SRch3

NASA Investigation Board Report On The INITIAL FLIGHT ANOMALIES OF SKYLAB 1

CHAPTER III : DETAILED ANALYSIS OF FLIGHT DATA

63 Second Anomaly - Loss of MS

The Investigation Board, evaluated the telemetry data in order to explain the various anomalies that occurred on Skylab 1. The first anomalous indication was an increase in S-II telemetry reflected power from a steady 1.5w beginning at R+ 59. 80 seconds. At this time the telemetry forward power remained steady at 58.13w. By 61.04 seconds, the reflected power had reached 1.75w, and by 80.38 seconds, the reflected power had stabilized at about 2.0w. This abnormal increase in power might be indicative of a vehicle physical configuration change which altered the antenna ground plane characteristic.

Shortly after the telemetry reflected power increase, the MS torsion rod 7 forward (measurement G7036) indicated a slight change toward the deployed condition (see fig. 2-5 for instrumentation layout). This occurred at R+60.12 seconds, and at 61.78 seconds the vehicle roll rate decreased slightly from a normal value of 1.1 degrees per second clockwise (CW) looking forward. Figure 3-1 is a graph of the roll rate versus range time during the time of interest. The next torsion rod 7 forward sample at about 62.52 seconds revealed a further relaxation. The increase in telemetry reflected power and the movement of torsion rod 7 forward tend to indicate meteoroid shield lifting between positions I and II (see fig. 2-5).

Between R+62.75 and 63.31 seconds, several vehicle dynamic measurements indicated a significant disturbance. A sensor on the OWS film vault showed an abnormal vibration at 62.75 seconds followed by disturbances sensed by X and Y accelerometer pickups in the Instrument Unit (IU), the pitch, yaw, and longitudinal accelerometers, and the pitch, yaw, and roll rate gyros. At 62.78 seconds, the roll rate gyro sensed a sudden CW roll rate resulting in a peak amplitude of 3.0 degrees per second CW at 62.94 seconds. A sensor at the X upper mounting showed a maximum peak-to-peak shock of 17.2 g's at 63.17 seconds. In addition, the S-II engine actuators experienced pressure fluctuations caused by vehicle movement against the inertia of the non-thrusting engine nozzles.

During the time the vehicle was sensing a disturbance, several slower-rate MS and SAS measurements experienced drastic changes. Because these measurements are sampled only once every 0.1 to 2.4 seconds, there is that period of uncertainty as to when the measurement has actually changed. Figure 3.2 is a graphic representation of the applicable measurements associated with the 63-second anomaly. Where only a single point is shown, the sampling is continuous or has no significant bearing on the hypotesization of the MS failure mode. For the MS and SAS data sampled at 0.1, 0.8, and 2.4 seconds per sample, the last normal and first abnormal times are shown. Figures 3-3 , 3-4 , 3-5 , 3-6 , 3-7 , 3-8 are pictorial representations of the status of the MS and SAS measurements at the indicated period of time. Figure 3-3 is a time slice at R+60.90 seconds where all measurements are known to be normal for the last time (except for the slight movement of torsion rod 7 forward beginning at 60.12 seconds).

Figure 3-4 is a time slice at the first indication of a measurement failure (R+62.78 seconds). The measurements K7211, C70132, K7010, K7011, and K7012 can be considered normal here because they were normal during the previous sample and were sampled later than 62.78 seconds and found to still be normal. At this time period, C7011 (a temperature measurement) was lost. The cause of this measurement failure could have been due to the sensor or its cabling (shown in fig. 3-4 by dashed lines) being damaged. This was most likely a result of the NS failure in the area between the SAS-2 wing and the main tunnel (between positions I and H). Furthermore, both SAS wing secure indications and the ordnance tension strap indications are known to be good. This evidence leads to two conclusions at this point: the meteoroid shield failure began prior to the SAS-2 wing becoming unlatched, and the ordnance did not fire prematurely.

Figures 3-5 and 3-6 are time slices at R+62.89 and R+62.90, respectively, that show the failure of measurements C7012, K7010, K7011, and K7211, while K7212 (SAS-1 secured) and C7013 (MS temperature) were known to be normal by a later sample, The abnormal telemetry indications C7012, K7010 and K7011, like C7011 at R+62.78 seconds, could have been due to sensor or wiring damage. Measurements K7010, K7011, and K7012 are, in fact, only breakwires placed across the ordnance tension strap. Measurement K7211, however, reveals that the SAS-2 wing was no longer secure to the OWS. This is an indication that the SAS-2 wing had moved out at least between 0.651 and 2.821 degrees, or between 4.66 and 20.2 inches as measured at the aft end of the wing perpendicular to the OWS.

Figure 3-7 represents a later time, 62.97 seconds. At this time, K7012 (tension strap) was detected as failed. Slightly later, at R+63.04 seconds, the first indication of increased SAS voltage appeared. Measurement M0103 showed a slight increase in voltage which is attributed to sunlight illuminating exposed sections of the partially deployed (unlatched) SAS-2 wing. Other SAS voltages fluctuated throughout the remainder of the launch phase for the same reason. Between 62.97 and 64.92 seconds, an of the MS failure-related measurements became abnormal. Figure 3-8 shows that the SAS-1 wing secure measurement (K7212) was still normal.

The data indicate that the most probable sequence of Meteoroid Shield failure was initial structural failure of the MS between the SAS-2 wing and the main tunnel (between positions I and II). The initial failure propagation from this area appears likely since the wardroom window thermocouple indication (C7013) remained normal at 62. 94 seconds after SAS- 2 indicated unlatched at 62.90 seconds and after the K7010 and K7011 tension strap measurements failed.

593 Second Anomaly

As a consequence of the MS failure at approximately 63 seconds, the SAS-2 wing was unlatched and partially deployed as evidenced by minor variations in the main SAS electrical voltages and SAS-2 temperatures. Full deployment was prevented due to the aerodynamic forces and accelerations during the remainder of powered flight.

At the completion of the S-II phase of flight the four 35, 000 pound thrust retro-rockets fired for approximately two seconds commencing at R+591.10 seconds followed by spacecraft separation at 591.2 seconds. The effect of retro-rocket plume impingement (refer to fig. 3-9 for location and orientation of the retro-rockets relative to the SAS-2 wing) was observed almost immediately on the SAS-2 temperature and on vehicle body rates.

The time sequence of observed changes in the affected measurements is demonstrated in figure 3-10. The response of the vehicle and the corrective action of the attitude control system may be seen in figures 3-11 and 3-12.

An analysis of the impingement forces on the wing was made and compared to the force required to produce the observed vehicle motion. This comparison provides a reasonable fit for the first 50 to 60 degrees of wing rotation as shown in figure 3-13.

At 593.4 seconds the wing imparted momentum to the vehicle, probably by hitting and breaking the 90 degree fully deployed stops and at 593.9 imparted a final kick as it tore completely free at the hinge link. In-orbit photographs show clearly the hinge separation plane and the various wires which were torn loose at the interface (see fig. 3-14).

Interstage Second Plane Separation Anomaly

Post-flight analysis revealed unexpectedly high temperatures and pressures in the S-II engine compartment following ignition and continued high after interstage, separation command as shown in figures 3-15 and 3-16. The unusually high temperatures from S-II ignition and until the S-II interstage separation signal are considered by MSFC to be caused by a change in the engine heat shield skirts introduced on this flight, and therefore do not indicate a problem. However, the increasing temperatures after the time of normal S-II interstage separation are indicative of an abnormal condition. More detailed investigation based on performance evaluation and axial acceleration time history revealed that the interstage had not been jettisoned; however, due to the vehicle performance characteristics and performance margin, the desired orbit was achieved.

Data analysis confirms that the primary ordnance command was properly issued at R+189.9 seconds. The back-up command was issued 100 milliseconds (ms) later but the exploding bridge wire circuit discharge was characteristic of an open circuit consistent with separation of the interstage disconnect by a minimum of 0.2 5 inch as shown in figure 3 -17.

The linear shaped charge (LSC) is mounted circumferentially around the S-II interstage as shown in figures 3-18 and 3-19. When fired by the primary command, the charge cuts the tension straps (in the direction of position II to position I) allowing the skirt to drop away. Normal propagation time of the LSC is approximately 4ms. Assuming a failure to propagate completely around the structure, analyses were made by appropriate contractor and the government personnel to determine what area must remain intact in order to retain the skirt and what area must have been cut to allow rotation of the skirt sufficient to disconnect the connector panel. An example of the results of one analysis is shown in figure 3-20. The various analyses isolate the region of failure to an are extending from approximately 0 = 100 degrees to as much as 0 = 200 degrees.

This ordnance installation was different from prior Saturn flights. Previously, a single fire command from the instrumentation unit was issued which simultaneously detonated the LSC from both ends allowing the charge to propagate from both directions. On this flight, in an attempt to provide redundant firing commands, the detonators at each end of the LSC were separately connected to two command channels spaced 100 milliseconds apart due to the characteristics of the airborne equipment. As a result of the partial cutting of the interstage, it rotated sufficiently to separate the electrical connector prior to issuing the back-up command.

A review of the history of manufacturing, acceptance, checkout, qualification and flight environment revealed no basic cause for failure. The most probable cause is secondary damage as a result of the MS failure, attributed to falling debris as evidenced by the various shock and acoustic disturbances occurring in the 63-second time period.

The redundant mode of ordnance operation of all prior Saturn flights in which both ends of the LSC are fired at once from a single command would probably have prevented the failure, depending on the extent of damage experienced by the LSC.

Forward Interstage Internal Pressure Anomaly

Flight data indicated a deviation of the S-II forward interstage pressure from analytical values commencing at approximately 63 seconds. Inasmuch as the deviation from the analytical curve of internal pressure versus time appeared to be coincident with the MS failure (see fig. 3-21) it was postulated that a portion of the shield had punctured the forward interstage. On this basis, it was possible to correlate the flight data with either an assumed 2.0 square foot hole in the conical section or an assumed 0.75 square foot hole in the cylindrical section.

Range Safety Receiver Anomaly

During the S-II portion of the flight, the signal strength indications from both range safety receivers showed drops in level. From liftoff through R+259 seconds, both receivers maintained relatively stable values above range requirements. At R+259.57 seconds, receiver 2 signal strength began to drop and between this time and 522.1 seconds, both receivers indicated various degrees of signal strength shift. These signal strength shifts dropped below the 12 db safety margins required by Air Force Eastern Test Range (AFETR) Manual 127-1. At R+327.81 seconds, the receiver 2 signal strength dropped briefly below its threshhold sensitivity. At this instant this receiver probably would not have responded to any range safety commands. Receiver 1 was, however, capable of receiving commands. At R+521.16, receiver 2 strength again dropped briefly to its threshhold sensitivity. None of these drops could be correlated to ground system performance.

Analysis indicates that the most probable cause of the S-II receiver signal strength dropout was a variable phase shift within the vehicle's hybrid coupler due to the changing aspect angle produced by the moving vehicle and the fixed transmitting site. Because the decrease in receiver signal strength occurred with only one receiver at a time, range safety commands could have been received continuously throughout power flight. During two of these drops, however, the planned redundancy of range safety receivers was not available.

During this investigation, it was revealed that the Wallops Island and Bermuda ground stations did not continuously record ground transmitter power levels. The Board considers that such continuous recordings would be of value.

Figure 3-1 - Roll rate versus range time.