Comment by Gravityloss | Hacker News Reader

Comment by Gravityloss | Hacker News Reader

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.

Reply on news.ycombinator.com