig. Notice the boxes containing the sandbags surrounding the airplane.
To satisfy the proof test requirements, we determined that we would need to show the airplane would not fail with a total of 2,913 lbs of sand placed on the wings, over 3.8 times the total loaded weight of the original 1903 Flyer! This did not mean we wanted to test our replica to 3.8 g's. NASA requires that any object placed in one of the wind tunnels have a safety factor on its strength to insure that during a test the object will not come apart and potentially do damage to the wind tunnel, because quite frankly wind tunnels are hugely expensive and take a long time to repair if something were to go wrong. For our test, NASA decreed that we prove the model would not break if subjected to forces three times what we would expect during the test - since we wanted to test up to slightly less than 1.3 times the weight of the Flyer: 1.29 x 3 = 3.87
There are seventy-six wing ribs and eighteen canard ribs (see the picture above), each of which required loading with a specific weight and pattern of sandbags. Loading a rib improperly, either by putting a sandbag in an incorrect location or simply placing the correct bag out of sequence could potentially damage the replica, so a great deal of planning and coordination was established prior to the actual test. Six teams, a knowledgeable member of the Project as team captain and two volunteers each, were established and assigned the responsibility to load a specific group of wing ribs. The canards required two similar two-person teams. Supporting the loaders was a team to measure deflections of the various surfaces, a group dedicated to documenting all proceedings with photographic and video coverage, and a final team of runners to pass all results along to be recorded on a large status board visible to all. The entire test effort was directed by Howard Marx, project Co-Chairman, as Load Control Chief.
For every rib, a box of carefully weighed and labeled sandbags was prepared that would permit four loading levels that approximate the shape of the airloads that the wing would experience in the wind tunnel. For the wings, 1,368 bags were prepared ranging from 0.7 lbs to 4.7 lbs each. For the canards, 144 bags ranged from 1.22 lbs to 4.77 lbs. Our Project alone did not have enough people to perform all the tasks required to load this many sandbags carefully, consistently, and methodically, so we asked friends, family, co-workers, the members of the men's fellowship at the local Covenant Church, and anyone else whom we thought would be interested to lend a hand for a day to help us conduct the test. All told, roughly sixty Project members and volunteers participated.
Howard Marx (left) supervises as a volunteer carefully places a bag of sand on the lower wing of the upside-down Flyer
On the morning of January 30, 1993, Project members and volunteers gathered for the first test attempt. Howard presided over an hour long briefing that detailed the test organization, basic plans, test setup, procedures, and cautions. There was even an independent set of ribs set up for "rehearsal" for Project members to demonstrate proper loading and for volunteers to practice. After this preparation, the official countdown began with loading sequencing being announced rib by rib via bullhorn. As the test proceeded from the first load point (32% of maximum test load) to the second load point (58% max load), two interplane strut fittings pulled loose from the wing spars. It was determined that the cause for this failure was the use of #6 wood screws rather than the #8 screws specified in the drawing. The wings were repaired and reinspected as was the rest of the machine, and a new test date was set.
At 9AM on February 27, 1993, the test team reassembled for a second go. After a brief pause to re-rig a center section flying wire, the test proceeded smoothly up to the maximum load condition required by NASA. As the last 2.38 lb sandbag was placed on the canard, one of the two center pitch control lever arms cracked under load, but did not fail. The maximum load condition was maintained for six minutes to measure all deflections and give the video/photo team time to completely document the achievement. Then, all surfaces were systematically unloaded under direction of Load Control.
Team Members crowd around looking at the canard after the test
Post test inspection found a slight looseness in two ribs that did not affect the results of the strength test and were quickly repaired. Likewise, although the canard control arms cracked and required repair, they did not fail. These results were documented and presented to NASA Ames, and our test was deemed a success. The results were also presented for public review in AIAA paper 93-3937, "Static Proof Load Tests of a Full-Scale 1903 Wright Flyer Fitted for Wind Tunnel Testing" by Howard Marx.
After the proof test was performed, the scheme chosen to mount the model to the wind tunnel was changed. This necessitated further analysis to insure the load paths to the new support system would not exceed the airplane strength. Since the proof test had proved the basic structure was adequate for the test, we could focus on the details of the attachment points using modern structural analysis tools.
Above is a screen capture of one of the Finite Element Model analyses performed on the AIAA Wright Flyer Replica.