Is the F-35 a Trillion-Dollar Mistake? | Hacker News Reader

Is the F-35 a Trillion-Dollar Mistake?

bloomberg.com

In a lot of ways yes.

It was likely a mistake to base a multi role fighter on a design aimed primarily the first strike role that were the first of the airfoces roles to be handed over to robots(cruise missiles), and it was most definitely a mistake to pick the one role where manned planes are least likely to be replaced by drones as the lowest priority roles.

The problem with the F-35 is that it's good at low level first strike missions and bad at visual range fighting weather it's dogfighting or close air support as it's what had to be sacrificed to make radar stealth possible, despite the fact that those are the primary roles where you cannot replace the pilot with a smart missile or drone.

And all of this in a plane that have the same crucial weakness that got the F-22 canceled, in that it's too expensive to build in number and requires way to much maintenance to fly as often as some of the less sophisticated airframes that can be pressed into service and do as good a job as the F-35 in the primary roles where you need to keep pilots in the battle space for prolonged periods, at a far lower pr hour cost.

In a lot of ways the US air-force is setting itself off for the same kind of failure as the luftwaffe had doing WWII where their advanced planes was bombed out of existence without ever getting in the air because, cost and maintenance issues prevented them from being in a constant state of in-air rotation.

In a lot of ways no as well. Take for example F-35B. That would've never happened on its own. Too few customers for a separate development project. Whereas now combined with amphibious assault ships it will be a quantum leap for USMC capabilities. Otherwise they'd have to retire Harrier for just V-22, AH-1Z and CH-53K.

It already costs less than its contemporaries Rafale and Typhoon. Remotely operated planes are not cheaper. Take for example MQ-4C Triton, which can only be used for intelligence and recon. That costs around 190 million apiece. Quite the sticker shock when you can get almost two F-35As for the same cost. Basically even if you could get similar capabilities (there are no air to air UCAVs atm), it will cost at least as much for quite some time.

According to pilots flying it, it's equal to clean F-16 within visual range.

F-35 program is also invaluable in utility. Nine countries became partners. Eventually most Nato-members and even some affiliates will adopt it. That's standardization of capability across the board on an unprecedented level.

Quick overview of test results: https://youtu.be/zgLjNsB_hyM?t=4m17s

The previous article linked had about 20,000 words. It spoke a lot about various problems.

Let's look at the gun for example.

There's a small door that opens so the gun can fire. Apparently for stealth reasons. The opening door creates an aerodynamic disturbance and the plane yaws or does something else and the aim is off. So that has to be corrected in the flight software - the plane must countersteer with the tail surfaces when the door is opened. And for some reason it's been very slow to create these rules.

Now, let's assume the F-35 program is canceled. There's a few scenarios after that.

Maybe some non-stealthy aircraft can be created for ground support missions. They don't need such pesky doors. On the other end of the spectrum, maybe some stealthy pure air superiority fighters or strike craft can also be created that don't need guns at all.

Or then you decide that you need a gun on a stealthy aircraft and you need a door again (or then you design it differently, AFAIK the F-22 doesn't have such problems). Will it be easier this time? If you think it will, what will be different? Why does it take years and years to develop software for F-35? Why wouldn't it take years and years, and actually longer for these new aircraft?

The gun problems are probably not caused by the "commonality between different versions", that's always trotted out as "the fatal flaw in the program", as it only exists in one version of the aircraft.

There might be some underlying reasons why progress is slow and expensive. But I don't think many of them would be fixed by procuring just a new aircraft. Maybe the department of defense could buy from Northrop or Boeing... but I doubt the results would be that different with a roughly similar process.

I would likely look at radical software development methods (would still have to be safety conscious). There are things like Skunk Works, Faster Better Cheaper, Agile Manifesto... two out of these three are from the aerospace world.

They don't last forever because the small group of competent people will always turn into a huge mass of mediocre people governed by a mass of bureaucrats. That's just nature.

For what it's worth, the F-22 has a similar system. Perhaps it's not such a problem because the aircraft is significantly larger.

I don't know why the cannon door issue is taking so long to rectify but I can guess: Whoever is responsible for the weapons systems can't fix this themselves, it requires a modification to the flight control law, which is handled by a different team and any modification to that system requires extensive testing. That team is likely to prioritize work items that actually impact systems they're responsible for, so the cannon languishes at the bottom of the queue, doing nothing but attracting a storm of blog commentary.

One ought to keep in mind when reading the DOT&E report that it's their job to tear the fighter limb from limb and take the program administration to task for every problem. Of course the general tone is going to be negative, we know the program is behind schedule and over budget. They still seem to be on track to build a multirole fighter that is at least tolerably competent at all of its assigned tasks (even CAS) and that's a win. Maybe it's not the best possible solution but it's what the Pentagon wanted.

I'll never understand the impulse to cancel programs when they're just turning the final corner to completion. I guess that's when they start to gain the highest public profile. There is no magical second system that will solve every problem, and at this point the thing to do is apply lessons learned from the procurement and R&D processes going forward, maybe plan a Block II to smooth out the rough edges. We're about a decade too late to flip the table in a fit of rage and start over.

But the program was massively delayed and technology has progressed dramatically. Why bring a jet to a drone fight? Or to a laser fight? Seems obsolete.

There are essential comms problems that will face any unmanned system that tries to take over the F-35's role. In the absence of a solution that is extremely low latency, immune to interference, and undetectable as radiated emission, you'd have to lean on computer AI to run the bulk of the mission autonomously. Nobody is ready to throw the switch on that.

A possible solution is to augment flights of manned fighters with unmanned drones as missile/bomb trucks. This gets you the force multiplier while keeping all communication short range and within line-of-sight. But you still need a stealthy, survivable manned fighter.

Overall it's a bit like saying why bother with a new generation of conventional cars when universal self-driving is right around the corner. Maybe it is, or maybe it isn't quite as close as you think. These are questions that are more appropriately asked of the N+1 generation fighter that may or may not be on the drawing boards at this time.

P.S. I've noticed the F-35 takes flak from both the extreme technological pessimists (drones/fighters can't take over the A10's role, low & slow manned flight is the only way to do CAS) and the extreme optimists (drones will make all manned flight obsolete next year). Seems to me neither faction has a very strong grasp of the state of military tech.

How would you fly a drone in a heavily jammed area?

Who says you can't replace the cockpit with a computer?

>There's a small door that opens so the gun can fire. Apparently for stealth reasons. The opening door creates an aerodynamic disturbance and the plane yaws or does something else and the aim is off. So that has to be corrected in the flight software - the plane must countersteer with the tail surfaces when the door is opened. And for some reason it's been very slow to create these rules.

This thing? https://i.ytimg.com/vi/AJhTDzeMYkI/maxresdefault.jpg

That "door" really struck me as absolutely insane the first time I saw it. I simply cannot understand how a competent engineer could have designed this

When you watch the associated video (https://www.youtube.com/watch?v=JD8mqs93y1s), The thing literally flaps up and down freely when the cannon is firing in static position on the ground. Could they seriously not design it to retract?

This is the first I had heard about the software compensation they added specifically to deal with the drag though. That just makes the entire thing hilarious.

Here is a really great write-up on the F-35 from Quora: https://www.quora.com/What-are-your-thoughts-on-the-F-35-Lig...

That was indeed a great write up. Conclusion it's not so bad.

Great read...you should consider submitting this link as an article

How useful are fighter jets? Shouldn't we be building swarms of unmanned drones instead?

It's a multi-role fighter aircraft, not simply air superioity. It needs to be able to carry air to surface ordnance

We're not quite in the sci-fi future yet Drone swarms exist, but mostly do intelligence, surveillance, and reconnaissance in the field. The perdix drones that the us military does have are powered by propeller, so so they can't keep up with a fighter jet that can go Mach 1.6 or carry or current supply of air to surface missiles. I believe that the way they would attack a fighter aircraft is turn themselves into Kamikaze missiles.

They're still working the bugs out and trying to perfrct their autonomous behavior...which scares the shit out of me as I write this. This is the only application of AI that causes me fear and anxiety. Thankfully, they don't have any weapons platforms yet

Basically in any reasonably capable enemy airspace, you need to fly with emissions control - ie. no radio, no radar else passive detection systems will find, triangulate and guide a SAM on you.

If you're doing drones, this means programming fully autonomous AI to identify and kill targets - which no one quite wants to do just now. For the same reason, "swarms" are tricky since you need radio coordination. Drones have proven useful against significantly less technologically capable adverseries but don't forget the drone captured by Iran by what appears to be control signal spoofing

>Basically in any reasonably capable enemy airspace, you need to fly with emissions control - ie. no radio, no radar else passive detection systems will find, triangulate and guide a SAM on you.

A cruise missile on final approach don't need forward radars or datalink back to base, and the same goes for almost any kind if missile used since the 2nd world war. Just because remote controlled drones don't work under heavy ECM don't mean the military don't have robots that can. And some of the US's drone use a directed signal towards an satellite instead of omni directional RF signals for control.

The Problem here is that stealth don't add a whole lot of value to the observe first then shoot kind of missions where you really do want an human in the pilot seat, so multiplying the cost pr air hour by a order of magnitude to get some level of radar stealth is the kind of economics that causes mission failures in real conflicts where real life supply problems dominate.

For the F-35 with it's dependence on BVR missiles for defense going silent also means that if it can be spotted by a ground observer(which it likely can) it faces a high risk if the enemy have enough time to scramble an interdiction patrol into visual range, a tactic that was perfected by the RAF doing the battle of Britain(using radars that could probably see the F-35).

You can forget the ground observer. Stealth is the reason why it can fly high all the time to avoid visual/audio detection. Non-stealth planes like F-15E are forced into High-Low-Low-High altitude flight profile. where low level ingress and egress will e.g. drastically reduce range. F-15E going around threats both ingress and egress uses about 30% of its fuel to do so. F-22 would use around 6% in comparison. So basically stealth means that you can waste less fuel on threat avoidance.

Yes. I expect manned fighter jets will become obsolete within 20 years. A few thoughts on the subject:

1. Drones can handle much higher G-forces than manned fighter jets. This is a huge advantage.

2. AI technology is progressing very rapidly so there will soon be no benefit to having a human in the aircraft. Even today the US military probably has AI that can beat any human in a simulated dogfight.

3. Pilots are very expensive. Not just to train but also in terms of the political cost of having pilots killed or captured by the enemy.

4. Eliminating the pilot opens up a range of options. For example, instead of a $200 million fighter jet it may be more effective to build 10 drones for $20 million each, or even 100 drones for $2 million each. Maybe half the drones will have stealth capabilities and half will have AWCS-like communication and control features.

5. During WWII German and Russian tanks were roughly equivalent. But the Russians could build tanks at a far greater rate than the Germans, making the German loss almost inevitable. Similarly, in future the country that can produce many drones quickly and cheaply may dominate air warfare. In this regard China will likely have an advantage over the US.

6. The military-industrial complex is a huge bureaucracy in which many people have a financial and psychological incentive to continue development of manned fighter jets. For this reason I expect such development to continue well beyond its use-by date, probably until there is a real-world battle where a large, previously dominant air force is soundly beaten by a small power using new drone technology.

3. Pilots are very expensive. Not just to train but also in terms of the political cost of having pilots killed or captured by the enemy.

In terms of ethics, there is a need to ensure that war has a human cost or there is very little to stop countries doing it all the time. If the cost of war is purely financial, then I think perhaps that might be the beginning of the end to the world as we know it.

As long as the US defense budget is bigger than the next 10 nations combined, might as well throw money around.

Just imagine if NASA's budget was that big what we could be doing instead.

Question: many people prefer the A-10 Warthog for ground support, and argue that the F-35 is both overpriced and insufficiently capable for the task. A big part of the reason is the A-10 cannon, which can be used to great effect on enemy targets very near friendly forces, and in a timely manner.

During the Iraq War I remember reports about about AC-130 loitering in circles above a battleground, presumably just out of the effective range of shoulder fired ground-to-air missiles. More importantly, a gunner would use a side mounted cannon on ground targets.

My question is: could be we replace the A-10's (and even the F-35's) ground support role with a similar gunship, but with much a much more accurate cannon; one that could reliably and consistently hit with, say, 1 meter accuracy. Do the existing cannon mounts on the AC-130 utilize accelerometers and mechanisms to compensate for the movement of the plane? Is there existing technology to use, say, lasers to measure wind and air turbulence to improve projectile accuracy? How high could a gunship theoretically loiter with improved technology? (I assume ground-to-air missiles will only get better, and cheaper, and more portable with time.)

Theoretically, such gunships could be flown fully autonomously, with a remote human doing targeting.

A-10 has been known to be a MANPADS magnet since the Desert Storm. If you have air superiority, you can keep them, use UCAVs or buy new Super Tucanos. But in contested environment they are useless. Multiroles or bombers with precision weapons are better suited for CAS. Cannons are legacy, which is why you haven't seen a uav concept matching your description. In the future targeting platform and the one doing the shooting might not even be the same.

"Part of the reason the A-10 has enjoyed the amount of success and notoriety it has is because there has been unopposed use of the airspace over Afghanistan, Iraq, and now Syria. It is not a contested environment. It is not a degraded environment. It’s not an operationally-limited environment–save for bad weather and some really big, nasty mountains in Afghanistan. All of that means our air assets have an ability to fly and operate unmolested in those AORs.

Tomorrow’s fight will NOT be that." https://fightersweep.com/2038/the-a-10-warthog-debate-a-fate...

Contested environments do not make the A-10 useless. In a "real" war, you expect to lose lots of people. Right now we are fussy, but think back to the world wars. People were dying left and right, and we accepted it as the cost of winning. Literally millions of people died.

The A-10 also happens to fly just fine with large chunks missing.

  Tomorrow’s fight will NOT be that.

All of our wars since Vietnam (after, but arguably during) have benefitted from largely uncontested airspace. That's been the case for over 40 years, and through many conflicts.

Why should we expect otherwise? The only conflict the U.S. might be engaged where that wouldn't be the case is an an all-out war with Russia or China, either on their territory or, like in Korea, an awkward proxy war where we're unable or unwilling to control airspace. (Hopefully we don't ever repeat the mistake of Vietnam, where we refused to control either the ground or airspace in North Vietnam for flimsy political reasons.)

It seems pretty dumb to spend hundreds of billions of dollars on such a theoretical conflict, especially given that so far the nuclear powers have managed to avoid such direct conflict.

In every last one of our conflicts with substantial American ground troops in the past half century, we've always controlled the airspace in short order. If it's ever the case where we won't be able to control the airspace, we'd have bigger problems than bike shedding this sort of technology. Any difference between the F-35 and the cheaper alternative (Super Hornet, A-10, etc) would be negligible.

I just don't understand wasting the money. I understand the principle is to plan for the next war, not the previous war. But after a half dozen previous wars with identical circumstances, and several similar on the horizon, at some point not taking _reality_ into account seems grossly short-sighted.

Yes, MANPADS may be getting better, but there are more effective and cheaper solutions to maintaining effective control of the airspace (excepting the above exception) than the F-35 or similar platforms.

See also previous discussion about this topic from 3 days ago:

https://news.ycombinator.com/item?id=14013203 (232 comments)

Story: "F-35 Continues to Stumble"

http://www.pogo.org/straus/issues/weapons/2017/f35-continues...

The underlying problem seems to be that they tried to satisfy too many requirements, and are now ending up with a mediocre product that does nothing well. At this point I would seriously be considering reducing the requirements. Tell the Marines to go find another plane. Maybe the Navy too. Focus on the big customer, the Air Force. Do whatever it takes to make it work for them. (Although I kind of suspect the Air Force would rather have the F-22.)

The Navy would probably go with the Advanced Super Hornet. God knows what the Marines would do.

The underlying problem seems to be that they tried to satisfy too many requirements, and are now ending up with a mediocre product that does nothing well

Ah, my old foe. We meet again. I have to admire your resilience. Sometimes I wonder what we could achieve if we were on the same side.

Maybe the Marines should take the A-10 from the Air Force since they don't give a rat's ass about infantry support.

This is surprisingly complex. I expected more points to clearly lean towards it or away from it. There are cost overruns and delays, but renegotiations happened and corrected at least some of that. The plane is real and works, but has weaknesses. It has tentative bipartisan support.

We built another Jack of all trades for Veitnam, the F4 phantom. Largely it was regard as a failure because it wasn't good at anything. Is the F35 advanced enough that not being the best still leaves the USA with better than everyone else and therefor less than the best is good enough?

> Is the F35 advanced enough that not being the best still leaves the USA with better than everyone else and therefor less than the best is good enough?

The other question is "and for how long?". Is it worth the price for a 5 or 10 year edge?

Is the F-4 really regarded as a failure? I thought it was a pretty big success. It certainly wasn't trouble-free. Infamously, the missile-only design ended up being way premature, but that was ultimately solved with a gun pod, and then a redesign to integrate a gun. Its combat record was good, it served in the US military for three decades, and it still serves in several foreign militaries today.

However, I'm far from an expert, and would love to know more if I got it wrong!

No, you're right. I suspect the plane squeaky was really thinking of was the F-111. The F-35's development process was reminiscent of the F-111, with features incorporated to create something for everyone, along with the concomitant weight and cost increases.

Yes, the F-111 would fit much better! Seems like the F-4 and F-111 occupy opposite extremes of the potential outcomes for a jack-of-all-trades aircraft.

> Is the F35 advanced enough that not being the best still leaves the USA with better than everyone else

I thought the F-35 was never intended to be the best, but rather the best the US is willing to sell?

That's kinda pedantic, but I'm actually meaning it as a question not a correction.

That's about accurate. F-35 is the best multi-role fighter the US is willing to sell to allies. The F-22 by comparison is the best air superiority fighter the US keeps to itself.

Does the world need F35...by the sounds of it f22 was good enough

This article violates Betteridge's law of headlines.

Actually it doesn't.

I think we need to compare the benefit the USA's population gets from this Trillion-Dollar spending with other possible program like free college tuition[1] who would around 70 billion/year.

I think democracies need a tool that help explore and explain what we gain from a law/program and what are the cost associated. But consequence and indirect cost is difficult to foresee: Obamacare (ACA) increased the number of people going freelance (author, artist, start-up ... ) because they were able to get insurance without a part-time job.

[1] https://www.theatlantic.com/business/archive/2014/01/heres-e...

Reply on news.ycombinator.com