Wow. Sometimes this thread sits untouched for weeks and sometimes we get 30 posts in 24 hours.
I’m going to try to address @Dead_Cat’s comments from a couple days ago about the Pinnacle accident. Without writing a book.
Service ceiling:
As noted by the other pros up-thread, there are really two service ceilings. One is regulatory, where a limitation is published based on manufacturer’s testing versus a new lightly loaded airplane under specified test conditions. The other service ceiling is the one your jet has today, based on the current air temperature, turbulence, engine performance, gross weight, trim, and actual and intended speed. One is easy to remember. The other is nebulous and hard to determine except by trying it to see. Perhaps assisted by intuition informed by training, personal experience and also by estimates from your FMS.
Operating above the former might be (barely) possible today but is always a violation. Operating above the latter is practically physically impossible, and even operating too closely below it may be dangerous, even if it’s fully legal.
The CRJ FMS was known to not provide useful service ceiling and power setting info at high altitude. Why that was allowed by FAA is an interesting question in its own right. What’s certain is the pilots had been provided paper charts to compute the things the FMS did badly. The NTSB report says they couldn’t be certain whether those charts were consulted by the crew, but on balance they beleived they had not been.
Training:
There is a problem today in the industry supplying pilots. The basic flight schools teach what’s needed to be able to fly straight-winged Cessnas at low altitude. The airlines (both majors & regionals) teach newbies how to operate the specific aircraft type as a machine, and a certain amount of jet-specific swept-wing high-altitude general knowledge. But there is not really an effective QC at the industry level to ensure there’s enough of that general knowledge covering all the bases, and that it’s well understood and well-retained by everyone.
Every business in every industry likes to assume they’re hiring workers with all the skills to be productive right away. With FAA oversight the airlines are less so. But they still like to believe the people they’re hiring come with a large and gap-free basis of general knowledge. Because it costs time and money to supply that knowledge. FAA mandates many topics be covered. But not necessarily all, and not necessarily at the right depth for the less well-prepared and less-insightful members of the potential new-hire community. It may not be obvious who has which gaps in their knowledge and skill base until a crisis shows them wanting.
High altitude engine failure:
An engine can quit at any time. Sometimes, as when a fan blade comes off or there’s a fire warning, a restart is obviously inappropriate. Other times the failure may be completely unexplained. It certainly will be completely unexpected. Was it turbulence, volcanic ash, internal flow-path icing, anti-ice system failure, fuel contamination, fuel pump failure, fuel control failure, a sensor failure, ice or other clogging in the fuel filter, or maybe some small compressor / turbine blade(s) failed and went out the tailpipe?
When one engine fails, the general plan of attack is to set up a glide at a lowish speed that minimizes altitude loss. As a generic example, in the 737 we usually cruise between 30,000 & 41,000. On one engine the highest altitude it can maintain at typical weights is around 25,000. On just one engine we will be going downhill, it’s just a matter of how quickly pointed in which direction.
Once the thrust asymmetry is compensated for and a glide is established, the next step is to sort out your mechanical problem and ATC problem, figure out where to go, then go there. If you later decide you want to try a restart that can be done by diving for airspeed to windmill the dead engine. Or on some types, including most modern jets and the CRJ, at lower altitudes you can start the APU and use that to help motor the dead engine. In either case an airborne start is something you’ve probably never done before, not even in a simulator. And it’s quite a different experience from a ground start. Much slower and much more uncertain. Ref my post a few posts ago on turbine starts.
With dual engine failure it’s a bit more exciting. Most of the lights & instruments go dark / blank as the generators drop off line and you revert to battery power for a limited subset of cockpit stuff. As much as you’d like to set up a glide to buy max time and range to locate a dead-stick runway, instead you’ve got to dive steeply to accelerate enough to get and keep the engines in their windmill start envelope. If you do get one started, you’ve (probably) saved yourself from a dead-stick landing probably off-airport. If you don’t get one started, you’ve vastly reduced your range of alternatives for landing on-airport.
If you’re already majorly slow because the engines stalled / flamed out as a result of airflow distortion from the flailing around following an aerodynamic stall (i.e. like Pinnacle had), then you’ve got to dive like a banshee for 10-15,000 feet right friggin now. As in immediately pissing away half of your altitude buffer to ground impact, then continuing to burn the rest at an alarmingly good clip.
Core lock = engine seizure:
When I first read that the engines seized I was flabbergasted. In my then-30 years in the biz having flown umpteen models of engines from every manufacturer there is, the very idea that a failed engine could seize just from thermal stress and you’d never get it turning again was totally news to me. No aircraft manual I’d ever read had included any concerns about that in any discussion about inflight engine failures or restarts. The idea that once the RPM gets low it may take an ungodly amount of speed and altitude loss to get rotation back up again was pretty commonly stated. The idea it might be flat impossible was not.
I am still amazed that this seizure event wasn’t a shock through the industry. A bit like lawyers say “bad cases make bad law”, this accident, which was a bad example of bad pilots piloting badly, served to completely mask the IMO much larger issue of inherently dangerous engine failure modes that are unknown to the pilots using them.
IMO this ought to have been a cause celebre, akin to the MAX MCAS fiasco. What did the manufacturers know when and why was it kept secret from the pilots and the regulators? Instead we got … crickets.
Follow-ons from Pinnacle:
One of the things that changed as a direct result of this accident is the FAA mandated more background info in the aircraft manuals about high altitude stalls and high-altitude performance in general.
They also mandated, for the first time in my career, that high altitude simulator work be included. That had never been done either. Hand flying up at / near service ceiling. Doing slow-flight and seeing how much engines can’t help you; descending is the only way out of the trap once you’re in it. Failing engines in cruise & dealing with the vast differences versus the traditional engine failure practice during takeoff on the runway or at lift off. Stalling in cruise. Hitting severe turbulence in cruise & being rolled sideways or upside down or pitched way up or down.
A LOT of gaps in knowledge and in skills were uncovered as the crew forces all around the industry went through all this. Some corporate embarrassment ensued as they realized how un-ready much of the crew force was for dealing with these things that simply had been assumed to be easy and simple and obvious. Not so.
I’ve got to get clean & go to work now. But that’s plenty on Pinnacle.
