Window damage on Atlantis threatens six month delay to STS-129

by Chris Bergin

Meetings have been taking place on Wednesday into evaluating damage to the pressure pane on Atlantis’ number 5 window, after a work light knob was observed to be embedded between the pane and the dashboard panel. The damage can only be fully assessed once the knob is removed, with the threat of a six month schedule impact to STS-129 noted, should the damage prove to be unacceptable for flight.

STS-129/Atlantis Processing Latest:

Recently returned to the home comforts of her Orbiter Processing Facility (OPF 1) after the successful STS-125 mission, Atlantis is being processed for the November 12 NET (No Earlier Than) launch date for STS-129.

The final mission of 2009 will see Atlantis deliver the Express Logistics Carriers 1 & 2 (ELC-1 and ELC -2), along with SASA and MISSE (Materials on International Space Station Experiment) 7A and 7B, to the Space Station.

Processing is proceeding to plan, with functional testing and post flight deservicing and troubleshooting currently taking place.

“The Aft Propulsion System (APS) pod functional checkout continues through next week,” noted Wednesday processing information on L2.

“Waste Management System post flight servicing is complete less post ops. The Star Tracker Lightshade inspection was rescheduled from yesterday to today. S-band trouble shooting (IPR-0014) begins today.”

Engineers are also working one issue noted with the orbiter cooling system, specifically the Freon Coolant Loop (FCL) 1 Radiator Flow Control Assembly (RFCA), which “failed to get into radiator flow on orbit.” However, engineers have been unable to recreate the issue inside the OPF.

“Functional testing resulted in nominal operations,” added the processing update. “The RFCA was cycled eight times without an anomaly. Data will be reviewed with JSC (Johnson Space Center) and the vendor.”

As per normal in an orbiter’s flow, all of the vehicle’s windows are being inspected for potential damage to the thermal pane – the outward facing pane – due to the usual occurrence of small MMOD impacts. These panes can be replaced reasonably easily.

“Window 1 Loctite application began yesterday and is scheduled to complete today,” added Wednesday’s update, following last week’s installation of a replacement thermal pane on that window.

However, it was during those inspections where potential damage to the inward facing pressure pane was observed on Window 5.

Window 5 Evaluations:

w2“Quick shoe mount knob from a crew work lamp was found wedged between pressure pane 5 and dashboard panel. Knob must be removed to evaluate pane condition before flight,” noted one of two expansive presentations, dated June 24, and created for high level managerial overviews – both presentations are available on L2.

Several attempts have already been made to remove the knob by applying dry ice in the hope the knob would shrink, thus allowing for it’s removal without causing additional damage to the pane. However, those attempts have been unsuccessful.

“The Shuttle expands when on orbit. While on orbit, a knob got stuck between the window and the frame on Window 5 (Pilot window),” added another memo on Wednesday (L2). “They have tried a number of tricks to remove it with no joy so far.

“If they have to remove the window, there is a schedule hit. Initial estimates are long, but initial estimates always are long before they sharpen their pencils.”

Those estimates speak of up to and over six months being added to the processing flow of Atlantis, which would have a major impact to the overall manifest that already has a full five flight schedule in 2010.

Even if Atlantis was retired as a result of a half year-plus processing hit – one absolute worst case scenario noted by one engineer on Wednesday – two orbiters could complete the manifest. However, due to LON (Launch On Need) requirements, a manifest stretch deep into 2011 would be the fallout.

Such discussions will not take place unless unacceptable damage is found on the pane after the knob is finally removed.

w3Inspections of knob have been carried out via borescope inspections. Photography shows the mounting plate appears to be dug into close-out panel. Further inspections will be required ahead of potential plans to remove the knob without causing additional damage.

Unique engineering plans would have to be drawn up in the event of unacceptable damage being found on the window, given a pressure pane has never been replaced at the Kennedy Space Center (KSC) over the history of the shuttle program, and ferrying Atlantis to Palmdale is obviously no longer an option – after the Orbiter Major Modification (OMM) facility at Plant 42 was shut down over six years ago.

The bottom line is Atlantis would not be allowed to fly – in the event of the pane breaching unacceptable damage criteria – due to the risk of the window failing during her next mission, which would result in a LOV/C (Loss of Vehicle and Crew) event.

“Knob wedged between pressure W5 and dashboard could have caused damage to pane and its currently loading the pane at two points. Removal of knob is required to evaluate condition of pane for flight,” added one presentation.

“‘Fly as is’ is not an option, unknown damage and loads to the glass could result in failure during the flight, with no redundancy; dynamic failure could result in redundant pane failure.

“Induced damage of the knob being wedged between the glass and the dashboard closeout panel structure, or from removal of the knob could result in unacceptable damage.

“Consequences of unacceptable damage to the glass pane: Replacing the pressure pane would result in a significant impact to ground schedule (potential 6+ month impact). Requires de-configuring dashboard structure and instrumentation to remove window assembly for refurbishment. Windshield pressure pane removal has never been performed at KSC.

“Knob removal must be performed carefully; exhausting all risk free options first, then attempting more intrusive (higher risk) options, if others fail.”

w4Damage to the pane is confirmed in the main overview presentation. Although the full extent won’t be known until the knob has been removed.

“Photos show evidence of existing damage to the pane. Cannot be assessed until knob removal. Glass is an aluminosilicate tempered pane (t=0.65). Tempered layer 21 percent of thickness or around 0.13 inch. Tempered glass has a built in residual stress, tempered layer penetration will result in failure.”

As far as unacceptable window damage criteria, ding depth greater than 0.0015” and/or coating scratches with a width greater or equal to 0.050” are listed, along with “no acceptable criteria for bruises or chatter checks.”

The redundant pane – which is located between the thermal and pressure pane – is also considered an inadequate redundancy for pressure pane failure.

“Analysis shows up to 70 percent negative margin of safety on redundant panes as a result of a dynamic failure at cabin pressure. Energy from pressure pane failure may be enough to break redundant pane Constitutes a catastrophic failure for the mission as thermal pane is not certified to hold cabin pressure.”

As to how the knob managed to find its way into the area between the dashboard closeout panel – which is the extension of the glare shield – and the window pane, changes in pressure for pre-launch to orbit operations, leading to the expansion of the orbiter’s skin, is once again deemed as the root cause.

w5Crew module skin expands while in orbit due to 14.7 psi internal pressure; flight deck floor deflection may also contribute to the relative movement between the console’s dash and the CM window area,” added one presentation.

“Gap between the dashboard closeout panel and the window may have opened wider while in orbit (to accommodate the knob). Additional 7 psid and vibration from ferry flight might have wedged knob further. Gap reduced when CM (Crew Module) internal pressure reduced (descending). Knob is exerting a preload on pressure pane and closeout panel.”

An array of techniques have been listed as potential methods for removing the embedded knob from between the window pane and the console dashboard, although most are classed as holding risk of causing additional damage to the window – and are thus classed as last resorts.

“Pressurize Crew Module and use dry ice on knob: DeltaP from pressurization would be much less than seen on orbit (~3 psid instead of 14.7 psid). De-configure dashboard/console and instrumentation to free up knob. Use an air bladder between window and dash,” the presentation added.

“Cut knob enough to yield part. Cut out part of dash where knob is wedged: Would damage dash likely requiring repair. Cutting vibration could induce further damage to window. Use pry bar to deflect dash down: Viewed as the riskiest removal method for all hardware involved.

“Further damage could increase the potential for unacceptable damage, resulting in significant schedule impacts.”

w6The options relating to the application of dry ice to shrink the knob have already been attempted, and failed. However, the same method, along with a pressurization of the Crew Module, may be enough to free the knob from its lodged position.

The only problem is the amount of pressurization that can be conducted in the OPF is far less than the pressure that played a part in allowing the knob to become embedded in the first place.

“Pressurize crew module and dry ice on knob to TBD (To Be Determined) pressure. Pro’s: Could allow for uniform structural deflection to increase gap between pressure pane and dashboard; enough to free up the knob non-destructively. Less potential for inducing further damage to the pane.

**UPDATE: The above option was selected on Thursday as the opening process for an attempt to remove the knob. The cabin will be pressurized to 3 psid, before an engineer will apply dry ice to the knob. This option is not deemed to be a likely solution, but more so the opening option that avoids additional damage to the window.**

New article, click here:
http://www.nasaspaceflight.com/2009/06/atlantis-window-pane-begins-longerons-damaged-accident/

“Con’s: Maximum allowable pressure might be limited by crew module equipment and systems May not be able to exceed 3 psid; as compared to 14.7 psid nominal (orbit). May not be enough pressure to dislodge the knob.”

Removing dashboard and console components to free the knob from below the window is another option, but would lead to a large schedule impact for Atlantis’ OPF flow.

“Pro’s: Less potential for inducing further damage to the pane. If damage is found to be unacceptable, console is partially de-configured to continue with window assembly removal.

“Con’s: De-configuring console and instrumentation could represent a significant schedule impact. Influence of panel removal on the knob are not known at this time.”

w7A pressurized air bag method – involving one (or several) air bladders between window and the dashboard closeout panel adjacent to knob – could be used to apply pressure to deflect the dashboard enough to free knob.

However, without knowing allowable local pressure that can be applied, this technique could also result in additional damage to the window.

“Con’s: Risk of inducing additional window damage (TBD by Stress Group), not quantifiable. Current load on glass is unknown, additional load from air bladder may result in additional stresses to the glass.

“Not enough open surface adjacent to the knob for bag contact. Pressure may not be enough to deflect hardware and release the knob. Possible to inflict permanent deformation to dash secondary structure.”

A visit by the Dentist may prove to be a solution for Atlantis, via cutting the knob with a high rpm/low amplitude dental drill, just enough to allow for the knob to be freed from its current location. However, the vibrations from the drill may transfer to the window pane and again increase sustained damage.

“Pro’s: Method can be accomplished quickly. Does not involve manipulating window or dash hardware.

“Con’s: Potential for tool vibration to be transferred through knob into window induce additional window damage. Debris release from cutting. Knob would only be cut enough for tech to yield the piece.”

Vibration concerns are also cited for another option relating to cutting out part of dashboard structure where knob is wedged, while another option of using a pry bar to deflect dashboard structure down; enough to release the load imparted by the knob on the glass, again threatens further damage to the window pane.

w8Once the knob is finally removed – which has to be carried out – engineers will use their predefined criteria to hopefully clear the window for flight.

“After knob removal, will assess/evaluate window damage. Inspectors will inspect window per MT0501-514 requirements. Mold impressions, visual examination, etc.

“Stress evaluation will be required if surface damage reported is beyond the acceptable criteria 0.0015” for tempered panes or if bruising of glass has occurred. Unacceptable damage will require a window assembly R&R, significantly affecting the flow schedule. Bruised pressure pane will result in an automatic scrap.

“NOTE: Damage is in an area difficult to reach; if window inspectors are unable to assess the flaw, engineering is not able to perform an evaluation. Access to the lower region of the pane is limited and removal of crew module dashboard and panel structure may be required for proper assessment.”

For the interim, engineers will continue to meet, in order to discuss the best forward plan on removing the knob, before moving forward with the option that holds the least amount of risk of causing additional damage to the window.

“Removal of the knob is necessary in order to evaluate the condition of the glass for flight. Some of the options being considered involve risk to the hardware,” concluding the main overview presentation.

“Knob removal attempts should be performed by exhausting least risk options first. More evaluation is required.

“Forward Plan of Action: Will continue to evaluate and further develop these options and other feasible options in order to remove the knob with the least amount of impact to the hardware.”

L2 members: Documentation – from which the above article has quoted snippets – is available in full in the related L2 sections, now over 4000 gbs in size.

Related Articles