Polymath A (mostly) technical weblog for Archivale.com

February 18, 2010

Personal Submarines, Bottom Crawlers, Amphibians

Filed under: Engineering,Hydronautics — piolenc @ 10:49 pm

These are chapter notes for book to be entitled Personal Mobility in an Age of Restriction.

3. Water

c. Submersibles and Submersible Amphibians

It might surprise many readers to know that many people, in different countries and from various walks of life, own personal submarines, most of them built by the owner. There are even organizations devoted exclusively to this underwater sport.

See for instance: http://tech.groups.yahoo.com/group/international_psubs_minisubs/

and: http://www.psubs.org/

True, these craft are mostly designed for brief excursions in sheltered water, not for cruising, but their mere existence proves that safe submergence is not beyond the skill of “ordinary” people, which puts them within the scope of this book.

As we will see, the need to safely submerge, operate underwater and surface again imposes constraints far beyond those that apply to surface vessels and surface skimmers. To justify our interest, then, there must be compensating advantages.

We can start with the most obvious one: stealth. Even on the surface, a careful choice of materials, shapes and paint schemes can make a small submarine very difficult to detect, whether with radar or visually. With surface vessels, the only protection that the owner has is that private yachts are fairly common, and the ocean is wide. If a armed vessel chooses to stop him—even in international waters—there isn’t much that he can do about it except run if he can. It’s hard to assert the freedom of the high seas with an Oerlikon gun pointed at you. A submersible has some hope of not being found in the first place, and if found, of escaping. This is all the more true if the submersible is disguised as a surface vessel, in which case its submergence after being challenged on the surface may pass as a sinking or scuttling. Such a disguise is not too far-fetched, as we will see.

A less obvious advantage is immunity to bad weather offshore. Only a few wave heights below the surface, deep water is essentially undisturbed. A submarine submerged at that depth, running at just enough speed to maintain control to stretch battery life, can ride out the most severe storm in comfort and safety.

A still less obvious advantage is access to the bottom and to underwater features and installations. Here we are mostly talking about the bottom-crawling type of sub pioneered by Simon Lake, which were equipped with airlocks that allowed divers to leave the vessel and explore wrecks, collect shellfish, perform salvage or just sightsee. With the negative-buoyancy tanks dry, such subs behave just like conventional ones.


Simon Lake’s submarine Protector was perhaps the highest known development of this type, having a retractable two-wheel undercarriage and a lockout chamber, as well as a fair hull for regular surface and submerged cruising.

Why would freedom-seekers be interested in such a machine? Well, one reason might be emplacement of and access to stores of fuel and other provisions on the bottom, perhaps hidden among wrecks or other known (and therefore unremarkable) hazards to navigation. A small submarine has limited storage and bunkers, so being able to replenish without risking the shore or a harbor might be a good thing.

submarine supply station

If the circumstances were right (sheltering feature located close to land, “friendly” parties living on the seafront), buried cables and even hoses might be run out, allowing direct replenishment and access to the communications and power lines.

What’s wrong with submarines is, basically, that they have to submerge. This means that they have to be able to take on sufficient water to cancel their reserve buoyancy, and even to become negatively buoyant if rapid submergence or the ability to rest or crawl on the bottom is needed. That water ballast takes up space. The more reserve buoyancy the sub has, the less usable space it has inside! But shaving any vessel’s buoyancy margin is dangerous; a slight error in loading, or a wave running over an open hatch can send it to the bottom. Many early submarines ended this way. A narrow margin also makes a submarine very “wet” in heavy seas, because instead of rising to a steep wave it punches through. This is less important today because a deck watch is probably not needed any more, when a video camera mounted on a snorkel- or mast-head can see farther than any watch-stander. But the fundamental objections to low reserve buoyancy—vulnerability to even trivial accidents and intolerance of overloads—stand.

Simon Lake, whom we’ve already mentioned, may have been the first to solve this problem by a method variously called the “double hull” and “saddle tanks.” This layout relies on the fact that a submarine’s main ballast tanks have two normal conditions: full and empty. They are never filled partway, other than during blowing and flooding, which are transient conditions. Therefore they need not resist external pressure. No matter how deep the boat dives, the pressure outside and inside these tanks can be the same, both because water is substantially incompressible and because the tanks can be left free-flooding until it is time to blow them empty to surface. Therefore, they can be housed outside the pressure hull, in a lightweight casing that sometimes surrounds it completely, but often just straddles the top, like a saddle, and in modern submarines is usually housed at the bow and stern, fitting into the overall hydrodynamic contours of the hull. Other stores can go there, too; fuel oil floats on water, so fuel tanks can be provided with openings at the bottom that allow seawater to replace the volume of fuel that is consumed (these days it would make sense to enclose the fuel in a bladder, to completely separate it from the water). Again, pressures are equal inside and out at all times. In naval submarines, spare torpedoes and other stores that tolerate exposure to seawater could also go in the saddle, and with the casing serving as a hydrodynamic fairing, piping and other equipment that would otherwise take up space inside the pressure hull could go outboard. In one stroke, Lake solved the central problem of submarine utility (ironically, although he illustrated this innovation in patent drawings, he never claimed protection for it in the text), and this has been the standard configuration for fleet or cruising submarines ever since. It allows very rapid diving. The Kingston valves, which admit water to the bottom of the tank, are first opened, ready for diving, but water does not fill the tanks immediately because of the air trapped inside. When the sub is ready to dive the air vents are opened and the tanks finish filling quickly. It is important to fill the main ballast tanks quickly even in non-naval applications because, while the tanks are partly full, water is free to slosh fore and aft and the sub may be difficult to control. When surfacing, air vents are closed, the Kingstons are opened and compressed air is introduced to blow out the water. Once the sub is “decks awash” a low-pressure air pump is used to remove the last of the water. Or the Kingstons can be closed and the remaining water pumped overboard with a conventional bilge pump. The latter solution is more efficient, at least in theory, but requires more plumbing.

[Added later: Simon Lake didn’t patent saddle tanks because he was not the first to use them; that honor belongs, apparently, to the French engineer Laubeuf, another talented early sub designer.]

Some tankage has to stay inside the pressure hull, namely the trim tanks – fore and aft – and the negative-buoyancy tank if there is one. The trim tanks are for fine adjustments to ensure that the boat is neutrally buoyant and level with the main ballast tanks flooded. Theoretically, their capacity is equal to the difference between the maximum and the minimum loading of the sub. In practice, they need to be somewhat larger because no fore-and-aft adjustment is possible if the two tanks are either completely full or completely empty. In either of those conditions solid ballast or stores would have to be shifted to trim the boat. Housing the trim tanks inside the pressure hull is mandatory because generally, they will have in them some air, which is compressible. Since water is admitted or pumped out only on the surface or during a very shallow “trim dive,” the pressure differential between the trim tanks and the interior of the pressure hull is always low, and can be made nil by venting air from the tanks into the boat with the outside valves closed. Because the tanks are housed inside the pressure hull, compressed air should not be used for pushing out water; instead, water should be pumped overboard.

Finally, there is the negative tank – a special case if ever there was one. It may be flooded or emptied at any depth within the submersible’s operating range, so it requires special arrangements. If it is filled at depth to allow the sub to settle to the bottom, the water cannot be admitted to it at full ambient pressure or the internal partition between the tank and the sub’s interior would rupture. Admission of water has to be through a restriction, with a relief valve venting air into the boat as needed. Likewise, the tank cannot be blown with compressed air like the main ballast tanks; instead, water has to be pumped out of the boat through the pressure hull. That pump, operating in reverse with a brake on it, can serve as the intake restriction. In a bottom-crawler, the capacity of the negative tank is determined by the bottom pressure required to get adequate traction for the drive wheels; in a regular sub, by the maximum static rate of descent needed for crash dives.

Now here’s the beauty of the outer casing: it doesn’t have to withstand pressure, so it can have any shape desired, as long as it provides the necessary volume for ballast tanks and bunkers. It can look like a submarine…or like something entirely different and much less remarkable. It makes good sense for somebody who wants to be discreet to make his casing look ordinary.

Argonaut II

Lake chose the shape of a sailing sloop (though the two “masts” – snorkels, really – ahead and abaft the “deckhouse” would have raised some nautical eyebrows). For us, a motor yacht would probably make more sense, though in some places a trawler or some other kind of workboat would be more appropriate, and would make the snorkels easier to conceal or disguise. The hull simulated should be a displacement type, preferably with round chines, because the sub will definitely not be capable of planing! We want to keep submerged drag as low as possible, so excrescences should be minimized. A traditional displacement motor yacht, with a low deckhouse, should do the trick.


It makes sense to measure cost in relation to a surface vessel of equivalent volume or payload. Because of tankage, the total volume of pressure hull will need to be about 40% larger than the living quarters of a surface boat to get equivalent usable space. This in itself increases cost. The cost per unit of displacement will also be higher because of the greater amount of machinery used. Now add the casing, and it is probably safe to assume a cost multiplier of about three.

Bottom-Crawlers and Amphibians

With the possible exception of rumored clandestine reconnaissance craft built under the Soviet régime and the recently retired US Navy research sub NR-1, modern submarines have completely abandoned the bottom-crawling mode of operation championed with great success by Simon Lake. Lake’s brainchildren remain of interest to us for two reasons: our application requires the boat to operate in shallow water much of the time, and shallow water is the enemy of conventional subs; and adding wheels opens up the possibility of limited amphibious operation. This is very much in keeping with our preference for bi- and multimode operation. Here, however, the emphasis is on the word “limited.” An amphibious submarine, unlike a surface-bound amphibian, can never pass as an ordinary highway vehicle, simply because it is too heavy. To prove this, imagine an amphibious sub built to the maximum practical highway size of 40 ft x 8.5 ft x 8 ft overall, and give it a fullness coefficient of .60 for the pressure hull. That gives us a submerged displacement of about 55 tons, which is roughly what the beast has to weigh out of water. Putting all that weight on two axles violates every highway regulation in the civilized world.


One remedy, adopted in the late-WW2 German Seeteufel (Sea Devil) amphibious sub, is to use caterpillar treads to better distribute the weight, but the weight, cost and maintenance burden of a crawler rig makes it unattractive. The next best thing is low-pressure tires, but pneumatic tires are impractical for underwater operation; at a (shallow) depth corresponding to their inflation pressure, they will collapse. One could envision using pneumatic tires as part of the main ballast system, and pumping water into them when it’s time to submerge, but that would entail a good bit of complication. We’re back to Lake’s formula of solid-rimmed wheels, but with the addition of solid-rubber tires shaped and formulated for low ground pressure. This is relatively easy to do because of rubber’s unique and “tunable” properties. Hysteresis heating of the rubber will limit ground speeds out of water, but we’ve already seen that this thing isn’t going on the freeway in any case. Solid rubber is substantially incompressible, so there should be no change of buoyancy and trim with depth. Despite these limitations, the ability to leave the water, even if only to the extent of coming up on a beach or boat ramp, gives an operational flexibility that we’ve already commented on in connection with hovercraft. The ability to load and discharge cargo and passengers on dry land, independent of port facilities, lighters and stevedores, shortens turnaround time. To a commercial operator that translates into greater productivity; for us, it means greater security. The ability to operate in estuaries, over gravel beds, mud flats (with caution, because it is possible to get mired) and sandbars, without fear of going aground, is another big plus. And imagine the commotion when a cabin cruiser comes up the local boat ramp, under its own power, when nobody saw it approach. Amphibians do incur one penalty: To come out on land, we need at least three wheels distributed over two axles. Underwater, we could make do with two wheels in line, as Lake did in his later designs.


The design of any submersible is more complicated than the design of a surface vessel, for a number of reasons, some of which have already been discussed. Stability, in particular, has to be verified for at least three conditions: surfaced, submerged and the transitional phase, during submergence and surfacing, when the main ballast tanks are partly full and the water in them is free to slosh. In the case of a bottom-crawler with a bicycle undercarriage, the sub’s immunity to tipping also needs to be verified.

Taking the simplest condition first, fully submerged stability requires only that the center of buoyancy be above the center of gravity. This condition is usually very easy to satisfy.

On the surface, the c.b. is nearly always below the c.g. This is permissible, provided that the metacenter – the point where a vertical line through the new center of buoyancy intersects the plane of symmetry of the vessel when the vessel is heeled slightly – is above the c.g. This calculation is well covered in regular naval architecture texts, but has to be done very carefully for submarines because they can be marginal for lateral stability on the surface, depending on the presence and shape of the saddle tank. A more boat-like casing tends to give better stability than the casing forms that are optimized for underwater cruising.

The transitional condition is usually dealt with by subdividing the main ballast tanks fore and aft to limit sloshing, and by lateral equalizing pipes to ensure that the corresponding port and starboard tanks fill and empty at the same rate. Aside from that, generously-sized valves and air vents ensure that the tanks fill quickly so that the transition is brief. Surfacing is usually done by powering up to periscope depth using the hydroplanes, then surfacing quickly from there. The critical transition condition occurs when the sub must surface statically – that is, by blowing its tanks – from a considerable depth. This might be the case if a bottom-crawler needed to surface from a tight spot where it was unsafe to drive or swim ahead. In this case, there might not be enough air to completely blow the tanks at depth. Instead, enough air would be released into the tanks to completely blow them at periscope depth. That air will only occupy part of the tank volume at depth, and will gradually expand as the sub rises and ambient pressure decreases. This leaves a possibly long period during which there is a free surface in the ballast tanks, making it especially important to get this phase right for bottom-crawlers.

Bottom crawling on a bicycle undercarriage means ensuring that the moment of the boat’s net weight (basically, the weight of water in the negative tank) is less than the moment of the boat’s buoyancy when the boat is heeled. This is fairly easy to ensure and to verify.

Detail design is concerned with keeping costs down, ensuring safety and minimizing crew workload. Cost control is primarily a matter of maximizing the utilization of expensive machinery. There are, for example, many tasks requiring the use of a pump; it pays to arrange for as many of them as possible to be done by the SAME pump. Here, as in many other design tasks, there is software that can help. There are several commercial software packages designed to optimize the design of chemical processing plants by minimizing piping runs and avoiding duplication of machinery that can help in laying out machinery aboard a submarine.

Powered by WordPress