The Nuclear Potato Cannon, Part 2 (2006) | Hacker News Reader

The Nuclear Potato Cannon, Part 2 (2006)

nfttu.blogspot.com

57 pointsajarmst58 comments

Dr. Brownlee himself (the guy who designed the experiment) said that the plate was vaporized by the atmosphere and never got to space.

Sorry for the buzzkill, folks.

https://en.wikipedia.org/wiki/Operation_Plumbbob#Propulsion_...

You're misquoting the article:

> Dr. Brownlee believes that the plate did not leave the atmosphere, as it may even have been vaporized by compression heating of the atmosphere due to its high speed.

Well it sounds like the the only reasonable, responsible course of action is to replicate the original test but use modern high speed recording equipment to determine what would happen to that steel plate.

How critical is the nuclear explosion in this experiment? Could you use an equivalent traditional explosive? 1KT explosion can be achieved without going nuclear.

According to the wiki page the Pascal-B test was the equivalent of 300 tons of TNT.

hold my beer

I'm not a physicist but as far as I know a nuclear explosion releases its energy much more quickly than e.g. TNT, so just releasing the same number of joules wouldn't exactly duplicate the test unless you were able to release them in the same number of nanoseconds.

This video https://www.youtube.com/watch?v=T2orn6T2QFM is a good DIY experiment along those lines and is interesting if not necessarily correct at the nuclear scale.

I'm curious, what speed is necessary for compression heating to be a significant factor? Obviously a regular rocket doesn't melt on its way into orbit, while it does need special protection for the trip back.

Rockets do actually heat in flight. But their launch profile is carefully chosen to keep the heating to managable levels. Liquid fuel rockets typically throttle down somewhere around a minute after launch, at a point called "max-Q", which is the point of maximum aerodynamic pressure. They throttle back up again once they have gotten above enough of the atmosphere.

As a point of reference, the SR-71 was one of the first major aircraft programs to use titanium extensively, the reason being that at the speeds the SR-71 flew, aluminum couldn't maintain structural stability due to the heat. And the SR-71 was "only" a Mach 3 airplane, flying at "only" 60,000 to 80,000 feet. It's fuselage would reach nearly 600 degrees.

An orbital launch vehicle typically reaches Mach 3 very roughly that same altitude, or somewhat higher (the figure I can find is Mach 4.5 at 28 nautical miles, which by interpolation I'm guessing is Mach 3 at between ten to twenty nautical miles; ten nautical miles is roughly 60,000 feet).

Another bit about the SR-71 (and supposedly concordes) is that their structures are designed with tolerances based around the forces they experience at speed. The effect of which is that when sitting on the ground not going anywhere there are gaps between some parts, and others are tightly compressed together (and then separate while traveling at speed). I don't remember if it was the last concorde, or one of the SR-71s, but there's a story out there about the pilot on its final flight sticking his hat into a gap in the cockpit, which on landing promptly closed up permanently sealing his hat in place inside of the cockpit (he did it on purpose to commemorate the last flight). Similarly there are stories out there that claim the SR-71 leaked fuel like a sieve while sitting on the ground because it needed the atmospheric heating from going supersonic to seal up all the various fittings in its fuel system.

I do recall seeing an image of an SR-71 on the ground with pans under it in various places, supposedly to catch leaking fluids.

Ironically there's a series of Russian dive watches that does approximately the same thing.

Instead of attempting to overengineer a high-depth tolerant system that's still sealed in open air, it's sealed more tightly as pressure increases.

An elegant weapon from a less civilized age.

https://en.m.wikipedia.org/wiki/Amfibia#Design

Here's the reference to the pilot's cap:

http://www.seattlepi.com/business/article/Concorde-s-arrival...

Any SR-71 story usually comes with the obligatory story about the Cessna v navy ground speed check. I'm replacing that story with the story of the time an SR-71 went very slowly.

https://www.google.co.nz/amp/foxtrotalpha.jalopnik.com/the-s...

Thanks! Tyler Rogoway can certainly relate a story.

IIRC they would refuel mid-air shortly after take off for long range missions.

Basically a rocket tries not to accelerate too much until it's above most of the atmosphere. (Which isn't as far as you might think - just a few tens of miles up.)

i.e. it climbs, but does not try to gain much horizontal velocity (i.e. the kind necessary to orbit).

On the way back the heat is just a way of shedding energy. So they shed energy "slowly", to keep within the thermal limits, they do that by shedding energy while still relatively high up.

This simulation of a Falcon 9 launch shows the heating rate peaking at 11 kilowatt (!) per square meter around maximum dynamic pressure.

https://youtu.be/oxtAa7QCbxw

Interestingly those kind of speeds could put it on the speed record list (if it was a vehicle) https://en.wikipedia.org/wiki/List_of_vehicle_speed_records#...

In grad school my advisor (who was teaching "Launch and Entry Systems") told us this story. A few lectures later he went over rail guns as a launch system, and mentioned that by his back-of-the-envelope calculations, a slug of titanium the size of a telephone pole, launched at earth escape velocity by a rail gun at sea level, would ablate approximately a third of its length away in the first thousand meters, or something close to that (it's been a long time).

Actually calculating the heating of a steel plate launched at that speed is a significantly nontrivial problem, because its aerodynamics are unstable. In other words, it would almost certainly tumble in flight, in an unpredictable manner. There's no guarantee that it would even fly in anything approximating a straight line. Even figuring out the effective coefficient of drag would be really hard. Figuring out the thermal profile would be even harder.

But, as several other comments have mentioned - no matter how you slice it, it would be subject to a hell of a lot of heating, and it's a pretty safe bet that it ablated completely away and did not reach space.

I'm familiar with this incident, but never saw this article before. I've seen articles such as http://nuclearweaponarchive.org/Usa/Tests/Plumbob.html#Pasca... and http://nuclearweaponarchive.org/Usa/Tests/Brownlee.html (the latter being written by Dr. Brownlee, the scientist who designed the test in question).

This article seems like an independent telling of these events, along with an independent recalculation of the results in the first link, but based on conversations with the same person, which is kind of interesting.

As far as the first man-made object in space, there's no real question that this was not it. There are two major problems: one is that the steel plate pretty much certainly vaporized in the atmosphere before it reached space. The second, even bigger problem is that the first man-made object in space was a V-2 rocket launched in 1944.

That V-2 rocket was indeed the first thing in space, but there's a distinction to be made: it only went suborbital. If our big steel friend wasn't tragically vaporized in the atmosphere, it might just have left the earth's gravity well--maybe even the solar system's well (escape velocity of 44km/s), which is in part why that incident is so cool!

Fun fact: the first ever picture taken from space was from a v-2 rocket launched in 1946. This is what it looked like: https://thumbs-prod.si-cdn.com/nnJLv8axPAJI3nScxyDExEeTQLk=/...

Sure, but the phrasing used here was "...the first man made object shot into space."

This whole story is awesome, no doubt. That's just not quite the right question to ask.

In any case, sweet picture.

Totally, I was just trying to be extra pedantic.

No problem, that's what the internet was built for, after all.

First object in space: 1944 [1]

[1] https://en.wikipedia.org/wiki/Spaceflight_before_1951

With the nuke mania of the early Cold War, I'm surprised that the military never tried creating a nuclear version of Project HARP: https://en.m.wikipedia.org/wiki/Project_HARP

(I've also always enjoyed the parallels of this story with "From the Earth to the Moon")

This -- or a rail gun similar to this concept -- in my mind is the key to true low-cost LEO cargo. The investment wouldn't be that great in comparitive terms, and it promises to bring an order of magnitude more cargo to space at an order of magnitude less cost.

All of the 'all impulse at the start' ideas for putting things into orbit die on the fact that air is a fluid. Further that fluid is very dense at the surface of the earth.

What that means in practical terms is that the harder you push something to go through the air, the harder the air pushes back. If you plot the air resistance as a function of energy applied, you see that the long before your payload has achieved orbital velocity + the amount you expect to slow down going up through additional air, you are dumping so much energy into the air that your payload vaporizes.

"But we'll shoot it straight up to minimize the time in the air!" Now you'll have an orbit who's perigee intercepts the earth again (aka highly elliptical).

The ideal trajectory can be calculated, (easiest at the equator but there are solutions for latitudes above and below the equator, to accelerate into an orbital plane such that on your first orbit you can "bounce" (trade excess velocity for altitude) into something that is lies entirely out of the atmosphere. But in that trajectory you spend more time at lower altitudes and that means more energy to get past that air and that makes you vaporize that much more quickly.

"We'll start from a higher altitude!" is another avenue to explore, the ideal altitude to start from is > 65,000' (20 km) which is defined as 'near space' but still 80km from the Kármán line. And we're still unable to build a 20km tall tower.

Good points, but anybody's who spent any time at all modeling the problem has probably worked through them.

Lots of various solutions to various problems. Sabot-type rounds, ablative coatings, small rockets to add the horizontal element of orbit, longer, mass-driver-like launchers, and so forth.

The thing is, none of this, pardon the expression, is rocket science. (It's tough, but there's not a huge amount of new ground to be covered and/or new technologies to develop). If you can accidentally shoot something outward at multiples of escape velocity, you can certainly do so on purpose.

ADD: In reference to GP's comment, with the right trajectory it's completely feasible to shoot something into lunar orbit. Maybe Verne wasn't so far off!

I have wondered if it might be feasible to deliver certain kinds of supplies all the way to the lunar surface using a light gas gun. In fact, you could take the idea to absurd levels by using lithobraking [1] to "land" stuff on the surface. One way to do this might be to explode the shell just above the lunar surface so that the contents will embed only a few centimeters into the regolith, so it can easily be recovered using very simple technology. Kind of like mining really rich ore.

[1] https://en.wikipedia.org/wiki/Lithobraking

This entire line of thinking is very reminiscent of the Air Force's X-Plane program in the 50s.

The Air Force was heading down the road of airplane-to-orbit back in the 50s, but everybody decided that rockets were the way to go so the X-Plane (at least in terms of rocket-planes pushing the envelope of manned flight) program was abandoned.

Who knows where we would be had we kept going? If nothing else, we could have seen major improvements in heat management.

> ... but anybody's who spent any time at all modeling the problem has probably worked through them.

Yes on the modeling but the consensus was 'not physically possible'. Not a solution to the problem. In a single impulse launch[1] the amount of energy that is dumped into the atmosphere around the mass you are accelerating results in temperatures that convert that mass into vapor. Any additional energy just moves it closer to being a plasma.

> If you can accidentally shoot something outward at multiples of escape velocity, you can certainly do so on purpose.

The point here is that the 'cover' in the case of the article did not get into space. A hypothesis that the cover went into space was proposed, but that hypothesis was shown to be false (or untenable) after additional calculations were made. At no time, even "by accident", has anyone sent something into space on a single impulse launch. And the math and models all say it isn't possible, these days you could run the CFD analysis yourself on a high end desktop if you had the software.

In the 80's when Reagan's "Star Wars" program was promising a huge market for launch capacity to LEO, a dozen or more companies sought to get into the launch business. The key feature they were shooting for was "single stage to orbit" but all sorts of things were contemplated, including single impulse ideas. Even for impractical (from a cargo perspective) long skinny javelin type shapes do not reach orbit before vaporizing.

I also have a faint memory of even an ASAT system (trying to 'shoot' things in orbit) that failed to do so due its inability to change the laws of physics. (that would have probably been an Aviation Week article, I haven't subscribed for nearly a decade so early 2000's?)

On the moon without an atmosphere it is a great idea, on earth it just doesn't fly (pun intended :-).

[1] The term "Single Impulse Launch" describes a launch system where all of the energy that is applied to a mass it done at once (and typically by a system attached to the departure point) such that once the mass and the energy system separate, no additional energy is added (post impulse, the mass is ballistic. During ballistic flight it is only being acted on by air resistance, gravity, and sometimes geomagnetic forces)

And the math and models all say it isn't possible, these days you could run the CFD analysis yourself on a high end desktop if you had the software.

Just spent about an hour poking around the internet. Fun subject! I believe you are stating this as far too certain of a thing when it's not. But hey, an hour doesn't mean much :)

There are three issues. Issue #1: without some way of correcting orbit, all you could ever get is some crazy long ballistic orbit (Which is another way of saying an elliptical orbit which intersects the ground). Solution? A sabot with standard space-capable gear inside.

Issue #2: hard acceleration needed. Solution? Not as big of a problem for cargo as you would think. Bull encased his gear in wet clay and it worked in a cannon setup. Some proposals calls for a 2km track with a 100G load, which is fine for a lot of cargo.

Issue #3: the atmosphere is basically a brick at the speed required. This sounds like much more of a problem than it is. Think about it: we have capsules re-entering Earth's atmosphere all the time, sometimes as fast as 12km/sec -- and that's at an very slight angle. Shooting more directly into space, the atmosphere would be mostly cleared in an extremely small amount of time. Bull was working with about 1/4 escape velocity decades ago. Even with gear like the Navy's new railgun, you're in the ballpark.

On my cusory search, I couldn't find other commenters as pessimistic as you are. If you have links, please share! I'd love to learn more. In the meantime, here's a paper that studies the issue. https://research.lifeboat.com/ieee.em.pdf

ADD: I by no means intended to imply this problem is easy or trivial, just as far as I can tell, it's nowhere in the neighborhood of being as open-and-shut as "the math doesn't work out"

I wonder if you could "blow a hole" through the atmosphere (very briefly) and send your real payload after. Sounds absurd, but then so did Orion. I expect you could only manage a transient rarefication behind a rapidly-slowing "rabbit", alas.

There's definitely some interesting math going on. For example, max aerodynamic pressure happens immediately, on launch, with each additional millisecond decreasing stress on the payload.

At high enough speeds, does the atmosphere supercavitate? Could you use a double-shot, with the first projectile basically ablating away and creating a "wake" for the second to fly through?

I think the aerospike concept is related to this solution.

https://en.m.wikipedia.org/wiki/Drag-reducing_aerospike

Interesting. The "air spike" referenced there is like another idea I was wondering about: cross a bunch of laser beams ahead of the bullet, from a spread-out set of ground-based lasers. Like laser launch, but for rarefying the air in front instead of for powering a rocket.

Your thoughts on the Lofstrom loop concept? It seems like the most practical non-rocket idea at the moment.

I like that it solves the problem of getting the mass up over the bulk of the atmosphere before you have to give up giving it any more energy. If you could mechanically build the thing (it looks to be on par with a space elevator in terms of material requirements) I don't see any obvious violations of the laws of physics.

Space elevators would be preferable from a footprint perspective caveat the seemingly infinite tensile strength cables.

I think that the Lofstrom loop is way more realistic than an orbital elevator, mainly because the cables "only" need to span ~100km, rather than the ~36,000km required for a geostationary elevator.

Also, I see that there are improvements in that part of the design: the cables could be replaced with counterweights, and those counterweights could be accelerated upwards and dropped, in order to dampen the waves generated by the release of vehicles.

http://launchloop.com/Counterweights

I don't see why this has a flattened middle segment at all - wouldn't it work better if it was a continuous parabolic arc, like an elliptical orbit that intersects the Earth at the locations of the base stations? That way, the working mass wouldn't have to make sharp turns wherever support cables are pulling on the loop.

For a relatively near-term non-rocket launch system I like Hans Moravec's rotating skyhooks. Similar to a space elevator but less of a leap. If you can get one going around each of the earth and the moon then you could power operations from then on by dropping moon rocks down to earth.

Had to poke around Wikipedia a bit to remind myself of the generic term mass driver.

https://en.wikipedia.org/wiki/Mass_driver

BTW, they seem to be a staple of sci-fi anime, Gundam shows in particular, and look quite a bit like a roller coaster with an abrupt upward ending.

There's another front-page article on HN right now about side projects.

If anybody has the time, many years ago I wrote a simulator for various launch concepts. It's a project that can start off quite simply -- and can get as complex as you'd like. I ended up with an earth-moon model that could handle various rocket/mass-driver/payload combinations.

It's both fun and educational. Great for trying out new languages.

You're not alone in thinking that. Quicklaunch [1] might have been the most serious attempt to actually pursue the idea.

[1] https://en.wikipedia.org/wiki/Quicklaunch

Project Thunderwell: http://www.abovetopsecret.com/forum/thread4434/pg1 - which was also mentioned in one of the articles linked by another poster - seems to be exactly that - I think it was determined to be too expensive / unlikely to produce a viable result!

It has been proposed that we could construct one of these nuclear potato guns to launch things into space[1]. One digs a multi-kilometer deep borehole in say a salt dome(easy to drill), lines that with steel, puts a bomb at the bottom, some reaction mass(essentially trash that will vaporize in the explosion), and a projectile we wish to launch into space.

So this solves several problems over other nuclear bomb based propulsion methods like Project Orion[0]. Foremost, we can contain almost all of the fallout. Second, we aren't accidentally triggering nuclear war by lobbing a bunch of nuclear bombs into orbit. In addition, if the nuclear charge is 150 kilotons or less and that charge is detonated underground then it's in full compliance with certain treaties.

So aside from the shear political impalatibility of using nuclear bombs, there are a couple problems. Namely, even if we use a very long shaft, the acceleration is still on the order of thousands of Gs. So no launching humans. Second, it's impossible to get into earth orbit purely with a cannon launch. And because it's difficult to make stuff that can withstand thousands of Gs of acceleration, orbital circularization will be difficult.

So one suggested use case is to literally shoot the moon. One fills the projectile with things that can withstand the high acceleration like water, construction materials, MREs, propellant, and carbon(extremely rare on the moon, but extremely useful for refining silicon for solar cells). We launch this projectile, which should mass ~3000 tons or so, to the Moon and then harvest everything we need to bootstrap a Moon base from the impact site. Perhaps we can even save money on our MREs by letting the acceleration do the food processing. For example, for apple sauce, all we need to is put cored apples in a plastic bag. The thousand G acceleration of the launch and lunar impact should be more than enough to reduce them to pulp.

Now, while we can't get into Earth orbit easily, if the projectile velocity is sufficiently high we can put the projectile into solar orbit.

It is a questionable proposal, but interesting nonetheless.

[0]https://en.wikipedia.org/wiki/Project_Orion_(nuclear_propuls... [1]https://www.nextbigfuture.com/2010/03/150-kiloton-nuclear-ve... [2]https://www.nextbigfuture.com/2010/12/sea-based-launch-optio...

I saw a talk about something similar (non nuclear) a few years ago. Could have made building up a supply of fuel in orbit cheaper. https://en.wikipedia.org/wiki/Quicklaunch

During my time at MIT, we referred to several explosions in the service and steam tunnels (caused by aging infrastructure) that launched street-level manhole covers into the air as the 'People's Republic of Cambridge Unmanned Space Program'.

Seems the PRoCUSP got several-upped.

The term potato cannon got me interested enough to look up the Wikipedia page for it, where I learned that they are also amusingly colloquially referred to as "spud guns". There's also a vacuum variant known as the "vacuum bazooka".

I always thought 'spud gun' referred to this kind of toy gun - https://en.wikipedia.org/wiki/Spud_gun#/media/File:Spud_Gun.... which fires small parts of a potato.

The vacuum bazooka, looks really neat, I'm surprised that according to wiki, they can apparently project the potato at 650mph.

Apparently according to Wikipedia the term is used in both contexts: to specifically refer to the toy, and more generally as another term for potato cannons.

If the fov was a 1/4 mile, and the plate near the center, wouldnt that halve the calculated velocity? I feel like I am missing the actual data here.

There isn't enough data to know the real velocity of it.

Since the plate only appeared in one frame, only a lower bound for it's speed can be given with observational data. A minimum of 2 frames with the plate visible would be required for determining its speed.

For me, this story, showing up here and there, is to discredit Russian Space Program in some part.

There are so many unknowns...

Do we even know if Russian had a situation like this with their nuclear tests? Most data is classified anyway, so how can we know?

This is not science, this is just "we cannot find a part that could have to reach the space".

Why do you assume this is a anti Russia story? To me this really was just a cute story that someone wants to believe in because it's fun and fascinating.

This story is surfacing every few months on many major outlets. The story itself is fun, but the fact that it often tries to "prove" that the US was first in space without any proof is misinformation. A lot of non-technical people will take this story and pass it as a fact.

Reply on news.ycombinator.com