The investigation of the Jan 2022 ramp strike by an F-35C
has become public and is presented in a USNI News website article.[1] I’d like to draw a lesson from it that the
Navy won’t and yet it’s the most important one.
To summarize the incident, a pilot was attempting to perform an abbreviated landing procedure, wound up with too low an approach speed, couldn’t recover, and suffered a ramp strike. The Navy’s investigation determined that the cause was pilot error for failing to switch on two automated landing aids which control approach speed (glide slope) and angle of attack. From the article/report,
The junior officer had performed a specialized landing approach to Vinson for the first time, but he did not realize a built-in aid that helped control the plane’s power during landing was switched off.[1]
… the [pilot] remained in manual mode when he should have been (and thought he was) in an automated command mode designed to reduce pilot workload during landings.[1]After the turn, the pilot didn’t engage the two landing assist tools on the fighter, the Approach Power Compensation Mode (APC) and the Delta Flight Path (DFP), which automate some of the pilot’s necessary tasks for landing on an aircraft carrier. When activated, DFP automatically adjusts the throttle to keep the aircraft on correct glide scope to land on a carrier, while the APC maintains the fighter’s angle of attack.
… investigators recommended that aviators stop performing Sierra Hotel Breaks, that policy require F-35C pilots to always use Approach Power Compensation Mode (APC)/Delta Flight Path (DFP) throttle assists and that heads-up displays include indicator lights showing when flight aids are activated.[1]This is just increasing our dependence on automated aids. What will future pilots do when the automated aid won’t work due to electrical failure or combat damage? How will they land, having never learned/practiced manual landings or never been forced to maintain a level of expertise? By increasing dependency on automation, we’re ensuring that pilots will become less capable and less skilled and will have more mishaps, not less.
We are dependent and addicted to technology and are losing our ability to function normally.
________________________
Note: We see this in society, in general, too. For example, the routine use of calculators
in school has produced a generation of students who have no mathematical
ability. They are unable to calculate
change at a register without aids. They
are unable to estimate large number manipulations (exponential manipulations
and order of magnitude manipulations). They
don’t know multiplication tables. In short,
they’re mathematically ignorant. I see
it every day.
_________________________
https://news.usni.org/2023/02/21/pilot-error-after-sierra-hotel-break-resulted-in-south-china-sea-f-35c-crash-investigation-says
So we need to get rid of fly by wire then and go back to cables attached to the flight stick no automated aides!
ReplyDeleteLooks at the accident rates for carrier landings before modern computerised aides yeah the US can't afford to return to that kind of aircraft loss rate.
Okay yeah that was too blunt but the F-4 Phantom had over 1800 accidents in it's time compared to 250 for the F/A-18 A-D and 720 for the F-16 to date.
DeleteModern flight controls and automation make the planes safer to fly and that is necessary.
Look at the F-8 Crusader nearly every one built was involved in an accident at some point, yes they weren't all landing accidents and yes flying isn't risk free but risking aircraft and lives because you want more manual control really is an 'old man yells at clouds moment'.
Don't be obtuse. I'm not talking about returning to WWI era biplanes. I'm talking about technological dependencies that degrade fundamental skills.
DeleteThere is a case to be made against fly-by-wire that is implemented to compensate for an inherently aerodynamically unstable aircraft. If damaged (or cyber-attacked or 'jammed' or otherwise electronically attacked), such an aircraft becomes unflyable and, in combat, that's not a good thing.
"F-4 Phantom had over 1800 accidents"
DeleteUnless you'd care to do a comparative analysis of the mishaps, your numbers are meaningless. For example, there were 5195 F-4's produced compared to 1480 F-18A/B/C/D so, naturally, there would be more F-4 mishaps. The F-4 was in combat nearly its entire career whereas the F-18 has seen almost no combat so, again, one would expect combat driven aircraft to have a much higher mishap rate. What were the types of mishaps? Also, the Navy has severely restricted aircraft flight envelopes in recent decades to prevent mishaps as opposed to the pilots training for actual combat in the F-4. With the F-4, we were still struggling to understand why a jet flies and all manner of equipment has improved since then, most of it having nothing to do with flight control computers. And so on.
Your comment lacks any relevant context and is meaningless without analysis.
Anonymous @2/23 0457,
DeleteThe Navy F-4’s did have an Approach Power Compensation System, basically an autothrottle coupled to an angle-of-attack sensor, for decreasing pilot workload on carrier approach. (My dad was an F-4 depot electrician at a NARF, so he got to repair and calibrate them when they failed, which was actually a very cool process….the functional testing involved sitting in a chained-down aircraft with the engines on and watching the engines respond to the AoA sensor with the autothrottle engaged, which vibrated your soul when they went into AB.)
The difference between then and now was likely that F-4 pilots back in the day were expected to be able to proficiently land the aircraft without the autothrottle (because it regularly failed), whereas now I guess the assumption is that the automation will always work so there is no need to practice without it.
"My dad was an F-4 depot electrician "
DeleteOutstanding anecdote! I love that kind of real world experience contribution. Thanks for sharing with us!
Quoting from the post by Anonymous on the prior thread:
ReplyDelete"This is a very similar critique Roger Thompson makes in his book on the US Navy, "Lessons Not Learned". His priorities are #1 Training, #2 Maintenance, and only then #3 Technology/Equipment. In other words, the exact opposite priorities of much of the current military."
USN, and military in total, need to get their priorities realigned properly.
Training is good but training to a dependency that degrades fundamental skills is wrong. A soldier who has been perfectly trained to use GPS but can't read a map or navigate cross country is wrong. A ship's crew that has been perfectly trained to sail a ship with GPS but can't shoot a bearing or manually plot a position/course is wrong. A pilot who is perfectly trained to use automated landing aids but can't land if those aids are damaged in combat is wrong.
DeleteI think I've made it pretty clear in prior posts that my concept of training turns off all the gadgets on day one and makes sailors go with the most fundamental and reliable methods. If I haven't, then I hope I have now.
DeleteHow about a compromise. Many of these systems do produce better results than manual operation when they work. However, obviously they occasionally don't. So how about making use of them but also training for how to operate without?
ReplyDeleteThat sounds great on paper but, human nature being what it is, it's impossible to avoid lazy, bad habits when using aids. The pilot in this mishap undoubtedly was trained to fly manually but had become dependent on automated aids without even realizing it ... human nature.
DeleteYour suggestion needs to be reversed. You train and operate manually, all the time, and you have a big, red, 'use only in emergency' button for automated assistance when your best manual effort isn't good enough. And, if you do use the emergency button, you get evaluated for competency because you shouldn't need it unless you and/or the aircraft have been injured/damaged. Similarly, if you're in-country and need GPS, one has to question your basic competency.
"compromise"
DeleteSo much of the military has become a compromise. Instead of doing the hard work of maximum training and fitness, we've introduced endless compromises, second chances, waivers, gender norming, sensitivity, and so on. We cloak it all in various wonderful sounding rationales but it's all just compromises. Compromise is the attitude of the weak. National security should have no compromise. You do the tough, hard, dirty work that's required and you do it without compromise.
It is hard to tell exactly what was required to execute this approach as I do not have access to the Pilot checklists. However, it is not uncommon for automated systems to be mandatory based on the procedure being executed. It does not excuse the pilot for not flying the airplane and recognizing a hazardous condition, Pilot in Command means just that, responsible for the safe operation of the aircraft. To say that we should eliminate automated landing aides is not the answer and is an extreme. Many aircraft use automation in some form or fashion during the entire flight envelope. For instance ... F/A-18s pilot inputs "demand" a condition they do not dictate what control surface gets the airplane to that condition, the airplane will (and is very good at) figuring out what mix produces the flight condition commanded. Another example is H-60 stability augmentation. Can a pilot fly without stab aug engaged, absolutely but, it is very laborious and requires a lot of pilot concentration. Point being, automation is not just "hands off" flying, automation makes aviation safer and comes in many forms and fashions. From what I read, this was pilot error, auto appears to be a required checklist item and it was not selected, pilot responsibility and should have been caught before extremis. Do not through the baby out with the bathwater ... Aviation, navigate, communicate, meaning always fly the airplane first. And during normal, abnormal and emergency procedures, make sure to complete your checklists. My two cents.
ReplyDelete"automation makes aviation safer "
DeleteYou've utterly missed the point of the post. The pilot failed to recognize and react in time BECAUSE HE WAS DEPENDENT ON AUTOMATION AND JUST ASSUMED IT WAS WORKING. He had become dependent and, therefore, manually deficient.
Automation is fine UNTIL IT ADVERSELY IMPACTS FUNDAMENTAL SKILLS. Our pilots have become automation-dependent and have lost fundamental flying skills. Our sailors have become GPS-dependent and lost their ability to manually fix their position and navigate. Our soldiers have become GPS-dependent and lost their ability to read a map and navigate overland with just a compass.
"Can a pilot fly without stab aug engaged, absolutely but, it is very laborious and requires a lot of pilot concentration."
DeleteExactly. And that labour and concentration turn Just Some Guy in an actually experienced pilot, over time.
Respectfully disagree with you CNO, he was not dependent on automation, that is not why he crashed the airplane. He crashed because he put himself in extremis and did not recognize in time he was in extremis. Pilot error, did not complete his checklist, in essence, stopped flying the aircraft. What you are not taking into account is the required systems required to fly the airplane, in this case, automated landing aides required otherwise, approach not authorized, proceed with normal approach; Pilot failed to recognize this, not that he lost his fundamental skills. 99.9% sure that if he had accomplished an normal approach, he would have landed safely.
DeleteOBTW I just read an article that referenced what this maneuver most likely was ... SHB.
ReplyDeleteThe report, which was provided to Military.com by the Navy, said that the pilot, identified only by his rank of lieutenant, wanted to try an "expedited recovery maneuver" that naval aviators know as the "Sierra Hotel Break" or SHB.
The report describes it as "when an aircraft initiates a turn to downwind from either behind the ship or over the top of the ship … [using] G-forces to decelerate over the course of a 360-degree turn, dropping the landing gear when the aircraft is below landing gear transition speed."
I am 99.9% sure that this is not a standard procedure or has operational relevance to Carrier aviation. Point being, not a training or operational requirement, end result, Pilot made a decision to attempt, this approach was at least in principle accepted by leadership and the result was F-35 loss for no good reason, Pilot error. Board convened, Pilot no longer flying for the Navy.
"OBTW I just read an article that referenced what this maneuver most likely was ... SHB."
DeleteThe maneuver was named and described in the cited reference in the post. Be sure to read the reference!
You've completely missed the point of the post. What was the point? Do you know?
Again with all due respect, believe your point is not valid. The real point is, we lost a plane because an aggressive landing approach was attempted and not correctly executed. Blaming automated landing aides for lack of pilot skill / proficiency is missing the point. I submit, agree to disagree here.
Delete" believe your point is not valid"
DeleteThe point is 100% valid and even the military has acknowledged it. GPS is an absolute, ironclad example. We've lost basic seamanship, navigation, and airmanship skills due to dependence on GPS.
Whether you believe dependency was an issue in this specific incident, is up to you but the phenomenon is indisputable.
I only had to read the first sentence to know that the real problem was lack of skill and training. The fact that the Navy and the rest of the military missed this obvious issue is why we will lose the next war. Automation is needed and necessary, but the pilots damn well better know how do it without automation. When I learned to fly, GPS didn't exist. It was charts and wind reports and compass's. Its minor in comparison, but when shit fails you need to be able to get back on deck without destroying a $100m aircraft in the process... and ego's have to go out the door... this pilot probably knew he was in over his head at some point and should have broken of the approach to reassess his situation and redo his checklist in the process.
ReplyDelete"this pilot probably knew he was in over his head at some point and should have broken of the approach"
DeleteYou've missed the point of the post. The pilot failed to recognize and react in time BECAUSE HE WAS DEPENDENT ON AUTOMATION AND JUST ASSUMED IT WAS WORKING. He had become dependent and, therefore, manually deficient.
CNO,
ReplyDeleteWhat do you think of doubling the length of an aircraft carrier hull? Long enough where catapults and possibly arresting wires aren't needed. Then there is no need for elevators because there will be plenty of space for ramps. Delete many of the extraneous features like the fancy radars. A Very Large Crude Carrier is roughly the size of a Nimitz class carrier and costs $150 million with a double hull. So the actual hull could be relatively inexpensive and you'd have a lot of empty space you could fill with fire retardant closed cell foam or something around the edge of the ship to protect the crew from fire and flooding. Theoretically this could be cheaper than an existing carrier by trading steel for complexity.
Are you Austin Vernon?
DeleteYes
DeleteThanks for the 'grumpy' label ! To be fair, it's not totally inaccurate although I tend to reserve the grumpiness for lower quality comments. :-)
DeleteTo answer your question, tell me, what is the ballpark takeoff distance for a combat loaded F-18?
DeleteI like your energy focused articles!
DeleteIt looks like somewhere in between 900 meters and 1200 meters and it the record for longest ship is ~460 meters. Some of the modern crude carriers are 400+ meters long. Of course there are ways to shorten that but entail other tradeoffs or operational changes like tanking right after takeoff (which would require a carrier tanker with extremely low stall speed).
DeleteThanks for reading! Energy is definitely my expertise with the military stuff being more of a side thing.
"Energy is definitely my expertise"
DeleteAnd it's evident! Well done. Perhaps an analysis article on your site of energy (fuel) logistics for the Navy relative to the anticipated operating are of the western Pacific? Number of bases (very few!), size of logistics fuel fleet (very small), location of energy/fuel sources in the Pacific, combat energy sustainment capacities, prospect for successful energy logistics in support of a China war, etc.
If you ever opt to do something along those lines or of general naval interest, feel free to advertise it here!
"looks like somewhere in between 900 meters and 1200 meters"
DeleteSo, given that a carrier is about 330 meters long, would doubling the size even allow operation of aircraft without cats and arresting gear? Also, bear in mind that a combat weapons load makes for a heavy aircraft which makes for longer takeoff runs.
Structural/construction issues aside, it doesn't seem as if even a double-long carrier could operate aircraft but what's your thinking on it?
Yeah, let's say triple the length of a ULCC. That is still around $450 million worth of basic labor, steel, and engines. You'd need a new dry dock, which would cost a few billion. Then how austere do you go on crew. If it was wartime you could do laundry off ship, use chemical toilets, have them eat MREs, and basically sleep in rooms on cots. The crew areas would be small enough in volume you could have them in the middle with 25 meters of foam and steel compartments on each side. There would be plenty of room on the flight deck below for all the aircraft and lots of crumple zone between the flight deck and the hangar. It'd basically be a flat top with ramps on each end. But now taking damage would be like repairing an airfield because its only steel and some foam.
Delete"It looks like somewhere in between 900 meters and 1200 meters"
DeleteKeep in mind that the distance goes like the square of the takeoff speed, and that includes the speed of the ship, so if the ship is going 30 knots into the wind, the distance on the deck will be shorter, since the ship is moving too.
Watching old YouTube documentaries about the Royal Navy, it has always struck me how seriously take their training. In their training schools they force the students to use the most primitive equipment possible.
ReplyDeleteIn fact, the more advanced the training, the more primitive the tools. For example, watching the documentary on their Perisher Course (The course all Attack Submarine officers have to pass before they will be considered for command), the students are reduced to visually measuring the height of the ship and doing a trig function in their head (no calculators allowed) to find the range as they create a targeting solution.
I guess their theory is if you are a badass riding a tricycle, you'll be even better on a 1000cc motorcycle. I think they might be right.
PS: Another great example is their grueling Navigation schools, as described by Mitch McGuffie in his article, "A Rude Awakening" at USNI Proceedings.
I am a long-time proponent of the types of training that the Royal Navy does. I think something like Perisher should be required of all PCOs of all ships, not just submarines. Interestingly, in their system, if you go to Perisher and wash out, you are ineligible for any future duty about submarines.
DeleteI also liked their SpringTrain series of annual exercises around Gibraltar. It was actually quite popular in the RN, because if you've spend the winter in home ports or at sea around the UK, then a trip south to Gibraltar sounds pretty good. The concept behind SpringTrain sounds a lot like the old USN Fleet Problems between WWI and WWII. Those prepared the fleet pretty well for the war in the Pacific.
Sandy Woodward writes in his book that he had just finished SpringTrain with a bunch of RN ships when Argentina invaded the Falklands, and because they were already at Gibraltar a number of them were selected to go south to the Falklands. He suggests pretty strongly that the training just completed in SpringTrain had them well prepared for the Falklands conflict.
Another thing they do is FOST (Flag Officer at-Sea Training, IIRC). This sounds a lot like our old Gitmo training.
I would strongly recommend a combination of Perisher, FOST, and SpringTrain for the USN.
However, automatic landing is very important on operating large drones on carriers.
ReplyDeleteI thought one off the "advantages" of the F35 was alot of the training is done on simulators. So shouldn't that training be done with different systems turned off, so pilots:
ReplyDeletea. get used to doing manoeuvres with different systems turned on/off.
b. Get used to checking what is turned on and off.
also the advantage of simulators is no one dies so extreme simulations can be practised. eg arrestor wires not picking up the aircraft so a fly around is required, Bad weather and bad visibility when landing etc.
Clive F
"simulators"
ReplyDeleteThe term 'simulator' is a misnomer. A simulator CANNOT simulate the reality of the stress in a pilot undergoing g-forces, seconds away from either an abrupt landing or a fatal crash. All the time in the world spent inside a comfortable simulator setting - with a coffee close to hand - with zero chance of a bad outcome (hit the reset button!) will not prepare a pilot for the real world of heaving decks, bad weather, g-forces, a postage stamp size landing area, and imminent death if anything goes wrong.
Simulators have a purpose (establish or review procedures) but they CANNOT take the place of actual flying and, indeed, statistics show that as simulator time increases, so too do mishaps as actual flight hours are replaced by simulator hours.
My post got featured on Marginal Revolution, congrats to your platform!
ReplyDeletehttps://marginalrevolution.com/marginalrevolution/2023/02/which-weapons-in-a-major-conflict-can-be-most-easily-scaled.html
Congratulations!
DeleteFor other readers, this comment links to an article about weapons production/usage and is authored by user 'Austin Vernon', of this comment, and is well worth a read. Here's a direct link to the article: "The Weapons That Win World Wars"
Check it out!
A good read!!!
Deletehttps://www.youtube.com/watch?v=owWU3VaMxY0
ReplyDeleteViewing this former Hornet pilot's review of the crash report, I was struck by his observation and opinion that the issue was that the pilot thought he was in DFP mode, where the aircraft automatically adjusts throttle based on the aircraft's angle of attack, when he was actually in manual mode, and his opinion that if the pilot knew he was in manual mode, he would have reacted differently.
Mover draws attention to one of the recommendations of the report, which calls for more significant landing mode indicators to the pilot, so that the pilot knows exactly what mode he's in, so he can act appropriately.
What I was really surprised about was his comment that DFP is not a new technology, and that it's present on Hornets and 737s - so that means it's been around for at least 30 years.
Yeah, the guy thought he was in DFP where pushing the stick forward adjusts glideslope forward, typically by increasing power, as opposed to other flight modes where forward stick tries to push the nose down, typically by deflecting the stabilators. His control inputs were thus counterproductive for a while as it took him too long to realize which mode he was in while trying to rectify the problems with his approach, and by the time he figured out which mode he was in it was too late to prevent the ramp strike.
DeleteThis reminds me of the F-35A crash at Eglin, where the pilots control inputs were discarded by the flight computer (yanking the stick back and increasing the throttle in an attempt to avoid a hard landing/crash) because the F-35A didn't like his parameters and took him out of landing mode without him realizing it, and then further decided his inputs didn't make any sense and put the control surfaces to a neutral position.
Overall, my gut feeling is the F-35 has too many flight modes with qualitatively different association between control inputs (stick, pedals, throttle) and desired control outcome, not to mention what control surface actuations it's performing to get there. The lack of indication may exacerbate the situation, but it doesn't create it.
Eg. imagine a car where, when cruise control is engaged, pressing the gas pedal increases the desired cruise control speed by 1 mph/second and pressing the brake pedal decreases the desired cruise control speed by 1 mph/second, and the gas and brake pedals retain their normal operation when cruise control is disabled. Does one have to be an automotive engineer or a credentialed interface designer to realize that this design is going to cause crashes, and that there is a reason no other car's cruise control works that way, regardless of how one views the merits or drawbacks of cruise control more generally?