User:Jeff Chan/Seadrome

From Seasteading
< User:Jeff Chan
Revision as of 02:39, 14 October 2009 by Jeff Chan (talk | contribs) (References)
Jump to: navigation, search

Description

Seadrome was a proposal by Edward R. Armstrong from 1927 through 1946 to build a series of floating airports in the Atlantic in order to enable trans-ocean passenger flights between U.S. and Europe before long distance, unrefueled flight was possible. Aircraft would land, refuel, and fly to the next Seadrome in a series of hops. Two things stopped it from happening: the great depression and improvements in aircraft range. Nonetheless, the Seadrome design appears to be very interesting and thoughtful. It was designed to be maximally stable in open ocean waves.

Detail

  • Overall Specifications (length, width and height are about the same as a modern aircraft carrier +10% or so):
    • Length: 1200 feet
    • Width: 400 feet at center, 200 at ends (i.e., narrower at fore and aft decks, much like an aircraft carrier). We could build to different shapes.
    • Draft: variable from 50 feet with ballast/heave plates retracted into vertical float columns, 160 foot draft with ballast/heave plates fully deployed
    • Air gap: 70 feet (or possibly 70 foot deck height, but images look like the deck may be 20 to 30 feet above that.)
    • Displacement: 64,000 tons fully deployed (compared to 100,000 ton displacement of a Nimitz class aircraft carrier)

Note that there were at least two slightly different designs for Seadrome and the specifications may vary between them.

Design Description

Seadrome has a trussed upper and lower deck, like a large double-deck bridge. The upper flight deck is a flat and open aircraft runway except for a hotel and control tower. The deck shape is very similar to an aircraft carrier in plan (top-down) view. Like an aircraft carrier Seadrome was to have an aircraft elevator to a hangar deck below the flight deck. The lower deck also had hotel space, lifeboats, living quarters, generators, machinery, etc.

Floatation is by about 30 large vertical floats with a diameter of 15 feet from the deck to some feet below the nominal waterline. Some feet under water the columns widen outwards to form to buoyancy tanks that are 30 feet wide and contain air, fuel and water tanks. Some buoyancy and levelling were to be trimmed (adjusted) by pumping around the contents of those tanks between cylinders. Below those tanks, an even narrower column leads another 160 feet below the waterline to iron ballast. The ballast was shaped as a simple cylinder in early versions, and a wide, inverted mushroom shape in later versions. The mushroom shape formed a heave plate of about 40 feet in diameter. The heave plate shape was pointed at the bottom to allow the column to fall more easily than rise and flat on top to resist rising for example due to a wave reaching the column.

The lower columns were designed to retract into the upper columns in order to reduce the draft to a possible 50 feet when maneuvering from shallower waters. They would be lowered in deeper waters. So the draft is variable depending on needs. The structure may be less stable with the ballast retracted up, but the winds and waters may be calmer closer to shore, and deployment from shore could be chosen during more stable weather than might eventually be encountered when stationed in the open ocean.

Since the buoyancy tanks and ballast/heave plates are inline in the same column, forces from the rising and falling wave water surrounding the column are mostly handled within each relatively strong column shape. Most of the forces may be handled locally within a given column and not transmitted to the deck or larger structure. Based on our model testing, it would seem that the heave plate may have a very significant effect on damping wave motion. Where a plain column would tend to bob up and down in heave after a vertical displacement, the Seadrome column settled immediately in something that may approach critical damping. Measurements would be useful to confirm this.

The large, underwater buoyancy tank raises the center of buoyancy since it's near the undersea surface. The deep ballast plates lower the center of gravity. Both of those features increase hydrodynamic and hydrostatic stability, especially taken together. That the buoyancy tank is underwater means that it is less affected by waves which mostly pass over them and interact with the thinner float column above. The thinner column offers less area for wave interaction. That the heave plate is very deep underwater probably means that it's operating in relatively very stable water. Both features reduce the response of the each individual column and thus the overall structure to waves.

The columns are doubly trussed together horizontally both above and below the water. There are also diagonal cable stays between the trussed columns. So some translational forces are shared between the columns and overall structure via the trussing and cable stays. This makes the whole stronger than the sum of the parts.

All of these design features together probably result in a structure that's strong, relatively light, and could offer a very favorable wave response. Model testing and/or software simulation should be done to confirm the design features and refine the design parameters.

Seadrome is an old design, but a very thoughtful and interesting one that could form a successful basis for seasteads.

Requirements Analysis

  • Safety - should be relatively stable in heave, pitch, roll. Minimal wave response in all directions. 70 foot airgap is nearly double ClubStead, but possibly still too low for rogue waves. Decks could be made higher, for example by scaling up the entire design.
  • Comfort - could be very stable in waves. Large horizontal deck area may feel like land or a large city block
  • Cost - ten million dollars per Seadrome was quoted in the 1930s. Call it half a billion dollars or more now.
  • Pretty - Industrial looking, kind of like a giant pier floating in the ocean, or stationary aircraft carrier, or a really wide oil platform
  • Modular - probably not modular. May be possible to raft them, or join with bridges. May require precise station keeping that should be possible with electric thrusters, accelerometers, computer control, etc. If joined or bridged, dispersal for storms should be quite viable.
  • Cargo - Was designed for housing people, storing aircraft, fuel, water, food, supplies.
  • Free Floating - Yes
  • Scalable - Smaller versions can be built, but for best response to full size ocean waves, it needs to be built to full scale. Subsections can also be built. Originally a middle third of Seadrome was meant to be built as a demonstration prototype. 1/3 of Seadrome, scaled down, may make a good Baystead (Seastead prototype model).
  • Standards - Unprecedented design overall, however, steel construction techniques and standards could be very similar to ClubStead, semisubmersible oil platforms, large bridge trusses, etc.
  • Mobile - Designed to be slowly mobile, possibly self-deploying. Meant to be anchored and point runway into the prevailing wind, swinging around massive anchor. Seasteads may not need anchor.
  • Draft - Variable draft. Ballast/heave plates retract up into vertical columns (nesting mostly inside them) for operation in shallower waters such as launching from construction near shore. Should reduce draft to about 50 feet. Fully deployed draft is 160 feet.

Model Testing

Jeff Chan and friends built a 1/100 scale model of a Seadrome foredeck (the front or rear third with 2 by 3 columns) at Ephemerisle 2009, but the materials were too heavy for its scale. Lightening efforts are underway, but a model with lighter materials would be a better test.

Heave response of a single model column was excellent, settling immediately, possibly near critical damping. Where a plain column would tend to bob up and down in heave after a vertical displacement, the model Seadrome column settled immediately in the first cycle. Measurements would be useful to confirm damping factors, for example by digitizing video of the column or model moving. Accelerometers could also be useful. Vince has some good ideas in this area.

References

References from wikipedia: