DAST: Exploring the Limits of Aeroelastic Structural Design

In the early 1970s, researchers at Dryden and NASA Langley Research Center sought to expand the use of RPRVs into the transonic realm. The Drones for Aerodynamic and Structural Testing (DAST) pro­gram was conceived as a means of conducting high-risk flight exper­iments using specially modified Teledyne-Ryan BQM-34E/F Firebee II supersonic target drones to test theoretical data under actual flight conditions. Described by NASA engineers as a "wind-tunnel in the sky,” the DAST program merged advances in electronic remote – control systems with advanced airplane-design techniques. The drones were relatively inexpensive and easy to modify for research purposes

and, moreover, were readily available from an existing stock of Navy target drones.[929] The unmodified Firebee II had a maximum speed of Mach 1.1 at sea level and almost Mach 1.8 at 45,000 feet, and was capa­ble of 5 g turns. Firebee II drones in the basic configuration provided baseline data. Researchers modified two vehicles, DAST-1 and DAST-2, to test several wing configurations during maneuvers at transonic speeds in order to compare flight results with theoretical and wind tunnel find­ings. For captive and free flights, the drones were carried aloft beneath a 9 DC-130A or the NB-52B. The DAST vehicles were equipped with

remotely augmented digital flight control systems, research instrumen­tation, an auxiliary fuel tank for extended range, and a MARS recov­ery system. On the ground, a pilot controlled the DAST vehicle from a remote cockpit while researchers examined flight data transmitted via pulse-mode telemetry. In the event of a ground computer failure, the DAST vehicle could also be flown using a backup control system in the rear cockpit of a Lockheed F-104B chase plane.[930]

The primary flight control system for DAST was remotely augmented. In this configuration, control laws for augmenting the airplane’s fly­ing characteristics were programmed into a general-purpose computer on the ground. Closed-loop operation was achieved through a teleme­try uplink/downlink between the ground cockpit and the vehicle. This technique had previously been tested using the F-15 RPRV.[931] Baseline testing was conducted between November 1975 and June 1977, using an unmodified BQM-34F drone. It was carried aloft three times for captive flights, twice by a DC-130A and once by the NB-52B. These flights gave ground pilot William H. Dana a chance to check out the RPRV systems and practice prelaunch procedures. Finally, on July 28, 1977, the Firebee II was launched from the NB-52B for the first time. Dana flew the vehicle using an unaugmented control mode called Babcock-direct. He found the Firebee less controllable in roll than had been indicated in simulations, but overall performance was higher.

Dana successfully transferred control of the drone to Vic Horton in the rear seat of an F-104B chase plane. Horton flew the Firebee through the autopilot to evaluate controllability before transferring control back to Dana just prior to recovery.

Technicians then installed instrumented standard wings, known as the Blue Streak configuration. Thomas C. McMurtry flew a mission March 9, 1979, to evaluate onboard systems such as the autopilot and RAV system. Results were generally good, with some minor issues to be addressed prior to flying the DAST-1 vehicle.[932] The DAST researchers were most interested in correlating theoretical predictions and experi­mental flight results of aeroelastic effects in the transonic speed range. Such tests, particularly those involving wing flutter, would be extremely hazardous with a piloted aircraft.

One modified Firebee airframe, which came to be known as DAST-1, was fitted with a set of swept supercritical wings of a shape optimized for a transport-type aircraft capable of Mach 0.98 at 45,000 feet. The ARW-1 aeroelastic research wing, designed and built by Boeing in Wichita, KS, was equipped with an active flutter-suppression system (FSS). Research goals included validation of active controls technology for flutter sup­pression, enhancement, and verification of transonic flutter prediction techniques, and providing a database for aerodynamic-loads prediction techniques for elastic structures.[933] The basic Firebee drone was controlled through collective and differential horizontal stabilizer and rudder deflec­tions because it had no wing control surfaces. The DAST-1 retained this control system, leaving the ailerons free to perform the flutter suppres­sion function. During fabrication of the wings, it became apparent that torsional stiffness was higher than predicted. To ensure that the flut­ter boundary remained at an acceptable Mach number, 2-pound ballast weights were added to each wingtip. These weights consisted of contain­ers of lead shot that could be jettisoned to aid recovery from inadvertent large-amplitude wing oscillations. Researchers planned to intentionally fly the DAST-1 beyond its flutter boundary to demonstrate the effective­ness of the FSS.[934] Along with the remote cockpit, there were two other

ground-based facilities for monitoring and controlling the progress of DAST flight tests. Dryden’s Control Room contained radar plot boards for monitoring the flight path, strip charts indicating vehicle rigid-body stability and control and operational functions, and communications equipment for coordinating test activities. A research pilot stationed in the Control Room served as flight director. Engineers monitoring the flutter tests were located in the Structural Analysis Facility (SAF). The SAF accommodated six people, one serving as test director to over­see monitoring of the experiments and communicate directly with the ground pilot.[935] The DAST-1 was launched for the first time October 2, 1979. Following release from the NB-52B, Tom McMurtry guided the vehicle through FSS checkout maneuvers and a subcritical-flutter inves­tigation. An uplink receiver failure resulted in an unplanned MARS recovery about 8 minutes after launch. The second flight was delayed until March 1980. Again only subcritical-flutter data were obtained, this time because of an unexplained oscillation in the left FSS aile­ron.[936] During the third flight, unknown to test engineers, the FSS was operating at one-half nominal gain. Misleading instrument indications concealed a trend toward violent flutter conditions at speeds beyond Mach 0.8. As the DAST-1 accelerated to Mach 0.825, rapidly divergent oscillations saturated the FSS ailerons. The pilot jettisoned the wingtip masses, but this failed to arrest the flutter. Less than 6 seconds after the oscillations began, the right wing broke apart, and the vehicle crashed near Cuddeback Dry Lake, CA.

Investigators concluded that erroneous gain settings were the pri­mary cause. The error resulted in a configuration that caused the wing to be unstable at lower Mach numbers than anticipated, causing the vehi­cle to experience closed-loop flutter. The ARW-1 wing was rebuilt as the ARW-1R and installed in a second DAST vehicle in order to continue the research program.[937] The DAST-2 underwent a captive systems-checkout flight beneath the wing of the NB-52B on October 29, 1982, followed by a subcritical-flutter envelope expansion flight 5 days later. Unfortunately, the flight had to be aborted early because of unexplained wing structural

vibrations and control-system problems. The next three flight attempts were also aborted—the first because of a drone engine temperature warning, the second because of loss of telemetry, and a third time for unspecified reasons prior to taxi.[938] Further testing of the DAST-2 vehi­cle was conducted using a Navy DC-130A launch aircraft. Following two planned captive flights for systems checkout, the vehicle was ready to fly.

On June 1, 1983, the DC-130A departed Edwards as the crew exe­cuted a climbing turn over Mojave and California City. Rogers Smith flew the TF-104G with backup pilot Ray Young, while Einar Enevoldson began preflight preparations from the ground cockpit. The airplanes passed abeam of Cuddeback Dry Lake, passed north of Barstow, and turned west. The launch occurred a few minutes later over Harper Dry Lake. Immediately after separation from the launch pylon, the drone’s recovery-system drag chute deployed, but the main parachute was jet­tisoned while still packed in its canister.[939] The drone plummeted to the ground in the middle of a farm field west of the lakebed. It was com­pletely destroyed, but other than loss of a small patch of alfalfa at the impact site, there was no property damage. Much later, when it was pos­sible to joke about such things, a few wags referred to this event as the "alfalfa impact study.”[940] An investigation board found that a combination of several improbable anomalies—a design flaw, a procedural error, and a hardware failure—simultaneously contributed to loss of the vehicle. These included an uncommanded recovery signal produced by an elec­trical spike, failure to reset a drag chute timer, and improper ground­ing of an electrical relay. Another section of the investigation focused on project management issues. Criticism of Dryden’s DAST program man­agement was hotly debated, and several dissenting opinions were filed along with the main report.[941] Throughout its history, the DAST program was plagued by difficulties. Between December 1973 and November 1983, five different project managers oversaw the program. As early as December 1978, Dryden’s Center Director, Isaac T. Gillam, had requested

chief engineer Milton O. Thompson and chief counsel John C. Mathews to investigate management problems associated with the project. This resulted from the project team’s failure to meet an October 1978 flight date for the Blue Streak wing, Langley managers’ concern that Dryden was not properly discharging its project obligations, repeated requests by the project manager for schedule slips, and various other indications that the project was in a general state of confusion. The resulting report indicated that problems had been caused by a lack of effective planning at Dryden, exacerbated by poor internal communication among project personnel.[942] Only 7 flights were achieved in 10 years. Several flights were aborted for various reasons, and two vehicles crashed, problems that drove up testing costs. Meanwhile, flight experiments with higher-profile, better-funded remotely piloted research vehicles took priority over DAST missions at Dryden. Organizational upheaval also took a toll, as Dryden was consolidated with Ames Research Center in 1981 and responsibility for projects was transferred to the Flight Operations Directorate in 1983.

Exceptionally good test data had been obtained through the DAST program but not in an efficient and timely manner. Initially, the Firebee drone was selected for use in the DAST project in the belief that it offered a quick and reasonably inexpensive option for conducting a task too haz­ardous for a piloted vehicle. Experience proved, however, that using off – the-shelf hardware did not guarantee expected results. Just getting the vehicle to fly was far more difficult and far less successful than origi­nally anticipated.[943] Hardware delays created additional difficulties. The Blue Streak wing was not delivered until mid-1978. The ARW-1 wing arrived in April 1979, 1% years behind schedule, and was not flown until 6 months later. Following the loss of the DAST-1 vehicle, the program was delayed nearly 2 years until delivery of the ARW-1R wing. After the 1983 crash, the program was terminated.[944]