Back to previous post: Awwwwww…

Go to Making Light's front page.

Forward to next post: Things I wish I had in my kitchen

Subscribe (via RSS) to this post's comment thread. (What does this mean? Here's a quick introduction.)

June 4, 2003

Hot jets!
Posted by Teresa at 12:16 AM *

Bruce Simpson, whose jolly My Jet Engine Projects page is appealingly subtitled Anyone can build a pulsejet or turbojet engine, and has articles which begin like this one:

If you’ve always wanted to build a pulsejet but have no access to expensive welding or machining equipment then I’ve come up with a new design that anyone can build with just a few simple hand tools and readily available materials …
has come up with something even more splendid:
A DIY Cruise Missile
Watch me build one for under $5,000

Some time ago I wrote an article in which I suggested that it would not be difficult for terrorists to build their own relatively sophisticated cruise missiles using off-the-shelf components and materials.

Not surprisingly, that piece has produced a significant amount of feedback from the tens of thousands of people who have read it so far.

Included in this feedback, I’ve received quite a number of emails from former and currently serving US military personnel who acknowledge that the threat is one they are very much aware of and for which there is little in the way of an effective defense available.

However, there have also been a number of people who claim I’m overstating the case and that it’s not possible to build a real cruise missile without access to sophisticated gear, specialist tools and information not readily available outside the military.

So, in order to prove my case, I decided to put my money where my mouth is and build a cruise missile in my own garage, on a budget of just US$5,000.

I like to think of this project as the military version of “Junkyard Wars”.
You can look at his project objectives here. (via Beth Meacham)
Comments on Hot jets!:
#1 ::: Alter S. Reiss ::: (view all by) ::: June 04, 2003, 04:06 AM:

I was actually a bit relieved to see that the person trying to do this appears to be a New Zealand resident.

While the current administration would be particularly peeved, I'm not sure any US administration in recent years would have been entirely sanguine about a private citizen owning a cruise missile, let alone describing how to build cruise missiles for relatively little money.

#2 ::: Charlie Stross ::: (view all by) ::: June 04, 2003, 05:07 AM:

Ahem: I seem to recall a news article about five years ago (not sure of the source -- AP, I think, via a British newspaper) saying that the BATF had raided a guy in Northern CA who had, it seems, purchased a pair of SCUD-B missiles and a launcher-erector second hand from an ex-Warsaw Pact country hard-up for hard currency. They were within flight range of San Francisco, although I suspect he was more interested in using them as lawn ornaments ...

#3 ::: Teresa Nielsen Hayden ::: (view all by) ::: June 04, 2003, 07:38 AM:

Well, Northern California, what do you expect?

#4 ::: Claude Muncey ::: (view all by) ::: June 04, 2003, 11:38 AM:

Well, (as a Central Californian) I would expect him to have the good sense to get missles with the range to hit Southern California . . . or Nevada, at least.

#5 ::: Dave Bell ::: (view all by) ::: June 04, 2003, 06:35 PM:

The owner of Budge, a British Construction firm, accumulated a fairly spectacular collection of armoured vehicles, from various sources, before his company went bust. Including a SCUD system. There was some quite unusual stuff.

This was in the Eighties, and there were rumours going around that there were depots in East Germany full of Panther and King Tiger tanks, ready to be used by their reservists, just as the Russians had lots of T34 tanks.

#6 ::: Robert L ::: (view all by) ::: June 05, 2003, 07:43 AM:

Doesn't Tom Clancy own his own tank? I remember hearing that...

#7 ::: Barry ::: (view all by) ::: June 05, 2003, 09:34 AM:

Actually, there are theoretical defenses against small cruise missiles (it gets harder as things get smaller, of course).

The theory is to take a vehicle, put a powered turret on it. In the turret there is an array of heavy machine guns/automatic cannon (from 2 - 4). There is also a radar set, and perhaps a thermal imaging system. Within the vehicle body there is a computer.

They way that this proposed system would work is that, when the cruise missile came over the radar horizon, the vehicle's radar would detect it. The vehicle's radio and computer would sent a code signal, while alerting the crew and turning the turret so as to face the guns towards the missile. If the missile didn't send back the right response code, the crew would be alerted again. Meanwhile, the computer would direct the turret/guns so that the trajectory of the shells would intesect the missile's trajectory. Upon crew authorization, the computer would fire the guns. The shells would explode on/in/near the missile, destroying it.

I'm working on a patent for such a system. I'm going to call it 'AAA', to make sure that I get first billing in the phone book :)

#8 ::: Dan Zlotnikov ::: (view all by) ::: June 05, 2003, 11:37 AM:

The theory is nice, Barry, but there are a few minor problems.

First of all, the minor issue of speed. An object moving at 167 m/sec doesn't really give much time for operator response. The range of a modern AA cannon is about 8,000m, and that's maximum range. So, let's assume for a moment that the cruise missile is, in fact, passing at zero altitude and in a straight line heading along the line of fire.
Time in range is 16,000m/167m/sec=95sec. Not too bad.

Second, the crossection of a much more massive missile, the Tomahawk, is 52cm, but the Tomahawk carries almost half a ton of explosives as standard payload. It's also nearly 6m long. A home-made cruise missile, with a mere 10-100kg payload, would not need to be so wide, nor so long.

Then, we address the issue of height. The whole radar horizon idea is great, as long as the missile is moving at a reasonable altitude. With a tiny bit of navigational programming, a missile can be fired to fly at 200m altitude. The Tomahawk, once again, can fly at 50m and dodge hills. I don't expect anything as sophisticated from a home-made missile, but 200m would drastically cut down on radar detection range and not involve too much in the way of hills. But for a ground AA crew? Terrain's a problem.

Third, there's the assumption that a missile is totally dumb, and won't react to an IFF ping bouncing off its hull. It's not very hard to tell it to start wobbling a bit in its course the moment a radar signal bounces off it. Good luck hitting it then.

*all of the above is speculation by the author, without much technical investigation. Feel free to refute*

Scientific calculator

#9 ::: Teresa Nielsen Hayden ::: (view all by) ::: June 05, 2003, 01:12 PM:

Did you catch the size of its payload? That's the smallest size that will carry a nuke.

Or so I'm told.

#10 ::: Graydon ::: (view all by) ::: June 05, 2003, 03:08 PM:

10 kg is a very good nuke, though; I don't think anyone but the US ever built one that small and the manpack stuff is all out of inventory.

So, yeah, possible, but not likely unless the USG decides to build nuclear cluster bombs or something of that character and someone manages to steal a single submunition.

#11 ::: Dave Bell ::: (view all by) ::: June 05, 2003, 07:03 PM:

I am not sure that the particular design reported on the web page will work. I'm sceptical about the strength of the structure.

On the other hand, even in open country, it needs to be _very_ flat before a 600-knot target at 200m is easy to spot at any great range. The RAF don't do as much ultra-low-level stuff as they used to, but that's comparable to the sort of height the USAF flies at when they do low-level over here.

And for terrorism you don't need to aim for a specific point target, such as the Oval Office. How many places are there on a Saturday with large crowds and live TV? How many places would need point defence AAA?

#12 ::: Graydon ::: (view all by) ::: June 05, 2003, 07:50 PM:

Firing the artillery does a lot of damage; argueably more than a missile that size could cause with an explosive warhead.

These things really are a good candidate for look-down, shoot-down energy weapons, which are (just) doable but the cost is many orders of magnitude greater.

#13 ::: James D. Macdonald ::: (view all by) ::: June 05, 2003, 08:28 PM:

Barry, you're trying to invent the Vulcan/Phalanx CIWS.

I won't go into detail about why no one leaves those things switched on.

#14 ::: James D. Macdonald ::: (view all by) ::: June 05, 2003, 08:35 PM:

Oh, yeah, and the T-34s. They eventually wound up in Nicaragua, which is about the world's worst tank country. Remember the infamous 5,000 Soviet combat troops who were supposed to be in Nicaragua during the time of the Contras? 4,999 of them were tank mechanics trying to make the darned things work.

#15 ::: Erik V. Olson ::: (view all by) ::: June 05, 2003, 11:26 PM:

I won't go into detail about why no one leaves those things switched on.

I will!

The problem, of course, is the timing loop between detection, ID, and resolution is too long -- by the time the smart humans have decided this is a bad thing, the missle's struck. 600 knots may be slow by many standards -- but you can cover a lot of ground at 600 knots -- almost 1000 feet a second, as a matter of fact.

So. The Phalanx and Vulcan CIWS (Close-In Weapon Systems) break that problem by a simple trick. They drop the "ID" part of it, simply assume *anything* they're tracking is a bad guy, and point at it and try to fire. If the humans are smart, they leave the saftey on. If not, well, the first radar points the gun at the thing in the sky and fires. In the case of the Phalanx, it fires a 20mm Vulcan automatic cannon -- a 7 barrel motorized gatling gun that fires 60 rounds a second. (Fun Navy Trick -- load the Phalanx with 1-1 shell-tracers, rather than the normal 1000-1, and test fire. Look, a laser!) This basically creates a stream of lead in in the sky. The second radar tracks this stream. The computer integrates the two tracks, and sweeps the stream of shells into the target.

It's very effective against missles. It's downright deadly against anything slower. And, it's very automatic. Leave this on in port, and the first plane trying to land at the nearby airstip dies. It has been nicknamed "R2D2 with a hard on" -- and that's exactly what it looks like.

The USS Stark, the frigate nailed by the Exocets in the Gulf, was caught with its one Phalanx offline for repairs -- and was hit at an angle the Phalanx couldn't cover. (These Figs only had one CIWS, covering the stern and beams. Bigger ships have multiples CIWS installations, with 360 degree coverage.)

Speaking of Figs. The USS Clark, FFG-11, visited Chicago, mooring at Navy Pier on a goodwill tour. Then, a huge storm swept over the lake. The bad thing about this is that these storms will, quite literally, force water out of Lake Michigan, into the other lakes. Suddenly, the Clark drawing 18' of water, finds itself tied to a pier -- in 15' of water. Oops. I was talking to the Master-at-Arms. "I hope you like Chicago -- or that your ship has wheels." "Nope. Chicago's great -- never been to a big city before. And we're not leaving any time soon -- we can't even turn the screws."

A week long goodwill stop turned into three months. Finally, they had 17' of water, enough that a USCG buoy tender could come up, fasten a hawser, and drag the Clark into deeper water.

Got to watch them lakes. They're clever.

#16 ::: Erik V. Olson ::: (view all by) ::: June 05, 2003, 11:44 PM:

And, in doing a little more research, I find this...

The Clark was commissioned in May 1980 and decommissioned in March 2000. Following its decommissioning it was transferred to the Polish Navy where it was re-named the ORP General Pulaski.

The fact that a frigate stuck in the mud in Chicago for three months is given to the Polish, and named the General Pulaski is perfect beyond words. For one thing -- the largest city in the world, by Polish Population, is Warsaw. The second largest is Chicago -- which celebrates, in a large way, Casimir Pulaski Day.

Google up General Casmimir Pulaski, next time someone tells you a Polish joke.

#17 ::: Dave Bell ::: (view all by) ::: June 06, 2003, 06:03 AM:

There are stories about Phalanx trials...

Yes, it can be rather indiscriminate.

And the trials which are often quoted for its effectiveness had the system mounted on a barge with almost no freeboard, getting the radar lower than the targets, so it was picking them out against a sky background.

Put it on a real ship, and it looks down on the incoming missile. This reduces the effective range a bit by making the missile harder to spot by radar.

#18 ::: Dave Bell ::: (view all by) ::: June 06, 2003, 06:11 AM:

I'm a little sceptical about the apparent design of this DIY cruise missle. Foam-core GRP structures are strong and light, but how will they stand up to the stress at 600 knots, low-level?

That is a rather harsh environment for aircraft: it's not smooth air. If the design is slightly wrong you could get all sorts of problems with flutter and other instabilities and feedback through the control system.

No, I'm not an expert. It might be possible, but would it be viable as DIY? Still, look as some of the late-war German schemes. Plywood winged jet fighters!

#19 ::: James D. Macdonald ::: (view all by) ::: June 06, 2003, 07:23 AM:

Other things that R2/D2 has evaluated as incoming missiles and shot at: the rotor tips of an incoming helicopter carrying the mail and movies; and a galley fan of a ship steaming alongside in formation.

An early problem was that the CIWS would only shoot down one incoming missile. It would shoot down the first one, and its pieces, and the pieces of the pieces ... but the second missile would get through because by then the weapon was out of ammo.

#20 ::: Erik V. Olson ::: (view all by) ::: June 06, 2003, 08:05 AM:

Other things that R2/D2 has evaluated as incoming missiles and shot at: the rotor tips of an incoming helicopter carrying the mail and movies; and a galley fan of a ship steaming alongside in formation.

Among other targets. The fire control logic was very simple -- if it's moving faster than foo, and it's moving generally towards the ship, it must be A Bad Thing. Spinning blades have this neat property -- the tips will be moving fast (sometimes supersonic, in the case of a helo) and for part of the arc, the radar return will look like it's heading right for you.

An early problem was that the CIWS would only shoot down one incoming missile. It would shoot down the first one, and its pieces, and the pieces of the pieces ...

The very first fire control tests had an issue with the CIWS declaring its own shells as threats. Turns out the programmers didn't think the shell returns would be strong enough to really be tracked by the search/target tracking radar, which operated at a much lower ERP than the shell tracking radar. They were wrong (many of the Mk15's problems were cause, paradoxically enough, by that radar working far better than originally designed.) Easy fix, though -- anything with a "negative" velocity (heading outwards) wasn't a Bad Thing.

The above problem was similar -- the target radar worked too well, and could pick up tiny things like bits-of-shells. The solution here was theoretically easy, but hard in practice. Any target that slows and diverges after convergence is Not A Threat. Problems with this approach are left as an exercise for the student.

but the second missile would get through because by then the weapon was out of ammo.

Which takes almost no time to happen. The orginal versions fired at 3000 rounds a minute -- but the magazine held 990 rounds. This gives you 1/3rd of a second continous fire. The newer version have a 1550 round magazine -- but a 4500rpm gun, giving you 1/3rd a second.

Folks wondering about this must remember that the Phalanx is designed as a last-gap defense measure. Incoming threats are to be engaged by other means while they're further away -- the system is designed to fight off the 1-2 missles that make it through, not waves of them. Fire 20 cruise missles at the Stark, steaming alone, and it'll sink -- R2 may splash a few, but after 1/3rd of a second, R2 can't help you anymore.

For those wondering why we call it R2? Just look.

#21 ::: James D. Macdonald ::: (view all by) ::: June 06, 2003, 11:41 AM:

More Navy fun:

While pierside in San Diego, take the ammo tray out of your Vulcan Phalanx, and switch it on. Watch it track the aircraft landing at San Diego airport, and lock on, and chatter as it tries to fire.

But do remember to take out the ammo tray first.

Which reminds me of the only time I saw a Flash Override message. A ship accidentally launched a live warshot Harpoon in the Carribean.

Lots of people switched on their CIWS that day, I betcha. (Didn't hear about any unexplained missing ships immediately afterward in the area, so I guess it didn't find anything.)

(Note on 'Poons: the RGM-84 Harpoon is an anti-ship cruise missile, active/passive homer, which will fly out on a bearing, find the best target along that line, and hit it.)

Other great moments in Naval History: the WRL-1 (whirly-one) is an electronic warfare suite which relies on a man with a set of headphones listening to the electronic signatures of things to identify missiles. It's very accurate, but slow. This was replaced by the SLQ-32 (Slick thirty-two) which is fully automated. In order to reduce clutter on the screen, it doesn't report friendly missiles (as if a missile in the air had 'friends'). Thus, if you shot at a US ship equipped with SLQ-32 with a 'Poon that you bought surplus somewhere, the first notice they had that a missile was inbound would be the explosion midships.

#22 ::: Jordin Kare ::: (view all by) ::: June 07, 2003, 02:59 PM:

Erik: I think you mean 1/3 of a minute... even Phalanx doesn't run through it's ammo in less than a second. (ObSF: It's still a stitching gun, not a splat gun. Oh, wait,this isn't RASFF. Never mind.)

On R2D2: Raytheon (IIRC) in the late 80's actually had the nerve to develop an optical beam director for SDI (aka Star Wars) under the official name of the Rapid Retargeting Precision Pointing system. Yes, R2P2.

On intercepting cruise missiles: The trouble Dan's "AAA" is that unless you know in advance what the target is and can put the defensive system right next to it (which is what Phalanx does) you need an enormous number of ground-based radars (and guns) to defend any significant perimeter.

On homebrew cruise missiles in general: I haven't signed up for the site, so I don't know details of this fellow's design, but there's a much simpler way to make a perfectly adequate cruise missile with a much larger payload, at marginally higher cost: buy a used Cessna, put a GPS and a laptop in the pilot's seat, and add some servomotors on the controls. It's slower and has a higher radar cross section than what I gather this fellow is building, but based on past data that won't be a problem.

What past data? Well, notably the fact that Matthias Rust was able to fly a Cessna through the world's tightest air defense net and land in Red Square. Gathering additional evidence from U.S. drug interdiction efforts is left as an exercise for the reader.

Back when Rust made his little jaunt (May 87) there was much argument going on over the cost of the B2 bomber. For amusemement's sake, I figured out how many Cessnas you could buy for the cost of a B2 -- roughly 10,000, depending on what assumptions you make -- and proposed replacing the B2 as a strategic strike weapon with a *very* large number of "cruise Cessnas" parked along the German Autobahns....

#23 ::: Dave Bell ::: (view all by) ::: June 07, 2003, 05:26 PM:

A V-1 with GPS.

I wouldn't be surprised if detailed plans for the V-1 were on the web. You just have to replace the crude 1940's guidance system. Getting a pulsejet up to flying speed is the problem.

But you can get bird-scarer rockets which fly about a third of a mile, and then produce a loud bang. Set off a salvo of those near some public event... Terror isn't always high-tech.

#24 ::: James D. Macdonald ::: (view all by) ::: June 08, 2003, 07:42 AM:

Getting a pulsejet up to flying speed shouldn't be too tough if you have a pickup truck and a stretch of open road.

Smaller type (our default)
Larger type
Even larger type, with serifs

Dire legal notice
Making Light copyright 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017 by Patrick & Teresa Nielsen Hayden. All rights reserved.