SRch5

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

CHAPTER V : HIGHLIGHTS OF THE METEOROID SHIELD HISTORY

The meteoroid shield of the OWS had its origin in a letter request of November 1, 1966 from MSM to the Douglas Aircraft Company to submit an expedited ECP for a description of "systems feasible as protection against probable meteoroids." A brief proposal was submitted to MSFC in response to this request on December 7, 1966. There followed a submittal of design criteria for the MS by MSFC stating, among other things, that it "shall be designed as a structurally integrated part of stage 209 capable of withstanding the dynamic forces imposed during the orbital workshop mission" and that "the weight of the bumper system shall be a primary design consideration." Protection from meteoroid penetration with a probability of 0.9950 of no penetration for a 12-month mission was also specified. The ECP of December 7, 1966 was approved on March 16, 1967 and work on the design and development of the shield was initiated, leading to the following project milestones and events.

May 2-10, 1967	Preliminary Design Review on the orbital workshop.

December 12-15, 1967	A "Delta" Preliminary Design Review on the orbital work shop.

July 22, 1969	NASA decision to change from a "wet" Saturn S-IVB launched workshop to a "dry" Saturn V launched workshop. At the same time, "Skylab" as the program was to be later designated, became a major line-item program in its own right, independent of Apollo funding and schedules.


August 20, 1969	Supplemental agreement No. 2 to contract NAS9-6555 authorizing the change from the "wet" to the "dry" workshop. In accordance with Skylab program policy that no unnecessary changes be made, and based on the full confidence of MSFC and MDAC-W management in the current design, the basic concept and general design of the meteoroid shield was retained.

August 1969	Provisions made in the meteoroid shield for the ground access door, the wardroom window and the scientific airlocks.

November 1969	The introduction, via a MDAC-W internal memo, of vents and internal baffles in the auxiliary tunnel to reduce aero-dynamic loads on the tunnel frames.

December 1969	Cluster System Review

September 14-18, 1970	Critical Design Review of orbital workshop.


November 1970 - April 1971	The "Mathews" Subsystem Review

January 1971	Rework of the butterfly hinges and adjacent panels of the MS to accommodate a misalignment in the bonding of the tension straps to the OWS.

March - April 1971	First reference to possibility of meteoroid shield 'burst" pressures in an internal MDAC-W memo; venting of shield by drilling holes in the panel joints proposed as solution in a responding memo.

March 17-18, 1971	Critical Mechanisms Review

April 29, 1971	Completion of re-qualification of expandable tube explosive assembly.

May 10, 1971	First deployment test at MDAC W. Shield failed to fully deploy because of gravity loading on the linkage mechanisms.

May 14, 1971	Test failure of expandable ordnance tube requiring re-design and retest.

May 18, 1971	Completion of qualification testing of MS ordnance system.


August 1971	An internal MDAC-W memo suggesting a change to an unbaffled auxiliary tunnel with the aft end sealed and a vent at the forward end.

September 14, 1971	ECP for design changes arising, in part, from the failure of deployment test. Changes included improvements to ordnance, in creased torque on swing links, increased auxiliary tunnel frame thickness and redesigned hinge pins. Decision to conduct all future deployment tests at MSFC on the static test article (STA).

February 10, 1972	Completion of re-qualification of expandable tube and strap assembly after tube wall rupture.

March 1972	Completion of series 1 deployment tests. Distance between fold-over panel and tank insufficient to provide meteoroid protection. Rigidity of shield also insufficient to cause panel to chord. Scroll springs accordingly added to four corners of the fold-over panel.

April 1972	Completion of series 2 deployment test. Latches failed to engage. Latch modified and spring relocated to provide greater angular motion.


April 15, 1972	Qualification of full-scale (22 feet) redesigned expandable tube and strap assembly on back-up structure.

May 31, 1972	Doublers added to butterfly hinges on SAS-2 side of main tunnel because of failure of hinges during tank pressurization test at MSM.

May 1972	Special development test on deployment latch mechanism.

June 19-23, 1972	Design Certification Review of orbital workshop. August 1972 The completion of four mechanical and three ordnance deployment tests at MSFC on STA using OWS- 1 MS flight hardware.

September 8, 1972	OWS- 1 shipped to KSC. Arrives on September 22, 1972.

October 28, 1972	Discrepancy Report DR 0136 developed at KSC due to difficulty encountered in rigging the shield tight to the tank (see next section entitled "Rigging the Meteoroid Shield for Flight").

November 10, 1972	Discrepancy Report DR 180 written at KSC on gaps between the shield and tank. Contains detailed mapping of the areas of such gaps and MRB disposition to "use as is."


April 3, 1973	Hardware Integrity Review

April 18, 1973	Flight Readiness Review

May 14, 1973	Launch

As may be noted from the foregoing, the meteoroid shield experienced a remarkable stability of design, both in basic concept and in most of its design details, throughout its six year history. Principal development difficulties, and hence engineering effect, concerned the achievement of an ordnance system that would release the shield without a rupture of the expandable tube and in several minor details of deployment such as locating latches or obtaining the desired deployment distances. The major structural failure experienced in testing was of the butterfly hinge, which was more of a ground test than a flight problem, and was readily solved by the addition of hinge doublers.

No deployment tests were conducted under vacuum conditions, which is quite acceptable in view of the low rate of motion of the deployment. Vibration, acoustic and flutter tests were specifically omitted in the test specifications because of the design requirement that the shield be "tight to the tank." This design requirement and pervading philosophy of design and development also served to omit all aerodynamic tests of the meteoroid shield. The major difficulty experienced with the meteoroid shield was in getting it stowed and rigged on the OWS. Handling such a large, lightweight structure proved difficult, requiring the coordinated action of a large group of technicians, and considerable adjustments to the assembly of the various panels were necessary in an effort to obtain a snug fit between the shield and the OWS wall. The specific procedure used for rigging the shield for flight is discussed in the following section.

Rigging the Meteoroid Shield for Flight

The condition of the MS at launch is sensitive to the rigging procedures used to secure the shield around the OWS. For this reason, the rigging procedures used at KSC to prepare the US for launch are summarized in the following.

The flight MS was shipped to KSC from MDAC-W fully installed, but not flight rigged, on the OWS except for the installation of the ordnance panel. This panel was later installed at KSC after the mechanical deployment tests were completed. The deployment tests which were conducted earlier at MSFC utilized a static test article (STA) version of the shield.

The rigging procedure that was to be used at KSC was developed jointly by MSFC and MDAC using the STA at MSFC. The STA shield was, however, different from the flight MS in four significant aspects. On the flight MS: (1) the double butterfly hinges on the SAS 1 side of the main tunnel were bonded to the tension straps while on the STA they were present but unbonded; (2) the butterfly hinges on each side of the main tunnel were cut in the middle of a longitudinal joint and refitted to the adjacent panels at a slight angle as mentioned earlier. The longitudinal edges of the panels were also modified to suit the altered hinge line. This change to the flight MS at MDAC was necessary to accommodate the misalignment which occurred in the location of the tension straps on the OWS; (3) a longitudinal misplacement of the tension straps of 0.15 inch too high also resulted in some binding of the forward weather seal and torsion rods that had to be refitted at KSC; and (4) the trunnion bolts. nuts and washers were initially not lubricated on either the flight MS or the STA. This lack of lubrication caused difficulties in the final rigging of the shield at KSC, which was subsequently corrected by applying a solid film lubricant.

The MS delivered for flight was therefore not identical to that used to develop the rigging procedures. In this sense, the flight shield was also the development and qualification unit.

An abbreviated version of the rigging procedure developed at MSFC on the STA is presented below. This procedure was used initially at KSC in rigging the flight MS.

1.	Install zero-g kit.

2.	Rotate shield against tank wall. (This process required the coordinated action of a group of technicians, one each at the forward and aft end of each of the sixteen panels.)

3.	Loosely install trunnion bolts.

4.	Remove auxiliary tunnel covers.

5.	Spread each of the 28 auxiliary tunnel frames 1-3/8 inches + /- 1/8 inch with a spreading fixture.

6.	Adjust trunnion bolt nuts finger tight, with distance between trunnion straps uniform within + 0.10 inch.

7.	By sequential end-to-end passes, torque the center 12 trunnion bolts to a value of 100 inchpounds. This condition to be reached while maintaining a uniform spacing between trunnion straps of less than + 0.10 inch. Sequential adjustments of all bolts to be such that at steps of 50 and 75 inch-pounds, an equal torque on all 12 bolts is obtained.

8.	Remove the auxiliary tunnel spreading fixture.

9.	Back off the 12 trunnion bolt nuts 3 revolutions.

10.	Retorque the 12 trunnion nuts up to 1 revolution, not to exceed 45.0 inch-pounds. (Revolution was the controlling factor.)

11.	Torque the top and bottom trunnion bolt nuts to 20 inch-pounds.

12.	Torque the trunnion bolt jam nuts to 100 inchpounds and lock wire.

This rigging procedure did not produce a satisfactory fit of the field to the OWS tank wall. Several "bulges" were evident where the shield was not snug to the tank and significant gaps, of up to an inch in extent at one point, existed between the shield and the tank along the upper and lower edges of the shield assembly. The rigging procedure furnished to KSC therefore had to be modified considerably in an effort to produce a tightly fitting shield. These modifications are noted in a discrepancy report (D.R. No. 0136), a summary of which is presented below.

1.	Loosen seal retainers on bottom of two pairs of panels.

2.	Loosen panel joint bolts along bottom 36 inches of a pair of panels and push panels against tank. Retighten bolts.

(A gap still existed.)

3.	Loosen 6 bolts in the other pair of panels, install a "puller" to one of these bolts, and pull panel out in mid-region and push bottom in against tank. Repeat this pushing and pulling 10 times.

4.	While pushing the shield against the tank, loosen 4 other bolts and tighten the others. Push and pull again in this region 10 times.

5.	Repeat step 4 with another pair of bolts loose. Re-tighten 2 bolts.

6.	Repeat step 5 with another pair of bolts loosened.

7.	Continue this procedure until one has worked his way up along the entire length of two longitudinal joints between two pairs of panels. Remove the "puller" and replace flight bolts.


8.	Tighten bottom two trunnion bolts 2 turns, with torque not to exceed 45 inch-pounds.

9.	Tighten third from bottom trunnion bolt 1 turn, torque not to exceed 45 inch-pounds.

10.	Remove shims on bearing blocks, loosen seal retainers on two more panels, and bolts from a strap on the ordnance panel.

11.	Loosen three bottom trunnion bolts (i.e., undo steps 8 and 9).

12.	Measure tangential length of ordnance panel, the gap between trunnion straps, and the spread of tunnel frames.

13.	Torque bolts to 90 inch-pounds.

14.	Repeat measurements of step 12.

15.	Install flight panel over access door.

16.	Remove top and bottom trunnion bolts.

17.	Loosen up and re-torque the center 12 bolts to 45 inchpounds.

18.	Remove these 12 bolts, apply lubricant, and torque to 45 inch-pounds.

19.	Torque up the trunnion bolts in sequential endto-end passes to 96 inch-pounds, maintaining a uniform gap between straps within + 0.06 inch.

20.	Back off trunnion bolt nuts and torque to 45 inch-pounds.


21.	Install top and bottom trunnion bolts, torque to 20 inch-pounds.

22.	Install jam nuts and lock wire.

23.	Measure and record torsion bar / strut fitting relationship (all okay).

24.	Put the bulb seal, back in place.

(A gap still existed along edge of panel.)

25.	Remove splice plate between two panels (one of the pair previously re-worked) and slot holes in shield flange.

26.	Reinstall splice plate loosely and push in on the panel joint. While pushing, tighten up the splice plate.

27.	Set the bulb seal against the tank.

28.	Verify torque on aft bolt at 20 inch-pounds, re-install jam nuts and lock wire.

At the conclusion of the rigging described above, the contact areas, that is, the areas of the shield which were snug to the tank, were mapped, and it was determined that only 62% of the shield was in contact with the unpressurized OWS. The OWS was then pressurized to 8 psi above ambient and the contact areas again mapped. In this condition, about 95% of contact was achieved. Much of the remaining gapping occurred along the forward edge of the MS. Since the flight differential pressure was substantially higher than 8 psi, it was felt that the contact area during the flight would be substantially higher than 95%. The condition of the MS was therefore formally accepted as satisfactory for flight on January 10, 1973. No further adjustments were made to the MS prior to the flight.

From the foregoing, it is apparent that the MS was very difficult to rig to the tank. Many hands were required to push and pull on various joints in the shield while groups of panel bolts were sequentially loosened and tightened in an effort to obtain a snug fit. As a result, the final rigging used prior to flight differed markedly from that used any previous time. More importantly, the actual condition of the shield in terms of final tension in the tunnel springs or in the trunnion bolts were uncertain at best. Some gaps undoubtedly existed between the forward and aft ends of the shield and the tank walls at the time of launch, which could well have increased as the flight progressed due to the non-uniform growth of the tank.