[L190] The investigation of the STS 51-L incident by the SRM Working Group involved numerous tests at both MSFC and at the Morton Thiokol, Wasatch Division.

A detailed description of each test is a part of the SRM Working Group files. The following is a listing of the tests performed followed by a brief description of each test, the test objectives, and results.

MSFC
Test No.	Title
.
1A	Characterization of SRM joint Seal Materials
1B	O-Ring Defect Analysis
2	Burn (Smoke) Tests
[3 and 7]	[O-Ring Resiliency Investigation]*
3A	Dial Indicator Rebound Resiliency Test
3B	O-ring Resiliency Investigation
3C	Short Term Resiliency Testing of O-rings
3D	Long Term Resiliency Testing of O-rings
3E	Dial Indicator Rebound Resiliency Test of Defect O-rings
4A	Conoco Grease Blowthrough Test
4B	Randolph Putty Blowthrough Test
4D	Sealing vs. Temperature Transparent Putty Behavior
4E	Sealing vs. Temperature of O-ring Static Fixture
4H	Sealing vs. Temperature; O-ring Discrete Increment No. (Piston-cone) Fixture, Part I and Part II
4J	Randolph Putty Blowthrough Evaluation Test
5	Scenario 4B. Ice Effects on Joint Seal
7	O-ring Resiliency Investigation Composition Variations**
8	SRM O-ring Stacking Damage Test
9	Ground Truth Photo Test

Note:

* This item did not appear in the original table above.

** Test 7 does not appear between tests 5 and 8 but after test 3E.

[Chris Gamble, html editor]

 

MTI
Test No.	Title
.
101	Leak Check Port Plug
103	SRB Referee Test
104	Full Scale Joint Cross Section Test Firings
105	Five-inch Motor Tests
107	Dynamic Vacuum Putty Extrusion Tests
108	O-ring Viscoelastic Properties Tests
109	O-ring Static Blowby Test
110	O-ring Dynamic Blowby Test
111	O-ring Resiliency Test
112	SRM Frozen Field joint Test
113A	Hot O-ring Extrusion Test
113B	Cold O-ring Extrusion Test
114	Ice in joint Tests
115	O-ring Twist Test/Demonstration
117	Seal Contamination Tests
119	O-ring Assembly Damage Test

Test No. 1A

Test Report Date: 3/31/86

Title: Characterization of SRM Joint Seal Materials

Objective:

Develop detailed thermal and mechanical characterization data primarily for the fluorocarbon O-ring seal material, as well as selected characterization data for other materials in the joint and adjacent to the joint.

Test Description:

• Thermal analyses performed by thermal gravimetric analysis, differential scanning calorimetry, and thermomechanical analysis methods
• Mechanical testing in both static and dynamic modes
• Development of comparative data for thermal stability, glass transition region, coefficient of thermal expansion (CTE), and mechanical properties

Test Limitations.

Limited to basic material characterization data-cannot simulate joint seal test configurations.

Test Variables:

• Temperature
• O-ring material composition, batch

Results/Conclusions:

The thermal analyses generally indicate an onset of decomposition of the O-ring material at 465°C to 470°C (870°F to 880°F), with decomposition of the other joint materials at progressively lower temperatures. Glass transition of the O-ring material has measured to be -11°C to -18°C by two methods, and the CTE was 98 x 10^-6 in/in F to 116 x 10^-6 in/in F measured at 75°. The thermal data further indicated relatively little variability based on the different O-rings and batches sampled.

Mechanical test data demonstrated a progressive increase in parameters such as tensile and relaxation modulus, as well as hardness with decreasing temperature. This trend was observed with dynamic as well as static modulus measurements. The mechanical properties (like the thermal properties discussed earlier) did not appear to vary significantly with different O-ring/batches. This is substantiated by the batch data comparisons run at the O-ring vendor plant. It would appear that the vendor presumption of equivalence of starting material and raw material compositions is reasonably valid.

The scarfjoints in the flight O-rings appear to fail at a somewhat lower stress (15 - 20 %) than the parent material, based on the five samples tested. The high heat rate testing of the joint areas indicated generally comparable performance to the parent material.

Test No. 1B

Test Report Date: 3/25/86

Title: O-ring Defect Analysis

Objective:

Define the type and severity of defects in flight-type case joint O-rings and assess any effect they may have on physical, thermal and mechanical properties of the O-ring.

[L191] Test Description:

Analysis/test activities specified below were performed.

• Assessment of x-ray films for low density inclusion and high density inclusion defect Circumferential location of defects
• Photographic analysis (light microscope and SEM)
• Mechanical/thermal characterization of defect areas

Test Limitations:

No correlation of defects with flight O-rings in STS 51-L hardware.

Test Variables:

Type and magnitude of defects

Results/Conclusions:

Based on the analysis and testing performed on flight case joint O-ring SN0001782, the observed high density, low density, and surface defects do not appear to significantly alter the physical and mechanical characteristics of the O-ring material. The effects of typical defects on the temperature response (resilience) characteristics of the material are to be reported under MSFC Test 3E. The tensile strength testing on surface scratch areas is inconclusive without additional appropriate test samples.

Test No. 2

Test Report Date: 3/30/86

Title: Test Burn (Smoke) Tests

Objective:

To observe burning characteristics of joint seal materials and obtain information on the potential origin of "black puffs" early in flight.

Test Descriptions:

• Oxy-acetylene welding torch (per parameters below)
• Samples suspended or in ceramic crucible
• In air or N₂

Test Limitations:

• Not SRM combustion flame
• Not SRM tortuous path

Test Variables:

• Materials

  1. Conoco grease

  2. Randolph putty

  3. O-ring

  4. NBR insulation

  5. Liner/inhibitor: record, video and film temperature and time

   

• Burn conditions

  1. Direct flame: hot

  2. Direct flame: fuel rich

  3. Indirect heating: air

  4. Indirect heating: N₂

Results/Conclusions:

The results of the tests are summarized in Table C.1. Some of the joint seal materials (i.e., vulcanized NBR, liner and inhibitor) upon decomposition in an inert atmosphere produced only white or light gray smoke, but upon introduction into an air environment produced large quantities of black smoke.

Tests No. 3 and 7

Title: O-Ring Resiliency Investigation

Introduction:

MSFC Test 3 and MSFC Test 7 were conducted to determine the influence of several factors, and their interactions on the resiliency of SRM O-ring materials. MSFC Test 7 focused specifically on compositional and batch-to-batch variations on O-ring resiliency. Five "subtests" were focused on different issues of the overall resiliency study, and these related investigations were reported individually under MSFC Test 3. The summaries of all tests have been attached to clarify the objectives, test fixtures, limitations, variables, and conclusions specific to each. However, the general objective and approach for Tests 3 and 7 may be summarized as presented below. The key conclusions from the tests are also presented in summary manner.

Objective:

The objective of the MSFC Resiliency Investigation was to determine the effects of temperature, compression, compression duration, compositions and their interactions, and the effect of various defects on this resiliency, or recovery capability of SRM O-ring materials.

Approach:

The approach followed in all experiments may be generalized as follows:

1. An O-ring sample (0.280-inch nominal cross sectional diameter) is contained in a groove of flight specification dimensions.

2. The O-ring sample is compressed.

3. The O-ring and test fixture are cooled to test temperature.

4. Compression is maintained for a specific period of time after the time of initial compression.

5. The load is removed rapidly and the O-ring recovery is monitored.

Conclusion:

1. Initial comparisons of the data generated by MSFC Tests 3A, 3B, 3C, and 3D indicate good agreement among the methods. These data correlate well with those generated by the Aerospace Corp. Further, segment tests conducted under MSFC Test 3B validate the more rapid initial recovery observed in MSFC Test 3C and correlate well with Test 3B circular ring results.

2. A reduction in the O-ring recovery rate is produced after a compression duration of 4 hours or longer.

3. Temperature is the dominant factor in controlling initial O-ring recovery capability. The rapid springback evident in the 66-mil response at 75°F is effectively eliminated at 25°F. The effect of the 72-hour compression duration evident at 73°F is absent at 25°F.

4. Recovery may be accurately modeled as being linear with increasing temperature in the range of 25°F to 73°F. Further, over the first critical second, the slope remains constant for the 40-mil and 66-mil initial compressions.

5. Recovery is linear with increasing compression during the first critical second, and the slopes remain constant during this period.

6. No significant differences were observed in the recovery characteristics of the samples, from the three different batches in tests at 75°F and 25°F.

7. The springback mechanism is unchanged by adding small inclusions to the baseline O-ring material or by adding a scarf joint.

Test No. 3A

Test Report Date: 3/26/86

Title: Dial Indicator Rebound Resiliency Test

Objectives:

To determine the resiliency and compression set of the SRM field joint O-ring material at 22°F, 30°F, and 75°F.

Test Description:

The O-ring rebound is measured and recorded using an LVDT and a multipen strip chart recorder. A 1/2-inch-long 0.282-diameter O-ring was placed in an REM configuration.....

[L193] .....O-ring groove as the test specimen. A drawing of the test fixture is shown in Figure C.1.

Test Limitations:

The test specimen is a straight, ½-inch-long segment of O-ring stock.

End effects may influence the response.

Test Variables:

• Test temperature: 22°F, 30°F, and 75°F
• O-ring compression: 22, 39, and 50 mils

Results/Conclusions:

Test results show that at room temperature an O-ring which has been compressed approximately 0.036 inch when released will spring back approximately 0.020 inch in 0.6 second.

At cold temperatures, springback of the O-ring is reduced. Data indicate that O-ring springback at 30°F is approximately 0.010 inch in 0.6 second and that 22°F, springback is 0.002 to 0.003 inch at 0.6 second.

Test No. 3B

Test Report Date: 4/2/86

Title: O-Ring Resiliency Investigation

Objective:

The objective of MSFC Test 3B was to determine the effects of temperature, compression, compression duration, and interactions of these factors on the resiliency, or recovery capability, of O-ring materials used in the SRM field joints.

Test Description:

A subscale O-ring (0.280-inch nominal cross sectional diameter) is contained within a circular groove machined to field joint specifications and compressed to a specific squeeze. Compression is maintained at the test temperature for a defined period. The compression plate is then withdrawn rapidly, and the O-ring recovery is monitored with an extensometer.

Test Limitations:

The O-rings are tested in a face-seal mode rather than the radial-seal mode of the actual SRM field joint.

The O-rings are compressed from an initial stress-free state; i.e., no stretch or hoop strain is applied.

Test Variables:

Temperature: 10°F, 25°F, 30°F, 40°F, 50°F, 60°F, 75°F

Compression:	0.020	0.035	0.040	0.066
.	(7.1%)	(12.5%)	(14.3%)	(23.6%)

Compression Duration: 2 hours, 12 hours

Results/Conclusions:

• Initial comparison of data generated under MSFC Tests 3A, 3B, and 3C and Aerospace Corporation data indicates good agreement. Sample data are provided in Figure C.2.
• O-ring materials exhibit a significant loss of resiliency between 75°F and 250°F.
• O-ring recovery is reasonably linear with compression (exception of initial nonlinearity. introduced by wall effects, high compression, at 75°F).
• The effects of compression and compression duration are strongly influenced by temperature.
• The influence of compression duration on O-ring resiliency was evident, at 75°F, in the reduction of the recovery rate after 12 hours duration when compared to the responses after 2 hours and also 30 minutes duration. However, at 25°F, the effect of low temperature dominated that of duration and no differences between the 2 hour and 12 hour responses were observed.

Test No. 3C

Test Report Date: 3/25/86

Title: Short Term Resiliency Testing of O-Rings

Objective:

The purpose of these tests was to determine the effects of temperature and compression on O-ring resiliency for short term compression durations. Samples from batch 268535 (25B ft, secondary composition) and batch 267768 were tested and compared.

Test Description:

Uses straight section of O-ring; compression applied over a 4-inch length. See attached Figure C.3. Uses light transmitted between opening of O-ring edge and top of platen to measure/evaluate both initial opening time and relative degree of separation.

Test Limitations:

The test specimen is a straight, 4-inch segment of O-ring stock.

The segment is compressed from an initial stress-free state; i.e., no stretch is applied.

Test Variable:

• Temperature: 10°F to 78°F
• Squeeze: 0 to 24% (unrestrained to metal/metal)

Results/Conclusions:

Short Term Compression Duration (approximately 5 to 10 minutes)

• O-ring springback deflection is proportional to load.
• The greater the compression, the quicker the springback deflection.

Long Term Compression Duration (30 minutes)

• Long duration compression of 30 minutes provides more consistent data than short term compression as observed in secondary O-ring tests.
• Compression regained at time and temperature in our test range is slightly reduced for longer compression time of 30 minutes compared to ~5 to 10 minutes of compression.
• No differences in batch 268535 O-ring versus batch 267768 O-ring material could be detected within testing uncertainty of ± 0. 0015 inch. For direct comparison initial compressions must be the same. Due to loading effects, the initial compression was not always the same. Therefore, only trends could be established.

Test No. 3D	Test Report Date: 4/1/86
	(Interim)

Title: Long Term Resiliency Testing of O-Rings

Objective:

The purpose of these tests was to determine the effects of temperature and compression on secondary O-ring resiliency for long term compression of 30 minutes to 30 days. Due to the long term duration, this report contains the testing accomplished to April 1 and is an interim report that will be updated when the series is completed.

Test Description:

O-ring segment (4 inches in length) contained within field joint dimensions groove is compressed to a specific squeeze. Compression is maintained for extended period of time (30 days). Load is released and O-ring recovery is monitored mechanically and optically.

Test Limitation:

Test specimen is a small, straight segment of O-ring stock. Load release is not at programmed rate.

[L196] Test Variables:

• Compression duration: 1 to 30 days
• Temperature: 10°F to 75°F

Results/Conclusions:

At 75°F, a trend of slower recovery rates was evident for compression durations of 4 hours and longer when compared to the recovery rates after only 30 minutes or 1 hour duration.

At 25°F, the length of compression time (up to 72 hours) does not affect the recovery rate during the critical first second after load release.

As the compression duration was increased beyond 30 minutes, the material required a longer time to regain 90% of compression and this behavior was more pronounced at 25°F than 75°F. Table C.2 and Table C.3 show the percent of compression regained after release of the top plate.

Test No. 3E	Test Report Date: 4/2/86
	(Interim)

Title: Dial Indicator Rebound Resiliency Test of Defect O-Rings

Objective:

To determine the resiliency and compression set of various defect SRM field joint O-ring materials at 22°F and room temperature.

Test Description:

Utilize resilience tester developed for MSFC Test 3A (Figure C.4) to test O-ring resilience in selected defect locations from flight case joint O-rings.

Test Limitations:

Not dynamic test fixture. Does not include effects of O-ring pressure actuation.

Test Variables:

• Temperature
• Defect type
• Compression during test

Results/Conclusions:

The data contained in this report are very similar to results presented in the final report for MSFC Test 3A. It is evident in the data that the springback mechanism is unchanged by adding inclusions to O-rings or by adding a scarf joint.

The variances in data for O-rings with inclusions is small and is within bounds of data obtained for baseline O-rings. It is important to note that there is a large difference between the springback rates at the two temperatures.

Test No. 7

Test Report Date: 4/2/86

Title: O-Ring Resiliency Investigation-Composition Variations

Objective:

The objective of MSFC Test 7 was to determine the effects of different compositions and batch-to-batch variations of the same composition on the resiliency of SRM O-ring materials.

Test Description:

A subscale O-ring (0. 280-inch nominal cross sectional diameter) is contained within a circular groove of flight specification dimensions and compressed to a specific squeeze in an Instron load frame. Compression is maintained at test temperature for a defined period. The compression plate is withdrawn rapidly, and the O-ring recovery is monitored with an extensometer.

Test Limitations:

• Availability of accurate pedigree on composition of batches of O-ring stock from Parker.
• The O-rings are subscale.
• The O-rings are tested in a face-seal mode rather than the radial-seal mode of the actual SRB field joint.
• The O-rings are compressed from an initial stress-free state; i.e., no stretch or hoop strain is applied.

Test Variables:

• Composition: Batch 268535 (25B aft secondary), batch 273043 (25B aft primary), batch 267768 (random)
• Temperature: 25°F, 75°F
• Compression: 0.040 inch-other compressions to be determined after analysis of 40-mil data (14.317 squeeze.)
• Compression duration: 2 hours

Results/Conclusions:

Seven tests were conducted on O-rings manufactured from each of the three batches at 75°F and 25°F with an initial compression of 40 mils. The compression duration for all tests was 2 hours. Results of the individual tests grouped by test conditions are presented in Appendix C of the Test Report. The averages of these tests with scatter bars, one standard deviation in width, are presented for each test condition in Figures 30-32 of the Test Report.

It is apparent from examination that the populations at each condition and over each time interval are in very close agreement. Results for the three batches at 25°F are presented in Figures C.5, C.6, C.7. Upon close comparison, the following qualitative evaluation may be made:

1. At 75°F over the first 1 second, the recovery of batch 267768 material (hereafter as P) greater but within one standard deviation of batch 268535 (hereafter S). S is approximately the same amount greater than the response of batch 273043 (hereafter H).

2. At 75°F over the first 12 seconds, P equals S and H is again low, but within 1 sigma.

3. At 250°F over the first 1 second, P equals H and S is slightly low, within 1 sigma.

4. At 250°F over the first 12 seconds, S equals H and P is more than one standard deviation above these.

For an accurate assessment of the equivalence of these populations, a statistical analysis must be performed. At present, a test analysis for comparisons of the means is underway.

Test No. 4A

Test Report Date: 3/15/86

Title: Conoco Grease Blowthrough Test

Objective:

Determine the cold gas pressure at which rupture or blowthrough occurs in the Conoco HD-2 grease as a function of......

[L199]....thickness and temperature. This test was to define the basic correlations of grease pressurization with thickness and temperature for a given sample cross section.

Test Description:

Test proceeds by forming layer of grease on ledge of bottom of test fixture. Next the fixture is assembled and a pressure sufficient to rupture or blow through the grease in less than 5 seconds is "dumped" on the inner face of the sample (Figure C.8).

Test Limitations:

Flight hardware grease application geometry and joint rotation cannot be emulated for grease samples in this fixture.

Test Variables:

• Grease thickness
• Experimental temperature

Results/Conclusion:

The Conoco grease appears to have sufficiently low viscosity not to be significantly affected by temperature and layup thickness in the comparative pressurization tests. This observation is drawn only for the temperature and thickness ranges used in the test. Based on the relative ease of blowthrough at very low test pressure, the grease would not be expected to significantly affect the case joint pressurization. No further testing is recommended in this test configuration.

Temperature Degrees F	Blowthrough Pressure, psi	Thickness Inches	Time sec
.
60	5	0.061	1.8
10	15	0.061	1.9
60	5	0.100	1.0
10	15	0.100	1.7

Test No. 4B

Test Report Date: 3/15/86

Title: Randolph Putty Blowthrough Test

Objective:

Determine the cold gas pressure at which rupture or blowthrough occurs in the Randolph putty as a function of thickness, temperature and moisture. This test series was to determine the basic response of putty to pressurization as a function of temperature and moisture for a controlled sample cross section.

Test Description:

Flat plate test fixture with milled out section was used for testing (Figure C.9). Putty was installed to correct thickness using shims. Fixture was assembled and pressurized at test temperature.

Test Limitations:

Flight hardware putty layup pattern and joint rotation cannot be simulated. This is a comparative type of test only to define putty blow through characteristics.

Test Variables:

• Temperature
• Putty layup thickness

Results/Conclusions:

The temperature/thickness effects on blowthrough pressure are summarized in Table C.4:

A number of successive runs had to be performed to determine the blowthrough pressure listed in Table C.4. Typically, a lower pressure was utilized initially and held for 2 minutes or until blowthrough occurred. If the putty held that pressure, it was increased and held an additional 2 minutes to determine the threshold pressure.

The results of the moisture conditioned pressure tests are summarized in Table C.5.

*RH = Relative humidity

The 0% conditioned putty held 1000 psi test pressure at 20°F for the 2-minute period and no failure time was established. It should be noted that the putty under these conditions was extremely stiff and dry, and did not represent a reasonable flight hardware application condition. It was included in the matrix to bracket the entire range of pressure response.

The test results confirmed the expected response of the Randolph putty. In general, blowthrough pressure increases with decreasing temperature, and decreases with increasing putty test thickness. Additionally, increasing amounts of moisture in the putty decreases the blowthrough pressure significantly at both test temperatures. This would be expected from the observed reduction in viscosity of the putty with increasing moisture content.

Test No. 4D

Test Report Date: 4/13/86

Title: Sealing vs. Temperature Transparent Putty Behavior

Objective:

Observe possible putty movement and other physical parameters using a transparent test fixture.

Test Description:

Observe putty movement, blowthrough or other motion that occurs during leak check and pressurization. Fixture is designed to allow video tape of putty and O-ring movement when pressurized (Figure C.10).

Test Limitations:

Complete visualization of test fixture not possible at high pressure (700-1000 psi) due to limitations on plastic parts.

Test Variables:

• Putty gap size

Results/Conclusions:

Three runs were performed on the transparent putty fixture at each of the two gap openings, 0.020 and 0.080 inch, using KSC conditioned putty (12 hours/80% relative humidity/80-85°F). The putty movement on pressurization was monitored by video, and in each case, putty uniformly extruded to the primary O-ring. The video tape data indicate within test fixture that the putty does begin to hold pressure off the primary O-ring during the 1-second time frame. This is consistent with MTI results on their putty fixture. Also, the 0.020-inch putty gap repeatedly held pressure off longer than the 0.080-inch gap.

Test No. 4E

Test Report Date: 3/14/86

Title: Sealing vs. Temperature of O-Ring Static Fixture

Objective:

The purpose of this test series was to determine temperature effects on the ability of SRM flight-type O-rings to effectively seal when pressurized.

[L202] Test Description:

See Figure C.11. O-rings and fixture are chilled down to specified temperature. O-rings are in a static condition when 1000 psig pressurization pulse, with a 10,000 psi/second ramp rate, is applied.

Test Limitations:

• Test fixture is subscale
• O-rings are maintained in a static condition
• Cold gas is used as pressurization force

Test Variables:

• O-ring percent squeeze
• Temperature

Results/Conclusions:

Based on analysis of accumulated test data it was shown that SRM flight-type O-rings tested under static condition will effectively seal at selected temperatures from ambient down to 10°F provided a positive squeeze is maintained. This conclusion is supported by the Data Plots, Static Fixture Test Matrix and Test Results Summary in the Test Report.

Test No. 4H

Test Report Date: 4/15/86

Title: Sealing vs. Temperature; O-Ring Discrete Increment No. 1 (Piston-Cone) Fixture, Part I and Part II

Objective

The purpose of this test series was to determine capability of uniformly loaded flight-type O-rings to seal when gap is varied in relationship to pressure, temperature and time.

Test Description:

In Part I, initial O-ring gap is set to specified amount. Fixture is then cooled to specified temperature. O-ring gap can be changed to a new setting just prior to pressurization. (See Figure C.12.) This test configuration simulates a delay in pressurization of the O-ring.

In Part II, dynamic testing was performed by opening the O-ring gap per specified gap opening curves. Maximum and minimum squeeze conditions were investigated. This test configuration simulates pressurization of the O-ring without a delay.

Test Limitations:

• Subscale
• Cold gas
• Both O-rings subjected to identical gap openings
• Circumferential varying gap not simulated

Test Variables:

• Temperature
• Initial gap
• Rate of gap opening
• Pressurization rate

Results/Conclusions:

Part I

Based on analysis of the accumulated test data, it was shown that at 25°F, at both maximum and minimum squeeze conditions tested leakage past the O-rings will occur at 0.250 second and 0.500 second into the SRM start transient when simulating a delayed pressure actuation of the O-rings. At the maximum squeeze conditions tested (0.004-inch gap), sealing is effective from ambient temperature down to 53°F, but becomes marginal at 50°F with a pressure delay of 0.500 second. Tests run at 70°F, with both minimum and maximum squeeze conditions, show no leakage past the O-rings.

Part II

Under dynamic testing conditions it was shown that with a maximum initial gap (0.020-inch), rounding, and pressure actuation, both primary and secondary O-rings demonstrated the capability to seal from ambient to 25°F. With a 0.010-inch initial gap, no rounding, and pressure actuation, the primary O-ring sealed with slight blow-by at 25°F. With a minimum initial gap (0.004-inch), no rounding, and pressure actuation, both primary and secondary O-rings demonstrated the capability to seal at temperatures from ambient down to 55°F.

Test No.: 4J

Test Report Date: 3/28/86

Title: Randolph Putty Blowthrough Evaluation Test

Objective:

Investigate the putty blowthrough characteristics across a gap with a putty buildup on the pressure side. This test series was to determine the basic response of putty, to pressurization as a function of temperature and thickness.

Test Description:

See Figure C.13.

Test Limitations:

• Not in same putty layup or configuration as in SRM joint.
• Pressure rise rate not as in SRM joint.

Test Variables:

• Gap width 0.010, 0.070, 0.200 inch.
• Temperature room temperature and 25°F
• Pressure 760 and 940 psi

Results/Conclusions:

In general, blowthrough pressure increases with decreasing temperature and decreases with increasing putty test thickness, and blowthrough occurs faster at 940 psi than at 760 psi.

The test results also show that the Randolph putty does exhibit some self-sealing capability demonstrated by audible "popping" and a stabilization or increase in pressure within the test fixture. The self-sealing capability is usually of short duration with a "pop" quickly followed by more "pops." This self-sealing capability existed at the 0.010-inch thickness at both 25°F and 60°F. At the 0.070-inch thickness the self-sealing capability existed at the 25°F, and to a much less extent it was evident at the 0.070-inch thickness at 60°F and the 0.200-inch thickness at 25°F.

The test results also show that with continuous pressurization a blowthrough may occur in the putty, and the self-sealing property of the putty will allow the Putty to reseal and the pressure to increase. This increased pressure was not shown when the pressure was cut off immediately after reaching the desired value ("closed" system) because there was no continuous flow possible.

Based on these observations, and the temperatures and thicknesses tested, it is reasonable to expect the putty to hold several hundred to 1000 pounds of pressure for several seconds; however, these results cannot be used to quantify this effect.

Test No. 5

Test Report Date: 3/29/86

Title: Scenario 4B. Ice Effects on joint Seal

Objective:

The objective of this series of testing is to assess the effect of ice formation in the gap between the tank and inner leg of the clevis on the secondary O-ring.

Test Description:

Figure C.14 is a sketch of the ice effects test fixture. Ice formation and O-ring movement if any, is recorded by video tape through plexiglass from the front side of the O-ring grooves. Gap between clevis and tang is filled with water up to just beneath secondary O-ring. Fixture is than cooled below freezing, and ice formation and O-ring movement are video taped.

Test Limitations:

Although the cross sectional volume of the annular space beneath the secondary O-ring in the fixture is very close to that in the SRM joint, the gap beneath the O-ring is much narrower

[L206] ....over a longer distance than in the joint, thus slightly enhancing the effect of water/ice.

Test Variables:

• Presence or absence of grease
• Proximity of water column to secondary O-ring

Results/Conclusions:

Based on the test results, water does not have to be in contact with the secondary O-ring prior to freezing in order to unseat the O-ring. In fact, significant hydraulic pressures greater than leak check pressures that unseat the O-ring may be generated upon freezing, and these pressures and/or mechanical blockage can prevent the O-ring from reseating until the ice has melted.

Test No. 8

Test Report Date: 3/31/86

Title: SRM O-Ring Stacking Damage Test

Objective:

Assess the potential for damaging the field joint O-rings during SRM stacking by simulating tang and clevis interference using a section of a flight joint.

Test Description:

A section of a flight aft center segment tang and its associated aft segment clevis are secured in rigid fixtures in a tensile test machine (Figure C.15). The clearances and interferences of mating surfaces are controlled for each test. The tang and clevis sections are mated and any effects on O-rings and metal surfaces are recorded.

Test Limitations:

May not be able to detect significant changes in O-ring temperature during test (if they occur).

Test Variables:

• Amount of clearance or interference between mating surfaces.
• Initial tang position (vertical or offset by springs).
• O-ring tension; cross sectional diameter.
• Line contact or scissor contact between tang and clevis.
• Joints; repairs.

Results/Conclusions:

In summary, initial O-ring tension (or lack of) significantly affects the probability of elongation and rolling of the O-rings during assembly. Elongation increases probability of damaging O-rings due to bulging, and roll may help drag debris across sealing surfaces and into grooves. Rolling of primary O-ring at ignition may move debris into a position between O-ring and sealing surface, promoting a leak path.

A test run without grease (No. 15) resulted in significant 0-ring damage (3/16 x 5/12 inch bite) and emphasizes the importance of liberal lubrication at assembly.

Mating metal surface interference and tang angle do not appear to affect ability of O-rings to seat without damage; however, increased interference increases the probability of generating metal debris as a result of tang or clevis damage.

Test No. 9

Test Report Date: 3/31/86

Title: Ground Truth Photo Test

Objective:

The objective of this investigation was to determine if the closeout photographs showed evidence of a damaged or defective O-ring or the presence of contamination in the joint or on the O-rings.

Test Description:

A 3-foot-long curved section of the inner clevis has been fabricated on nominal dimensions. Photographs of O-rings installed in this fixture are to be taken using the identical camera, lens, perspective, etc., as the closeout photos. O-rings with known artifacts will be used. A matrix of bare, greased O-rings, bare and greased grooves, scarf joints at various orientations, etc., will be photographed.

Test Limitations:

The closeout photographs document the full periphery of the clevis and O-rings, but the photography was lacking in quality such that only 1 to 1.5 feet of clevis and O-ring could he analyzed, primarily because of lack of focus. Eleven slightly overlapping photographs were taken, allowing about 11 to 17 feet of the total joint periphery to be photographically inspected. The condition of the remaining 50% to 70% was unknown. While not a limitation of the test, it certainly leaves a vast unexplored area.

Test Variables:

Any O-ring artifact - scarf joint, patch, repair, surface irregularity, etc.

Results/Conclusions:

1. The dark space that thickens as noted on closeout photograph Figure 8 (Test Report) was a shadow of grease on the edge of the land above the secondary O-ring, not a defect or a damaged area of the secondary O-ring.

2. The small scale structures could be easily reproduced by a surgically gloved hand applying grease to the clevis. The presence of these structures does not reinforce the hypothesis of an O-ring defect or damage in that area.

3. The circular artifact as noted in the closeout photograph of Figure 5 (Test Report) is a grease structure and not a contaminant.

4. No defects, damaged O-rings, or contaminants, were apparent from the closeout photographs.

Test No. MTI 101

Test Report Date: 2/27/86

Title: Leak Check Port Plug

Objective:

Characterize potential Leak Check Port plug failure modes including: assess if assembly of Port Plug can damage the O-ring; determine if the torque on an assembled port plug relaxes when the fixture is cooled to a lower temperature; determine the effect of various torques on the ability of plug O-ring to seal; and determine the leakage rate of a port plug without an O-ring.

Test Description:

A series of 10 tests were conducted using the fixture shown Figure C.16. All but one of the tests were conducted at ambient conditions. The test activities included the following:

Plug with new O-ring torqued to 60 in-lbs. checked for leak at 200 psi and 1000 psi.

Plug without O-ring torqued to 60 in-lbs at ambient temperature then cooled and plug checked for rotation at retorque.

Five plugs without O-rings torqued to 60 in-lbs and checked for leaks at 200 psi and 1000 psi and the leak rate measured.

Plugs with used O-rings torqued to 5 in-lbs and checked for leaks at 200 psi and 1000 psi.

Test Limitation:

All tests were conducted with cold gas rather than hot particle laden, motor exhaust.

Test Variables:

• Plugs and O-rings
• Torque
• Test temperature

[L209] Results/Conclusions:

No damage of port plug O-ring was detected due to assembly and limited use of plug. Cooling of the torqued port plug assembly without an O-ring did not relax the initial applied torque. Port plugs having an O-ring either new or used will not leak when pressurized to 1000 psi, if the O-ring is visibly seated. Successful seating of O-ring can be achieved by finger-tightening the port plug. Port plugs without an O-ring leaked at varying rates when fully torqued to 60 in-lbs. These leak rates were:

Comparison of protest prediction with these measured leak rates showed the following, which are within 26 percent of the pretest predictions.

Test No. MTI 103	Test Report Date: 4/03/86
	5/05/86
	(ECD)

Title: SRB Referee Test

Objectives:

To obtain structural data on the effects of pressurization upon the SRB motor case field joint including:

• Tang and clevis relative rotations and displacements
• The opening and/or closing of the gap at the O-rings
• Stresses, strains, and load paths in the tang and clevis

Test Description:

The test fixture (Figure C.17) is a stack of two full size SRB flight motor case segments (a lightweight attach segment and a lightweight cylinder segment) capped on each end by a proof test dome.

The tang, clevis, and membrane of the segments are instrumented with strain gages and deflection gages.

The stack is pressurized by hydropressure to approximately 1004 Psig.

Test Variable:

The tests are being performed in three sequences:

Sequence 1

The joint is instrumented to determine the following:

• Relative rotations and deflections between the tang and clevis measured at Pin holes (pins removed in 6 locations) and through new O-ring leak test ports (8 locations).
• Circumferential growth of each case at the joint (12 locations).
• Axial growth across the joint (8 locations).
• Effect of the following shim thicknesses (Pin retainer clip):
• 0.034-inch (flight shim)
• 0.018-inch
• 0.040-inch
• 0.045-inch
• 0.050-inch

Each shim configuration was run twice-first taking deflection through the leak test port, then monitoring the pressure between the O-rings.

Sequence 2

The joint is instrumented the same as Sequence 1 except as follows:

• Additional holes were drilled to measure the gap action at the alignment slots (3 locations at each slot).
• No axial deflection gages.
• Only pins at the alignment slots removed.
• Tests will determine the effect of removing the primary O-rings.
  • Test 1 - 0.034-inch shim

    Both O-rings installed

  • Test 2 - 0.050-inch shim

    Both O-rings installed

  • Test 3 - 0.034-inch shim

    Primary O-ring removed

  • Test 4 - 0.050-inch shim

    Primary O-ring removed

Sequence 3

Joint is instrumented to determine the stresses and strains in the tang and clevis and the load path through the joint.

Results/Conclusion:

The key results from Sequence 1 are summarized below:

It can be observed that for the standard (flight) configuration, the seal gap was opened an average of 0.033 inch which is in the mid range of the previous measurements (i.e. 0.028 inch to 0.042 inch at 1004 psig). It can also be observed that the addition of thicker shims did in fact reduce the relative gap displacement; however, the reduction was only slight.

The key results from Sequence 2 are summarized below:

[L211] Other results from Sequence 2 are:

1. The seal gap opening at the tang locator slots is less than at any other location measured and is within analytical predicted values.

2. With the primary O-ring removed the seal gap opening increases. The increases were 0.011 for the 0.034 shims and 0.008 for the 0.050 shims. Although removing the primary O-ring increases the seal gap opening, the secondary, O-ring seal retains pressure.

3. During test article assembly there was an insignificant load measured in the joint. The girth gage deflections recorded were less than 0.00001 inch.

4. The testing for Sequence 3 has not been accomplished yet. Plans are being implemented to install an External Tank attach ring on this (or follow-on) test sequence. The objective is to quantify the reduction in gap opening of the aft joint due to the added stiffness afforded by the ring.

Test No. MTI 104	Test Report Date: 4/03/85
	5/25/86
	(ECD)

Title: Full Scale joint Cross Section Test Firings

Objectives:

• The objectives of this test activity are to:
• Evaluate field joint response to induced hot gas leak.
• Evaluate potential causes for field joint leak.

Test Description:

The 70-pound charge motor joint test section and nozzle configuration used is shown in Figure C.18. Not shown is the 70-pound motor pressure chamber which contains the end burning grain designs. Both the insulation and joint thicknesses are the same as the full scale design. The configuration was designed to have the same heat flow characteristics as the full scale design when subjected to a single point O-ring failure. The design incorporates all the full scale joint features except diameter. These features are full scale pins, leak check port, Viton O-rings, pin spacers and pin retention band. The insulation material is silicafilled NBR instead of asbestos-filled NBR to avoid the health hazards associated with asbestos. The insulation gap is designed to nominal, full scale dimensions. Randolph putty and Conoco calcium HD-2 grease is used in all tests.

Results/Conclusions:

The tests that have been accomplished to date are shown in the attached Table C.6. Also shown are the conditions for each test. The observations resulting from these tests are:

• The diametrical gap between the inside clevis leg and tang surface has a significant influence on the time required to burn through, after leakage through the O-rings is initiated.
• Time to bum through with a 0. 040-inch gap was 11. 7 seconds for Test 3, and 28-30 seconds (estimated) for Test 4.
• Only minor leakage with no O-ring or metal damage with a 0.0035-inch diametrical gap (Test. 5A).
• Only minor leakage with significant O-ring damage but no metal (Test 5B).
• Generation of black smoke requires an intermediate size leak.
  • No black smoke was evident with diametrical gaps up to 0.02 inch and smaller.
  • 0.250 x 0.040 V-shaped groove produced 3 to 4 seconds of black smoke (Tests 3 and 4).
  • 5.6 inches of missing O-rings produced immediate flame (Test 1).
• Appearance of black smoke indicates the O-ring is being thermally damaged. Termination of smoke at an early time period most likely would require closing diametrical gap (Tests 5A and 5B).
• Closing of the diametrical gap is the most probable reason that the joint could leak at ignition and not sustain a bum through until 58 seconds (Tests 3, 4, 5A, and 5B).
• All tests which used putty conditioned at 80 percent RH and 75°F for 10 hours and tested at either 30°F (7A, 7B, and 7C) or 70°F (8A) showed the putty was extruded to the primary O-ring groove and through O-ring groove and through the 0.250 x 0.040 inch V-shaped groove in the primary O-ring. All joints sealed with no indication of leakage.
• Putty conditioned at 30 percent RH and 80°F for 18 hours also extruded to the primary O-ring groove and through the primary O-ring fault when tested at 70°F. As in the 7A, 7B, 7C, and 8A tests, the putty sealed the fault in the primary O-ring and no gas leakage resulted.
• The only test which showed no putty movement during static test was 8B. This test used putty conditioned to 30 percent RH and 80°F for 18 hours and the test section was conditioned to 30°F prior to static test. Sealing of the joint was achieved by the putty only with no evidence of gas flow to the faulted O-rings.
• There was no leakage of hot gas evident for test 9D which had 0.015 inch of negative O-ring squeeze, a 1.0-inch putty flow path, a 0.040 inch diametrical gap and putty conditioned to 80 percent RH at 75°F for 10 hours. The motor was conditioned to 30°F and fired for 20 seconds. Review of pressure data and post fired hardware showed the primary O-ring seated with no evidence of blow-by at ignition. The primary O-ring in line with the putty fault had a flat spot eroded to a maximum depth of 0. 0 11 inch which phase out on either side over a total circumference of 3.75-inch. The flat was eroded on the upstream side of the O-ring and appears to have resulted during sealing of the O-ring from flow into the forward portion of the O-ring groove. Review of the high speed motion picture films showed no leakage at ignition or during steady state bum. Some smoke was evident coming from the joint very late in tail off after chamber pressure has decreased to a level which allows the O-ring to unseat.
• Testing with a 0.015 inch space between the O-ring and the scaling surface resulted in a seal. This implies that the most critical feature of the seal performance is pressure actuation capability and not gap opening (joint rotation). When pressure can get to the "forward face" of the O-ring, the O-ring will pressure actuate into a sealing position.

Test No. MTI 105	Test Report Date: 3/13/86
	5/21/86
	(ECD)

Title: Five-inch Motor Tests

Objectives:

The objectives of this test activity were to:

• Evaluate field joint response to induced hot gas leak.
• Evaluate potential causes for field joint leak.
• Guide the more complex full scale joint cross section testing.

Test Description:

The test motor used for these tests is shown in Figure C.19. It was tested in either a center perforated grain (CP) configuration which provided approximately 3 seconds of burn time or an end burner grain (EB) configuration which provided approximately 22 seconds of bum time.

Test Limitations:

The test configuration was subscale and did not represent the configuration of the SRM joint.

[L218] Test Variables:

• Extrusion gap
• Putty and O-ring defects
• Temperature
• O-ring squeeze

Results/Conclusions:

The tests that have been accomplished to date are shown in the attached Table C.7. Also shown are the conditional for each test. The observations resulting from these tests are:

• The time required to obtain a burn through after a gas leak has developed is primarily a function of the diametrical gap between metal surfaces.
  • Flame within 0.40 second of ignition with 0.040 to 0. 030 inch diametrical gap, 1/8-inch section of O-ring missing.
  • Only smoke flow for full motor operation, 22 seconds, with 0.001 to 0.004 inch diametral gap with 1/8-inch section of O-ring missing.
• An O-ring with zero or negative squeeze up to 0.0004-inch will seal.
• The putty as configured in this test fixture could act as an effective seal.

Test No. MTI 107	Test Report Date: 4/03/86
	4/21/86
	(ECD)

Title: Dynamic Vacuum Putty Extrusion Tests

Objective:

Determine putty response to SRM start pressures at conditions representative of STS 51-L launch and assess O-ring pressurization during the motor ignition transient.

Test Description:

The dynamic vacuum putty extrusion test fixture consists of a full scale cross section representation of the SRM clevis inner leg and tang. A 10-inch-diameter model (Figure C.20) is used to complete the tests. Pressures and temperatures were recorded during assembly of the tang-to-clevis and during the simulated SRM chamber pressurization. Photos were made of all disassemblies to assess the putty condition and location. The test matrix is shown in Figure C.21.

Test Limitations:

The tests were representative of a full scale SRM joint except for three features: (1) O-ring inside diameter (9.33 inches), (2) volume between primary and secondary O-rings, and (3) volume between primary O-ring and vacuum putty.

Test Variables:

1) Tang-to-clevis end gap

2) Putty exposure to humidity

3) Simulated launch temperature

4) Putty back-blowhole condition

Results/Conclusions:

The pressures recorded between the putty and primary O-ring during normal stacking produced pressures consistently up to 46 psig. These assembly pressures created back-blowholes in the putty which resulted in O-ring pressurization times to 20 psig ranging from 0.159 to 0.572 second, with the earlier pressurization occurring at the lower temperatures. After a back-blowhole was formed during the fixture assembly, the putty did tend to reseal he damaged area. This was verified by the pressure between the putty and primary O-ring reestablishing subsequent to a blowhole during the 200 psig O-ring leak check prior to simulated SRM start. Also, some disassembly photos showed back-blowholes with no red chalk which was introduced in the inlet pressurization line during SRM simulated start.

Some test fixture assemblies were done with the volume between the putty and primary O-ring vented to prevent back blowholes. The ties required to pressurize the primary O-ring to 20 psig was highly influenced by temperatures, putty gap width, and putty humidity (water content.) Pressurization times to 20 psig ranged from 0.20 second to 20.72 seconds with the higher duration occurring with as packaged putty at 20°F in a minimum gap. Disassembly of the tang and clevis fixture showed some putty to the primary O-ring on approximately 80 percent of tests.

These tests imply that the putty may provide seal of the SRM joint during SRM start transient, which causes delayed O-ring pressurization.

Test No. MTI 108

Test Report Date: TBD

Title: O-Ring Viscoelastic Properties Tests

Objective:

To characterize the viscoelastic response of the SRM case joint fluorocarbon O-ring material.

Approach:

• Determine glass transition temperature
• Measure dynamic mechanical properties as a function of temperature