Investigation confirms vertical debris events during STS-124 launch

by Chris Bergin

An expansive investigation has been carried out on imagery footage that showed numerous debris events near shuttle Discovery during her launch on STS-124.

Thousands of bricks were blown out of the flame trench, with some of the debris confirmed as rising above the zero level of the Mobile Launch Platform – which has resulted in a major investigation to ensure the vehicle remains safe from potential impacts.

**Click here for sample of L2 menu and content**

**STS-124 L2 Sub Section (over 3400mb) – Continuing through post flight and post mission IFA/Mission reviews. 13 Post Flight IFA Review Presentations now available.

**STS-125/LON-400 L2 Sub Section Special already over 1650mb in size – special section Now Live, with over 40 superb mission presentations. STS-125 Flight Plan (newly created) and Flight Data Files now uploaded. Mission Overview in CGI videos building up – starting with payload overviews. STS-400 Flight Plan. STS-400 EVA and mission overview Presentations (14). Daily extensive processing latest.

**STS-124 Post Flight – STS-119 Processing Latest** — **STS-124 News Articles**

**LIVE updates: Endeavour Dual LON and STS-126 Processing**

**LIVE updates on Atlantis STS-125 Processing Flow** — **STS-125 News Articles**


Cause Of Flame Trench Damage:




The event occurred as Discovery’s Solid Rocket Boosters (SRBs) ignited, with the exhaust hitting what is now known to have been a weakened flame trench wall. A large section of the Apollo-era bricks were sent flying out of the trench in several directions. No debris hit the vehicle during this early phase of ascent.

‘The post launch inspection of the Mobile Launch Platform (MLP-3), Pad A FSS (Fixed Service Structure), and Pad A apron was conducted on May 31, from Launch +2.23H to 5.23 H (1925 to 2225 EDT). Excessive SRB flame trench wall debris was noted during this inspection,’ noted one of several expansive presentations – available to download on L2 – on the events.

‘East Brick Wall Damage: Estimated Total Bricks On East Wall around 22,000. Estimated Lost Bricks around 3,540 (16 percent Loss of East Wall). Structural Assessment: Failed wall section nearest flame deflector. No observed dove tail clips nearest break area. No epoxy nearest joint.

‘All upstream panel bricks (48) fractured at joint. 10 discolored bricks indicates older damage.’

An investigation into what caused the flame trench wall to shed its bricks is non-specific, classed instead as a wall system deterioration over time, along with weaknesses in previously repaired areas of the trench, and a natural bowing of the wall.

‘Comparison of Pre and Post STS-124 East Wall: Missing sections of repaired (since Pad activation) intact wall for post STS-124. Potential causes: SRB Plume induced loads on repaired section. Bowing of wall most probable cause,’ added the presentation, before elaborating on test results for a more expansive result.

‘Destructive Testing: Core samples,18 two-inch core samples taken from east/west flame trench walls. 11/18 failed, 7 resulted in tensile strength ranging from 54 PSI to 114 PSI. Lab testing, Brick flexure and compressive testing showed minimal reduction in strength from original specifications.

‘NDE (Non Destructive Evaluation): 6 NDE techniques analyzed, laser shearography most promising to date. Tap testing initially used to qualitatively determine areas of hollow or solid ringing. Laser shearography is labor intensive and provides no additional benefit over audible tap testing.

‘Observations/Results: Mortar/concrete joint strength reduced by carbonation and acid deposition. Epoxy adhesive strength still exists in localized areas – relative amount and specific locations indeterminate. Brick destructive testing indicates very little brick strength reduction in 45 year service life.

‘Acidic solutions have adverse effects on Portland cement concrete, depending on the type and concentration of acid. Hydrochloric acid, a byproduct of the SRBs, has shown pH readings below 2 during post launch field sample. Hydrochloric acid is commonly used to clean mortar from bricks and surfaces.

‘Based on current data and Fault Tree closures, the most probable root cause is ‘operational degradation’, as designed condition has demonstrated the ability to meet the load and environment conditions.

‘The Anchor Plate system was severely corroded in the area of failure initiation. The concrete wall/mortar interface indicated exposure to atmospheric carbon dioxide (damp conditions); this ‘carbonation’ of the cement paste resulted in a loss of strength between the mortar and underlying wall.

‘Epoxy either has lost its elastomer over time or it was not present as specified. Visual inspection was unable to determine the health of the wall system behind the brick. Unable to detect anchor / epoxy deterioration. Unable to detect slight bowing of wall structure.

Based on these findings, engineers noted a significant reduction in strength of Anchor Plates due to corrosion, reduction in mortar to concrete integrity due to carbonation and deposition, and degradation of epoxy along with existing areas of non-contact, as all contributing factors towards the bricks liberating from the wall.

Due to the aforementioned factors, when Discovery started to lift off the pad, the loads present caused the failure at the control joint and subsequent cascading brick loss.

‘Wall Failure Conditions/Failure Mechanism: Wall section with minimum tensile load capacity. Loss of anchors and epoxy bonding (minimum strength). Brick anchors connections provides for some load capacity.

‘Indication of load imposed by downstream wall on upstream wall. Inter wall pressure developed from gas migration between brick and wall and/or excessive vibration. Control joint loaded to failure prior to wall separation.’

Analysis of Pad Debris:

Thanks to post Return To Flight implementation of several forms of debris imaging, engineers have been able to review and compare several data points of where debris ejected towards during launch.

These include Avian and Southern Doppler radar data – the later installed on to a naval vessel located out to sea. These data points were compared to the launch of a few previous missions, in order to find any items of interest.

‘Avian Radar Debris Findings: The reflectivity of the plume is different from 124 and prior missions. A ‘jet’ of particles is observed out of the north flame trench after SRB ignition.

‘After 3 scans, or just over 3 seconds, the appearance of the northern plume returns to a more normal signature. The debris in the plume does not tend to follow the wind displacement of the exhaust cloud.’

As observed on an infrared video of the launch – which pointed to several items of debris rising above the pad and appearing to follow Discovery – the radar also noted debris tracks which appeared to be travelling in a vertical path.

‘Indications of discrete, near vertical, debris tracks can be seen in one pass of the vertical radar. These cannot be rendered at heights lower than the 295’ level of the FSS due to sector blanking for that sensor.

‘This signature is consistent with infrared camera indications of particles reaching high apparent altitudes around the pad near lift-off.

‘Further indications can be seen in both plume directions from the horizontal radar. Several unique particle trajectories are seen from this sensor. One ends abruptly at or near the north perimeter fence. These events precede the northern expansion of the SRB exhaust cloud, and follow arcing ballistic tracks north of the pad.’

The Southern Doppler radar results appear to show debris particles suspended within the SRB plume.

‘All observations occur from MLP level 0 (95’ level) or above due to earth curvature obscuring lower areas of the flame trench to this radar,’ noted the presentation relating to the radar ship’s findings. ‘STS-124 lift-off Doppler signature was compared to 120, 123 at lift-off.

‘Principle differences are a very prominent cloud of debris particles seen at lower initial velocities. These particles span the timeframe of 0.5368 to 1.2317 MET. These particles span a velocity range of 200 to 1000 fps (feet per second). These contacts are bright, (-14.7 dbsm), move out of the north trench, and are the brick/trench debris.

‘A series of discrete, high velocity – initially supersonic – plumes are seen shortly after SRB ignition on each flight. These are small particles, and likely indicative of expected behavior. (However), these plumes are more prominent on STS-124 than other missions.

‘The particles span the timeframe of 0.5907 to 0.7045 MET, and 2.0044 to 2.31 MET. The particles span a velocity of 500 to 3500 fps. These contacts move out of the north trench, and are small trench debris preceding the plume front.

‘The overall SRB plume signature is brighter, for a longer period, than comparison missions due to suspended debris material within the plume.’

The presentation noted that the available resources are not designed to track such events as the flame trench shedding debris, and normally are used to track the vehicle during first stage – in the event of a debris liberation from the External Tank.

However, once again, debris rising vertically – though away from the vehicle – is noted in the summary.

‘Neither the Avian or Southern Debris/Doppler radar are optimized for lift-off debris, but indicate unique debris signatures when comparing the STS-124 launch to prior missions.

‘The vast majority of the contacts are observed to move from the north flame trench away from the vehicle.

‘Some particles reach heights of at least the 295’ level of FSS. The debris signature is maintained throughout tower clear.

‘One debris track, moving at slower velocity, does not follow the trend of the other contacts, and moves in the west to northwest direction. Source material is reflective, but unknown.’

Infrared Video Footage Latest:


The most telling footage of the debris events – that somehow managed to rise vertically out of the pad – remains to be the infrared footage gained by a camera on the Vertical Assembly Building (VAB) roof – video available in full on L2.

Noted as ‘Experimental Infrared Imagery’ – the imaging systems are being tested for applicability to shuttle launches and landings.

‘FLIR Ranger III XR+ MWIR (the expanded name for the infrared camera) imagery showed debris (first observed around T+1.1s) shooting up and away from the pad,’ noted another presentation, specific to the debris events.

‘The camera quickly became completely overexposed as the SRB plumes emerged from beneath the pad.

‘Close inspection of the imagery also shows several pieces of debris shooting horizontally across the pad and another piece emerging from behind the facility and rising to an apparent height greater than the water tower height.’

A large focus was placed on what the imagery was capturing, with flame trench bricks (when specified as such) a serious concern, due to the proximity the debris flew upwards next to Discovery.

To aid the investigation, engineers reviewed similar footage from recent launches and found there to be similar – though very isolated – events of debris rising near the vehicle.

However, these debris sources – observed on launches STS-120, STS-122 and STS-123 are believed to be a combination of throat plugs being ejected from the SRBs, dark ice falling off the ET, and the aftermath of the water baggies that surround the SRBs pre-launch.

In relation to STS-124, engineers attempted to focus in on the SRB holes, to try and find relations with the large elements of debris seen on the wider view. All that is seen is small amounts of debris shooting up and away from the SRB holes at T+0.3 seconds – before the quality of the imagery is lost.

‘The noisy imagery resulted from the camera’s automatic gain control attempting to compensate for the SRB plumes. Despite this, the imagery shows debris moving up and away from the SRB holes after lift-off.’

Managers have previously noted that it would be impossible for the bricks from the flame trench to threaten the vehicle – which is based on analysis of the bricks finding a path upwards through the MLP. This is indeed impossible.

While analysis of the infrared footage is inconclusive, and requires further evaluation, engineers concur with management and conclude the debris has to be throat plugs and water baggie debris – at least for the events observed next to the SRB exhaust.

No specific claim is made for the large amount of debris seen rising up to 300 feet near the vehicle in the wider view – though that debris is seen moving away from the vehicle, hence the concentration on the debris located next to the SRBs.

‘At infrared wavelengths, considerable debris is observed shooting up and away from the general vicinity of the SRB holes after T-0. The majority of this debris is undetectable in visible imagery.

‘This debris is presumed to be throat plug and/or water baggie debris associated with a nominal launch. Analysis of debris trajectories may provide insight into this presumption.’

Analysis Of Debris Threat To Vehicle:

A major analysis effort, to confirm a flame trench related event could not result in debris impacting on the vehicle during the first few seconds of launch, resulted in another presentation that was tasked with confirming ‘previous assessments of unexpected debris liberated from the flame trench hold no transport potential to vehicle.’

The findings opened with historical notes on previous debris events that had the potential to be close to threatening impact with a vehicle during launch.

‘Reports, records, and reviews performed prior to STS-124 identified a previous report of flame trench brick above the MLP deck: STS-71 (object, possibly flame trench brick, observed above MLP deck). STS-83 (‘firewall material’ liberated from SRB trench, found post-launch on apron, in trench, and on MLP 0 level).

‘Flame trench debris analysis was prioritized lowest before STS-124. Expected debris sources in this category not identified as risk drivers (e.g. throat plug, water bag, water bag rope).’

Noted as an objective of the investigation, engineers were tasked with identifying credible transport mechanisms to vehicle, and to provide critical times and release locations within the flame trench to substantiate risk assessment of trench debris liberation during launch.

‘Develop and illustrate with optical record and simulation results the timeline of the flowfield evolution, fluid dynamic events and rebound mechanism that provide transport path of debris towards the Shuttle. Address potential transport paths identified in infrared imagery. Address observations noted during record review.’

Potential sources of bricks flying towards the vehicle first looked at rebound scenarios, where a flying brick had its direction changed towards the rising shuttle, or being redirected by unsteady flow fields by the exhaust of the SRBs.

‘Goal: Explore rebound potential for bricks exiting the SRB flame trench and impacting the retaining wall. Determine whether impacts to the vehicle are possible from bricks exiting northward and rebounding off the retaining wall,’ noted the evaluation process on the presentation.

Results showed that velocity plays a factor, with any impact of a brick on to a surface – such as the retaining wall – it could rebound off, has to be less than 100 feet per second, given higher velocities would result in the brick disintegrating upon impact.

‘Results: Found that there is a critical coefficient of restitution below which rebounds are of no concern. Brick velocity at wall was found to be 500 ft/s. Critical point is 0.3 or less ( w/ rebound velocity of 150 ft/s or less).

‘Actual rebound physics of brick to stone impact are less extreme. Preliminary tests showed bricks disintegrated at impact velocities above 100 ft/s w/ partial rebounds. Rebound angle is a factor in risk – 45 degrees rebounds of most concern. Rebound mass is not a factor – Negligible variance in impact risk from large to small brick portions.’

Another element evaluated was the potential of bricks rebounding off each other during such an event noted on STS-124’s launch – and/or rebounds off the flame trench walls.

‘Goal: Explore rebound potential for bricks rebounding off one another or the trench walls. Determine whether impacts to the vehicle are possible from such impacts.

‘Considerations: Cycle-1 plume is only a rough approximation of actual plume as it is a steady flow. Brick on brick impacts are highly unlikely as likelihood is 4th order function of x,y,z positions and time. Brick to wall impacts are more likely and risky as they are independent of time and stationary bricks absorb limited rebound momentum.

‘Results: Found that even using stationary rebound surfaces for reasonable coefficients of restitution (0.1-0.3), impacts to the vehicle are improbable. Brick velocities are too low until they have exited the trench ( > 300 ft/s ). Angle, velocity combinations which result in impact are severely limited by MLP presence.

‘Time of flight of debris is long enough to allow the vehicle to move out of danger. Rebound mass is not a factor – Negligible variance in impact risk from large to small brick portions.’

To back up the claim that the velocity of the bricks is a major factor when considering rebound potential, engineers carried out simulated tests on fire bricks being shot into steel and concrete targets.

Those results backed up findings that the bricks tended to shatter, thus eliminating their threat of rebounding into the potential path of the vehicle.

‘Firebricks: Material Properties. Brick Material is highly amorphous and brittle. Compression strength is 6,000 psi, but tensile strength is only 1,300 psi. Density = 2.26 g/cc, porosity = 16 percent, elongation at break = 0.05 percent.

’14 tests were run for half cubes and rods at 100-500 ft/s velocities on steel plate and concrete block targets. Brick material tended to shatter, with little or no rebound. 96 ft/s impact shown in image.

‘Current models can reproduce most of the features of the brick tests. Models have correct density, strength, and porosity. Models have 16 percent random voids and statistical fracture model.

‘Tests and simulation show that fire bricks shatter on impact, leaving fragments that range from 80 percent of the brick size at 100 ft/s, to 10 percent of the brick size at 500 ft/s. Transport for fragments larger than ‘typical’ liberation within flame trench is improbable.’

Aiding the evaluations were computational models, that included detailed SRB Flame Trench and MLP. The models contained the entire height of the SRBs and the SRB trench exit, but excluded SSME (Space Shuttle Main Engine) plumes – given their exhaust is no where near as violent as the SRBs.

The model also included Secondary and Primary water troughs, nozzle water deluge pipes, LOX and LH2 TSMs – due to presence of pressure measurements, which allows future quantitative evaluation of results, and also included truncated SRBs to improve tractability of model in the short term.

Using these computation models, engineers were able to evaluate debris paths during the events of the SRB ignition and resulting changes to the exhaust as it aids the vehicle’s rise off the pad.

These models observed an upward jet preceding plume evolution along surface of plume deflector – which it is claimed is consistent with SRB throat plug material emerging form North end of SRB Holes.

‘Cause of upward jet is displacement of leading edge gases whose momentum has been spent accelerating column of stagnant gases in SRB trench,’ noted the presentation. ‘Only direction to get out of way of gases behind is to move upwards.’

Another interesting finding related to ‘Upward Flow Evolution’ – which is when the plume flows down and to the left (X-direction component of velocity). This is observed to cause an upward flow around the sides of trench.

Importantly, the findings concluded that such environments did not support the transport of heavy debris – such as bricks – in its path.

‘Simulate evolving SRB plumes in the flame trench (to 0.7 sec). Survey the unsteady flow field and identify transport potential to vehicle. Transport through SRB hole (before 0.3 sec) – ‘Fountain Effect’.

‘Transport of large heavy debris (bricks and large chunks of Fondue Fyre to 2 lbm) not credible. Transport likelihood for smaller masses down to ‘typical’ environment is improbable. Transport around North MLP deck (0.3 – 0.7 sec) under evaluation.’

What appeared to be the most credible explanation behind debris – some of which has been imaged as bricks – rising vertically out of the trench, is another characteristic of the SRB exhaust, known as the ‘Mushroom Effect’.

‘Physical Description: As initial wave emerges from under North MLP it expands unimpeded. Initially creates large upward flow velocities. As plume continues to develop, smoke is pulled down and to the north. Large upward flow velocities are a transient event – ‘started’ trench flow has significantly less upward flow (volume and intensity).’

This is not classed as a credible transport mechanism for brick to vehicle, which confirms a vehicle is safe from all possible events where a large debris sources can liberate from the flame trench and transport themselves to an impact with the vehicle.

However, the findings do note this as a phenomenon that has a limited ability to transport large debris in its path, and thus is the likely explanation behind how bricks and debris managed to fly vertically in the infrared video.

‘Transport around North MLP deck (0.3 – 0.7 sec) – ‘Mushroom Effect’ (Preliminary) Transient phenomenon with limited transport potential for large debris.’

Based on these findings, the debris observed in the infrared video points to a mixture of debris sources, including the potential of bricks, rising from outside the MLP, rising up into the air vertically via the ‘Mushroom Effect’ of the SRB exhaust.

However, and most importantly, this transport phenomenon is not directed towards the shuttle’s path, meaning the vehicle remains at no threat of being hit by debris from the flame trench during launch.

Final summaries are being sent to managers this week, which will likely close the issue with regards to STS-124.

An update on the status of repairs to Pad 39A will follow this week, as preparations ahead of STS-125 continue on track.

L2 members: All documentation – from which the above article has quoted snippets – is available in full in the related L2 sections, updated live.

**BRAND NEW STS-124 MISSION OVERVIEW MUSIC VIDEO – AVAILABLE HERE** (Free Video Section – Only requires to be logged into the forum)

Click here for a range of superb STS mission specific music videos created by Max Q videos 

Related Articles