KSC Next Gen Site ___Questions? Comments?
A Review of Space Shuttle Data
Shuttle Orbiter Line Replaceable Units (LRUs) Replaced per Flight During Orbiter Processing Facility (OPF) Turnaround Operations
How many parts are usually replaced on a Space Shuttle Orbiter between flights?
On average about 100 unplanned problem reports of the type requiring a removal or replacement of a Line Replaceable Unit (LRU) level part are initiated on every Space Shuttle orbiter between flights. These 100 LRUs are usually changed out due to failing a test, criteria or some measure of confidence, as part of the process that prepares the systems for a safe next flight. The parts consist of an assortment of items ranging from valves and regulators to major sub-systems such as actuators. The actual parts that fail, or in which confidence has been lost, will vary from flow to flow.
The data shown below does not include 2 major sub-systems. One of these systems not included is the Space Shuttle Main Engine (SSME). This removal and replacement of whole engines is, nonetheless, a regular event between flights. The second major system not included in this data is the fragile Space Shuttle orbiter Thermal Protection System (TPS) components - tile, blankets, soft goods, and assorted parts.
It is worth noting the Shuttle reliability against loss of vehicle (RLOV) was originally specified as 0.98 (catastrophic loss of one vehicle in every 50 launches). Current estimates, post Challenger, are marginally higher. This relationship of hardware as it behaves in flight, and hardware as it behaves on the ground is a link, also related to design life, which must be improved upon.
A next generation Reusable Launch Vehicle (RLV), must improve on these major parameters in order to truly open the space frontier. For comparison, a Shuttle with a reliability of 0.98, and with a design life per orbiter of about 100 cycles or flights, compares to airliners designed at 0.999999+ reliability (a probability of loss of life/vehicle in the 1 in a million's) and design lives for airliners in the 10's of thousands of cycles (70,000 flights is not uncommon).
This link between hardware reliability in flight, and the ground processes that are required during turnaround, is inevitable and inseparable.
Future RLV’s must have design and development processes that will result in higher reliabilities across all hardware. Low Shuttle like reliabilities, though relatively the highest of all launch vehicles, will not be able to achieve, because of this basic linkage to ground processes, any opening of the space frontier. Low costs and fast turnarounds, goals such as $1000 or $100 dollars per pound of payload, or flights per week per vehicle, will require HIGH vehicle / ground system reliabilities against catastrophic failure. Gains in safety, as through reliability (versus escape systems), will translate into proportionally low maintenance and turnaround costs (or vice versa, as today).
Lack of confidence, a result of low reliability against loss of vehicle, otherwise creates ground processes that seek to “find the broken parts” meaning planned as well as unplanned work.
- Planned work seeks to test systems that have likely had previous failures too often to simply assume they will function next time.
- Unplanned work results when these tests and checkouts find the failed parts. Bad parts must be removed & replaced.
- New parts purchases at low volume create considerable expense.
- Parts deemed reparable to save monies may still incur, by definition, a hi benchwork cost.
- Collateral work results simply from damage from having to access areas to perform work on a system that was not designed with maintainability in mind.
- Having such a unique, low volume design also makes changes difficult as parts changes only introduce more uncertainty. This over time results in obsolescence and lack of supportability (parts unavailable at any price).
Another factor to consider is robustness to failure. Is there a requirement for being able to lose an engine in flight and still assure vehicle and crew safety? How many of these failure modes actually result in collateral damage that means loss of vehicle? Can a vehicle loiter on its return landing flight? Can any one system fail rather completely and still allow safe flight by proper design? These are the goals that must become requirements of next generation designs if access to space is to become routine and low cost. For a vehicle to land from a space flight, be turnaround in a few days, and be launched, such simple metrics as reliability must be dramatically improved upon. Rigorous certification, testing, test to failure, fix, redesign and retesting processes, as expensive as they may be, will be a path to reduced operations costs and decreased turnaround times, resulting in productive, responsive space fleets.
_____________________
Also see:
_____________________
Website Contact: Edgar Zapata, NASA Kennedy Space Center