Hacker News Reader

Comment by olympus | Hacker News Reader

Missiles go extremely fast. Like stupid fast. Most ABM systems today use hit-to-kill, which means that you actually have to make contact with the missile to kill it. Physically hitting something that small and fast from a long distance away is hard even if you know exactly where it's going to be. Terminal guidance needs to be very accurate, since in the last second the missiles are a mile apart. With an update rate of about 10 hZ, you have to make your final correction that is accurate to about two feet (the diameter of the missile) over about 1/10th of a mile. That's tough.

Most other missile systems (like surface-to-air or air-to-air systems designed to kill planes) have proximity fuses, and don't need direct contact, they just blow up nearby and throw shrapnel into the plane. That's enough to kill a plane, but doesn't work very well to knock a ballistic missile off it's trajectory.

View on HN

To add to this, the closing speeds are so fast that you would need to detonate explosives far in advance for anything to happen anywhere near the target. If you detonate at the usual anti-aircraft distances (a few meters), the time it takes for the detonator to fire and for the shock wave/chemical reaction to propagate though the explosive mass would leave the shrapnel cloud way behind the target.

Wow, I hadn't actually thought of this. So, I assume that's why you go for hit-to-kill kinetic warheads, then, and just solve the accuracy issue rather than accuracy plus tricky timing issues?

That's part of it. There's also a point where the kinetic energy of impact surpasses the chemical energy of explosives for equivalent mass. This is around Mach 10 as a huge ballpark figure. So with intercepting high mach and ablatively shielded targets like ballistic missiles, it might be better to budget your projectile mass to increasing likelihood of a hit vs attempting to damage via a proximity explosion.

The UK Starstreak [0] missile seems to be trying for a compromise with a combination of relatively high speed (Mach 4.0) kinetic tungsten penetrator with a very small (again, relatively) 450g high explosive charge. It's an odd looking thing, because the sub-munitions aren't faired over at launch so you get this normal looking tube shaped missile with three tiny missiles stuck on the front. It seems they rely on proximity fuzing with time delay, so once the penetrator has breached the aircraft skin, causing a reasonable amount of damage, it waits until it knows its inside and then detonates for maximum effect. I guess three of the penetrators hitting your plane at M4.0 within a ~1.5m group is enough to make you have a very bad day, even if the explosive warheads turn out to be all duds!

0. https://en.wikipedia.org/wiki/Starstreak

Starstreak is a really clever design. They use a neat clutched rotation mechanism with skewed fins to get 2d attitude control via a 1d clutch signal. The US adopted a similar mechanism to build a smart fuse that can be used with existing dumb artillery shells to give them gps guidance at low cost.

I wouldn't think so. Hit to kill is going to have the same but worse timing issues as a proximity hit since the size is much smaller.

I would guess the real reason is that if the target is ballistic, you need to actually "kill" it, not just wing it like you can with an airplane. If you wing it the warhead will still fall in roughly the same place with a probably intact warhead. To be safe you want to actually destroy the warhead. Worse case it goes off at point of impact, which is better than on the target.

Yes, Scuds go Mach 5 (3800 mph) which is indeed stupid fast. That Patriot PAC-3 only goes Mach 4.1 but then it's not chasing its target down.

Can you explain why the update rate is 10 Hz? It seems orders of magnitude away from current tech, and even from tech of 20 years ago.

Perhaps I misunderstood what the update rate was, but I interpreted as the rate at which the "chaser" can update its trajectory or take action (kind of like the main loop running 10 times a second).

It's only "about" 10 hZ, on that order of magnitude. Indvidually you can have faster components and even though the process is pipelined, the last guidance correction that matters takes the sum total of all component times in the system. The seeker has to see the target is off-axis, then the computer has to calculate the proportional control input, and then the servos actuate the fins.

You also might be overestimating what "current tech" means in military hardware. The PAC-3 Patriots were entered into service from 1995-2000, meaning they were designed with late 1980's technology. The Saudis don't get to have the true top end hardware, the US won't sell their "best stuff" until they have "new best stuff" to beat it.

Whats our current "best stuff" in missile tech?

I think its the THAAD https://en.wikipedia.org/wiki/Terminal_High_Altitude_Area_De...

I wonder if the laser weapon systems will go anywhere in the end. Certainly seems easier to target, and cheaper to shoot. But there's still a lot of energy that needs to go into a missile to be certain it's disabled.

https://en.m.wikipedia.org/wiki/Laser_Weapon_System

An ICBM warhead is going to be a tough target for that. Very fast so you don't have long to heat it and fitted with heat shielding for reentry already.

It depends on what kind of missile you want to hit. A scud is very different from an ICBM, which are both quite different from anti-ship and some of the area denial weapons.

In perspective: The Russian Satan 2 is indefensible, and also the older SS-N-27 Sizzler shot from very nearby submarines. Not enough time to react on those.

There's really no such thing as an indefensible missile, just like there's no such thing as an impenetrable computer network. Difficult to defend against, those exist. Missiles with an extremely low probability that we can physically intercept, those exist. But defending against a missile starts as soon as a possibility of launch occurs, and you can bet that a lot of people spend a lot of time figuring out how to defend against "indefensible." They aren't perfect, but they aren't going to bat .000 either.

Well, Russians claim otherwise and Wikipedia confirms. It's of course highly contra-productive to tell such things the American public.

To add, more specifically to ICBMs, missiles also have counter measures. A common one in nuclear missiles is multiple and decoy warheads. So you really want to hit the missile, and before the warheads deploy. Which really sets a timelimit on when you can hit it. After the warheads deploy, they spread out pretty fast, making it much harder to hit with an explosion instead of kinetic force. Really you only have a few minutes to detect the launch, track, launch, and hit your target (generally the sooner the better). That's a real time crunch, which is one of the reasons they go STUPID fast.

Tldr: kinetic force is definitely the way to go.

Ballistic missiles are fast, but satellites are faster. In 2008 the US showed that an SM-3 missile can destroy a satellite [1]. this is the same type of missile that is deployed in the Aegis ABM defense system.

[1] https://en.m.wikipedia.org/wiki/RIM-161_Standard_Missile_3#A...

So I understand that it is quite hard to catch a missile when you fly in the same direction as the missle, but couldn't you just intercept the missile by flying towards it and park you own missle in the tradjectory of the incoming missile.

Why is that not possible?

That's exactly how Patriot missiles work. They just get in the way. The entire trick is, to be in the right place at precisely the right time.

It's still a very small target going very fast. That doesn't become easier to hit if you fly against it's direction of flight and thus have even higher relative speeds.

Reply on news.ycombinator.com