By the time ATC sees the lateral maneuver isn’t going to work, decides to stop SWA’s climb, says something, and they react to it, they’re already committed to either exceeding the altitude or pinning people to the ceiling.
Meanwhile the TCAS logic only knows what your instantaneous state is. It does not know what you’re about to change. It is semi-common for a last moment ATC directive to come in just about the time the TCAS goes off. Or for you to start reacting to ATC, and about then the TCAS logic is triggered even though you’ve already stared doing what the TCAS will moments later insistently recommend.
TCAS also goes off when there won’t be a conflict. Because it does not know that you intend to level off soon. That’s semi- by design, because it’s also there to guard against the possibility that you ought to level off, but are about to goof by forgetting to do so.
The NTSB has a mandatory report for an RA that truly saved the day. I’ve responded to a few dozen RAs over my career. I’ve never filed that report. And not for lack of diligence.
My bottom line: There are lots of RAs that happen every day. Very, very few of them are truly mid-airs avoided. The vast majority are sorta-close calls that are ameliorated by the pilot response to the RA.
It’s implicit in LSL’s post, but I’ll make it explicit: unless the traffic is in sight and and the pilot believes separation can be maintained, it is mandatory for commercial planes to follow an RA. Meaning, they must respond to it, even if it conflicts with an ATC instruction.
I’ve also had a number of RAs. When the initial warning (a traffic alert, or TA) goes off you put your hand on the yoke and try to spot the intruder. If it becomes an RA (resolution advisory) many modern avionics give you a display with red in the “no fly” zone as computed by TCAS, green for where you should go. That could be a climb, descent, or stay level. My current jet provides a voice too: “Climb, climb!” or “Descend, descend”. Or it might say, “Don’t climb!” Anyway, you disconnect the autopilot, do what’s required, then tell ATC what happened.
All that to say, the SWA plane did what was required. That a vintage warbird jet was involved just makes it interesting.
Granted the crew had to react to the RA. But if the altitude of 14,000 feet was assigned in a timely manner this is a different scenario than what is being portrayed in the news.
It worked out in the end but it’s the difference between a pat on the back for pilots or a kick in the ass. I’m wondering if the RA was for the plane in trail.
A question for the airline pilots, I always thought a flight of X was a formation flight. If there is a mile of separation that sounds like 2 related flights headed in the same direction with minimal separation.
Also, wouldn’t a commercial plane be assigned an even numbered altitude? !4,000 vs 14,500?
Normally in a “standard” formation flight only one airplane is squawking. The others are transponder off. Which makes them invisible to TCAS. Also, a “standard formation” has limits on how spread out laterally or vertically the whole formation can be. I don’t recall the numbers, but the idea is to keep all the planes close enough together that ATC can view them as just a single oversized sorta-blurry airplane for calculating required separation distances.
IIRC there are “non-standard formations” where everybody is squawking separately and visible to TCAS. And handled separately by ATC with an understanding that they’re all going to be given more or less identical instructions to keep the herd all moving in the same way together.
If the Hawkers were operating under VFR in appropriately less-than-fully controlled airspace then nearly anything is possible. Talking to ATC or not, all or just one squawking, cruising at 14,500, etc.
Not much more can be said about what really happened until some investigation gets done and some report gets written and released to the public. Which may not happen.
I think you’re assuming facts not in evidence. Or you’ve looked at a cite the rest of us haven’t seen.
From @JRDelirious’s cite, SWA was climbing to an unspecified altitude. They and ATC knew what it was, but the article didn’t say. ATC said to stop climb at 14,000. But said it too late, so the overshoot was inevitable.
Meanwhile the TCAS RA goes off. Which does not care about altitudes, only about vertical speed. So it commanded a rather aggressive descent and the pilots really shoved the nose over, leading to injured people. As a general rule, you don’t need to play fighter pilot in an RA. Just punch off the autopilot and smoothly alter pitch to get into the target pitch range that TCAS is showing. Which puts you in the target VVI that it’s really wanting. Even if the RA wants a steep descent and you’ve now got a steep climb, that doesn’t mean shove the nose over. It means keep pushing steadily at a measured rate of pitch change and G force for as long as it takes to effect this large pitch change without hurting people standing up in back.
Now if the pilots visually acquired the conflict airplane they may have reacted appropriately, or overreacted, to what they saw.
The altitudes 14,100 and 13,65X are not target values; they’re simply where SWA’s maneuver happened to apex and then bottom out.
Once the TCAS event is over the procedure is to smoothly go back to what you were doing before you were so rudely interrupted. That is, climbing or descending towards whatever target altitude.
Do we have a timeline on when SWA was instructed to stay at 14,000 feet? Also, I would expect SWA to be given the same instructions to look for traffic as the Hawker Hunter was given. ADS-B should have made short work of the process.
In the video you can see some clowns with their large carry on luggage, which means they thought it was a good idea to stop and grab their stuff from the overhead bin while the plane is on fire.
I saw that and noticed the same thing. On the other hand, I’ve seen footage of aborted takeoff tests which are conducted as part of the process of certification of a new aircraft. The plane has to stop using wheel brakes only, which heats up the brakes and tires enough to catch on fire, but the fire must not spread to the rest of the plane for five minutes.
They’d have done better to leave the right overwing exits closed so the billowing smoke wasn’t entering the cabin that was still full of people who wanted to breath.
But that’s graduate-level Evacuation, not kiddie-level Evacuation.
Back in the day we used to have a procedure for the cockpit to order the evacuation including limitations like “left side only” or “forward exits only”. It was later decided that any such instruction would probably be ignored, and hence was futile. To the degree it wasn’t ignored it definitely would reduce the number of exits and hence slow the evacuation of the last person off.
The procedure was eventually removed from our books.
The certification process actually would have included an evacuation test where half of the doors (one per pair) would have been disabled. The test would have shown that evacuation was possible in under 90 seconds.
I bet the test didn’t include entitled chuckleheads forcing their way up and down the aisle to retrieve their precious carry-ons and trying to push them out an overwing exit.
The test does include small carry-on luggage and pillows and blankets to be placed in the ailes and passageways. The typical safety briefing is given and the obstacles are then placed according to the test plan.
The test is also done under fairly dark conditions, using emergency lighting only.
I read somewhere that, to reproduce the chaos and desperation of a real evacuation, some tests have been conducted that included a promise of $100 to the first 10 (probably not the correct number) people off the plane.
In recent years it’s been pretty well established that the FAA standard evacuation tests are very much not representative of actual real world evacs. The problem is there’s no reasonable way to increase the fidelity at acceptable hazard to the test participants. And admitting the test is wildly optimistic is also admitting that substantially no airplanes actually meet the published standards.
Of course the historical record also shows that, of those accidents where evacuation is a plausible response, almost everyone who’s ambulatory when the wreckage quits moving gets out rather than dying inside trying to get to an exit. So regardless of any regulation and compliance process, the real world performance is adequate to the real world task at hand.
So we have a badly chosen arbitrary legacy standard and 75 years of functionally faked tests that meet the BS standard. Notso hotso overall.
We regularly have discussions in our work about how much is make believe when establishing assumptions and rationale for human capabilities and behaviors. People somehow never act the way the “rules” need them to!
Or you can do every test, add complicating factors to improve realism, etc, and ultimately….a plane still crashes with no survivors and all that work can feel a little pointless. But it would be so much worse without at least that level of safety, so we keep going!
I have strong opinions on how many of the regulations need to be rewritten, updated, improved or even discarded in favour of other ones. It’s a very slow thing to change.
Oh it gets really nerdy really fast! I’d end up writing pages of arguments! I’ll see if any that can be quickly explained come to mind, but I post on my phone and writing long posts is difficult.
One that’s irritating me today is that the FAA has certain equipment on a list of issues that they perceive has different rules in different countries (Canada and Europe) and therefore would need special approvals for FAA acceptance. In that list is one item where the rules aren’t actually different but where trying to get concurrent approvals turns into a multi-month paperwork exercise to document how same things are the same.
I’m also tangentially involved in a compliance discussion about how to show that a water system works after prolonged storage at -35C… water….below freezing. The argument that “you add water after you warm up the plane” is getting stuck on a very strict definition where the system must be shown to work.
It’s more nuanced than that, and I don’t disagree with the idea of the regulation in question, but sometimes it seems reality gets lost in the paperwork.
