11
$\begingroup$

I see many questions regarding the unreliability of thrust reversers. Why are these components so faulty across all airplane models and manufacturers?

Is there a technical challenge to making the the thrust reversers as reliable as other aviation systems? Are the A380's electronic thrust reversers more reliable than the traditional mechanical or hydraulic reversers?

Improve this question
$\endgroup$
7
  • $\begingroup$ There is redundancy just like nearly everything in aviation. Though you may wish to reselect your examples only 1 the crash is about unreliability. $\endgroup$ – ratchet freak Mar 15 '15 at 17:18
  • $\begingroup$ I think there is a fallacy in the basis of your question. Your first and third links to not point to questions about unreliable thrust reversers. $\endgroup$ – FreeMan Mar 16 '15 at 14:49
  • $\begingroup$ Are they really unreliable? In my experience I flew with an inop thrust reverser about 0.25% of the time. $\endgroup$ – casey Mar 16 '15 at 14:58
  • $\begingroup$ @casey: I drive my automobile twice a day, six days a week, at a minimum. Therefore I rack up 400 drives in about a month to a month and a half. Any component that I find is inop on the vehicle monthly I would call unreliable. Are airplanes held to a lesser standard? $\endgroup$ – dotancohen Mar 19 '15 at 7:43
  • $\begingroup$ @dotancohen I operated between 20 and 50 flights per month generally. Less on average when you consider training months, leaves and vacation. Racking up 400 flights took a while. Also note the fleet flew more than 3000 hours per day. $\endgroup$ – casey Mar 19 '15 at 9:29
27
$\begingroup$

It's not so much that thrust reversers (TRs) are more unreliable than other components of the plane. However, they have a very important impact on the airplane's performance. Hydraulics, flaps, and even engines fail regularly. But engines and hydraulics can be backed up by redundant systems, and an airplane can still land without flaps. But engineers learned from Lauda Air Flight 004 that an airplane may not be controllable if the TRs deploy in flight.

So in the interest of safety, the TRs have additional precautions. The reasons for not aborting a landing after they are deployed are discussed in your first linked question. Reliability is also not the only reason for having no TRs on the A380 outboard engines, as discussed in the second linked question.

So to answer your question, there is indeed a technical challenge in making the TRs reliable: they absolutely must not deploy in flight. This means many safeguards are put in place, and they are designed to fail by not deploying if there is any doubt about whether it is safe. A TR that won't deploy is more desirable than a TR that deploys when you don't want it to.

Improve this answer
$\endgroup$
2
$\begingroup$

I would disagree with the assertion or assumption in the question that TR's as "so faulty" and significantly less reliable than other systems. Based on my experience, the reliability of TR's is probably between 99% and 99.9% -- in other words the times when a TR fails to work is between 1 in 100 and 1 in 1000 flights. That, to me, doesn't look like a particularly bad record.

Modern aircraft in general are extremely reliable, so when you compare a component with a 99.x% record against a component with a 99.999% record (say, a windshield -- how often do you see those fail?), the former doesn't look good in the comparison. But when you consider that there are thousands of commercial flights daily in the US, and it's uncommon for anything significant to make the news, the reliability rate for the system as a whole is extremely good.

Some of this is fault tolerance; on most runways, unexpected loss of a TR doesn't affect things very much. The vast majority of the time, braking performance is entirely adequate to stop the aircraft well before the end of the runway even with no reverse thrust (not merely one of two or one of four failed).

But in over a decade of flying modern jet aircraft, the number of times I've seen a TR fail to deploy I could probably count on the fingers of one hand, so all in all I'd suggest that they tend to be quite reliable.

Improve this answer
$\endgroup$
5
  • 2
    $\begingroup$ Regarding your 3rd paragraph, braking performance alone is always sufficient to stop the aircraft without reverse thrust. This is a requirement. If the runway isn't long enough to land and stop safely w/o reverse thrust, you can't land there (except, of course, for the standard caveat of emergency situations.) $\endgroup$ – reirab Mar 16 '15 at 14:19
  • 3
    $\begingroup$ @reirab -- That's not correct. It is legal to land on a runway where contamination etc makes the conditions such that operative TR's are necessary to stop. That legality may reside in an OpSpec, which differs from one carrier to another, but the 'requirement' you mention is NOT universal. $\endgroup$ – Ralph J Mar 16 '15 at 14:49
  • $\begingroup$ The one valid failure example from the original question isn't a failure to deploy, its an uncommanded deployment in flight - a situation that's been rectified in the ensuing 24 years. However, I'd still agree with you that they don't seem any less reliable than other systems. $\endgroup$ – FreeMan Mar 16 '15 at 14:51
  • 1
    $\begingroup$ @RalphJ Ah, good point about contamination. I was thinking of a dry runway. The FAA does indeed require the assumption of inop engines when computing dry runway landing field length (FAR 25.125(f).) $\endgroup$ – reirab Mar 16 '15 at 15:04
  • $\begingroup$ Thanks, Ralph. In automobile terms, a 1/1000 failure rate is considered high. That would mean the same component in your car breaking every three to four months. But I understand that airliners are operated by professional pilots who train for such conditions, which sadly cannot be said of general passenger automobile failures. $\endgroup$ – dotancohen Mar 19 '15 at 8:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.