February 15, 1962

Aircraft: X-15-3 #672 Date: 1/17/62

Flight: 3-2-3 Takeoff: 1105

Configuration: MH-96 Flt Cont. System Launch: 1200

Pilot: Neil A. Armstrong Land: 1211

Launch A/C: B-52 #003 Total: :11

Crew: Major Allavie, Bement

Chases: : Capt. Gordon

Cdr. Petersen

Maj. McDivitt

Maj. Rushworth
1. All pre-flight checks were satisfactory. The MH-96 Airborne Analyzer check had temporary self-correcting malfunctions in the center stick trim, the auto-trim, and both power checks. On the basis of satisfactory system engagement and no discouraging remarks from the FCS monitor personnel, it was decided to proceed with the flight.

2. The launch was performed on the side stick with all axes engaged in the adaptive mode. All axes remained engaged during the separation. The launch was performed with a small amount of left stick displacement and the normal right roll did not exceed 15° bank angle. The engine light was smooth and chamber pressure checked at 470 psi.

3. The pullup was performed with side stick trim only (3 units). Maximum angle of attack indicated was 10°. The pullup was stopped by returning the rate trim knob to zero and pitch attitude remained at 30.5° ± 0.5°. Roll attitude hold was engaged at approximately wings level attitude without transient. The pushover was initiated at 54 seconds with 3 units of nose down trim. Roll over powers were performed to bank angles of 30 degrees each direction. The returns were rather slow, but without overshoot and stabilized within one degree of the reference attitude. Reference changes of 10° bank angle were performed using the side stick CSS button and were accompanied with a small amplitude (< 0.1 g) limit cycle of approximately 15 cps. Lock-ons were satisfactory.

4. Angle of attack was noted at 1-2° and nose down rate trim was increased to the stop. The pitch attitude outer loop was engaged at approximately 85 seconds and ±5° q (indicated). The vernier trim and CSS functions were checked and found operative although CSS was again accompanied by the previously mentioned limit cycle. Since indicated angle of attack (» 6°) was greater than indicated pitch attitude (» 4°), a negative flight path angle was indicated although the inertial rate of climb was positive.

5. The engine shutdown at approximately 97 seconds with no apparent malfunction conditions. Inertial indications at shutdown were 5650 fps and 117,000 ft. altitude. The velocity indication was suspected to be several hundred fps high on the basis of several previous velocity checks during the trajectory. No thrust misalignments were noted during shutdown and both pitch and roll attitude outer loops remained engaged.

6. Inasmuch as the flight path angle was still in doubt and the possibility of a higher than planned velocity existed, the airplane attitude was maintained at a value which would insure deviations from the trajectory to be in the direction of lower dynamic pressure. When radar confirmation from NASA 2 indicated such a deviation did indeed occur, speed brakes were extended (without transient) and were retained in that position until the aircraft was in the landing pattern.

7. On a number of occasions and in particular at high Mach number and low dynamic pressure, a flight control limit cycle was detected for approximately 3-4 cycles, although the frequency was not determined. The phenomenon was reminiscent of the early days on the simulator when temporary super-critical gains were a frequent occurrence.

8. All three damper axes were reduced to fixed gain at approximately 4000 fps without transient. The outer loops properly disengaged. A primary flight objective was realized when a sizable time period was flown in the fixed gain configuration through the region of minimum controllability (unaugmented). No significant differences were noted over the 8°-17° angle of attack range evaluated. Although damping was low, particularly in the pitch axis, conventional piloting techniques were adequate to maintain reasonably precise control. Comparisons of specific dynamics with the simulator are quite difficult inasmuch as the airplane cockpit presentation is of considerably higher quality.

9. With the dampers remaining in fixed gain, the reaction controls were energized in the 'auto' position. Reaction control operation was observed during both control inputs and motion damping. Although only the left yaw, right yaw, and pitch down rockets could be checked visually, there was no evidence to indicate that the remaining rockets were not also operating satisfactorily. After reaction control operation had also been checked in the 'on' position, the flight control system was re-energized in the adaptive mode (with reaction controls 'on'). No evidence of reaction control was immediately noted, as would be expected, since the airplane response to a pulse was not sufficiently large to exceed the reaction control deadbands. As the aircraft slowed to lower speeds, however, these deadbands could be exceeded, and reaction control operation was once more in evidence.

10. CSS operation at subsonic speed was accompanied by the same type limit cycle previously described. The reaction control switch was left in the 'on' position until turning downwind and may account for the sizable hydrogen peroxide usage from the #l APU peroxide tank.

11. Although the inherent dangers of decelerating on a pitch rate command are well known, it had always been assumed that airframe buffet would provide an adequate warning when high angles of attack were approached. It was surprising, then, to see an indicated value of 20° while in a turn at 200 KIAS at subsonic speeds with no detectable buffet. Angle of attack was permitted to increase until an indicated value of 24° was obtained. Although the turn was accompanied by a modest amount of sideslip, no other evidences of degraded stability were evident. The urge to record a rudder pulse was rejected.

12. Pitch pulses and roll step at 200 KIAS indicated that time constants were negligible and it was decided to complete the approach and touchdown with all axes engaged in the adaptive mode. Handling qualities were felt to be satisfactory throughout the pattern, although no specific tests were performed, with the exception of a moderate pitch step subsequent to the flare. Operation of the side stick during the approach and landing was satisfactory. Geometry and force gradients seemed awkward after touchdown when maximum lateral deflections were required for directional control during runout. Touchdown was at the intended 2-mile marker and lateral dispersion during the runout reached several hundred feet.

(Signed)

Neil A. Armstrong

Aeronautical Research Pilot

NAA:bjc