NASA FRC

March 12, 1965

MEMORANDUM to X-15 Project Manager

Subject: Cockpit film analysis of X-15 flight 1-52-85

The following observations were made during an analysis of cockpit film, internal data, and pilot comments of flight 1-52-85.

1. The computer malfunction light came on at launch.

2. The Vertical Velocity indicator jumped to -200 fps. at launch, then gradually worked its way back to a more reasonable value.

3. The pitch angle presentation on the 3-axis ball appeared to be zeroed in line with the optical axis of the cockpit camera. This would result in approximately 3° to 4° parallax error at pilot eye level. This error would indicate to the pilot a lower pitch angle than the airplane was actually flying.

4. An attempt was made to correlate readings of pitch angle from cockpit film with internal recorded data. Five reasonably good correlations were made as follows:

 Flight Time (sec) q from film Internal q Apparent

Presentation

Error

27 14° - 15° 17.3° -2.3° to -3.3°

82 18° 23.6° 5.6°

108 8° - 9° 12° -3° to 4°

125 6° 5.4° +0.6°

366 -20° to -22° -19° -1° to -3°

The largest apparent discrepancies appear during, or shortly after, the boost phase. The parallax at pilot eye level would increase the apparent error.

5. If the attitude instruments were operating properly, the q vernier should have begun to move at a pitch angle of 23°. The pilot stated that he did not see it move. The cockpit film does not show any motion of the vernier although the 3-axis ball is not discernible during a large portion of the boost phase due to over-exposure.

6. The phasing of the angle of attack crosspointer was reversed.

7. The pilot stated that the computer malfunction lights are below his normal field of view and that he did not see the malfunction light until after reentry.

8. The altitude indicator became erratic at about 26 seconds after shutdown. From that point on, it appears to be completely unreliable, although it did settle down long enough to read the correct peak altitude.

 Comments 1. Pitch angle is quite difficult to read from the cockpit film, although it is felt that on the five points listed, the accuracy is within ±2°.

2. The reasons that only five q check points were picked are:

(1) Time correlation is very difficult because the cockpit camera does not run at constant speed, therefore, other means of correlation had to be used.

(2) During a large portion of the boost phase the film is over-exposed, consequently, readings were not possible.

 3. It appears that during the boost phase the visual pitch angle presentation was reading lower than actual airplane pitch angle. Recommendations 1. Have the pilot correct for parallax error prior to B-52 takeoff. This can be done by having the pilot set the 3-axis ball to read +2° q prior to B-52 takeoff. Add this to the pilot's check list.

2. Consider changing the calibration of the q vernier so that each graduation would represent 2° q instead of 1° as it is now. This should have two apparent advantages:

(a) The 2° per division sensitivity is more compatible with airplane control system dynamics.

(b) The pilot would see motion of the vernier 10° before the programmed q rather than 5° before, as it is now. This would give the pilot more than twice the time to observe pitch rate as he approaches the programmed q. (Due to erroneous information, the simulator had the 2° per division sensitivity up until about two months ago).

 3. Make the necessary adjustments on the cockpit camera to ensure better pictures. As it is now, the film coverage of the major portion of the boost phase is not readable. Also, it would seem advisable to devise a means to ensure a constant frame rate. Presently, time correlation is impossible.

4. If it is important for the pilot to be able to see the computer malfunction lights, locate them so that they are in his normal field of view.

5. The recommendation concerning the phasing of the angle of attack crosspointer is self evident.

6. An attempt be made to establish the accuracy of the pitch angle presentation on the next scheduled flight.

7. The next altitude flight be programmed for no more than 180,000 feet.
 
 

John A. Manke

Aerospace Engineer