OK but once you have demonstrated your gearbox passes that test, once then its done. i.e its not an issue if you come out later and say our gearbox can't run for 30 mins or there is an accident where lubrication was lost and the gearbox was unable to run for 30 mins.
I guess what I'm saying is that there is an assumption that once that initial test is in the bag and the 30 min demo box is ticked then it is assumed that future failures are the anomaly and not the initial test..?
Pitts- If you read the relevant section of FAR 29 that dascanio posted, you'll note that there is lots of room for interpretation in the requirement as written. What you are saying about the FAA loss-of-lube qual test is basically correct. The regulation only requires a single conforming test article to pass a single test procedure. However, the loss-of-lube test itself is only one part of the total gearbox certification process. Besides the loss-of-lube test itself there are also analysis efforts associated, such as FMEA/reliability.
One thing that the regulatory agencies can do to address this issue in the near term is to revise the regulations. Even without increasing the "run-dry" capability beyond the current 30 minutes, it would be a good idea to require testing of more than a single gearbox due to the large number of variables that can affect the results of this particular test. Conducting this qual test on 3 or 4 conforming test articles would help to minimize the potential for basing a pass/fail decision on a statistical anomaly. Since the loss-of-lube qual test results in destruction of the gearbox, requiring a larger number of test samples will increase certification costs. But spending an extra 2 or 3 million to improve the level of confidence regarding critical MRGB loss-of-lube operational capability is probably a wise investment.