One-Person Dry Submarine

 Photo of the sub. 
 Sketch of inside the sub. 
 Better sub design. 
 Better port design. 
 Karl Stanley's 2-person, 600-foot-depth sub. 
 Karl's sub's interior. 
 Hull-strength GW-Basic program. 
                    THE FLORIDA MANATEE
   The design for the Manatee incubated after reading one of the first
books about diving, Jacques Cousteau's "The Silent World".  In it he tells
how he and collaborator Emile Gagnon developed the Aqualung, and about some
of his early dives in the Mediterranean.  I was in my early teens in Coral
Gables, Florida at the time.  From the book, I realized that skin diving with
all its limitations was not the answer.  Major problems include the bends,
nitrogen narcosis, embolism and oxygen poisoning.  But the biggest limitation
is that you just can't go very deep (safely).
   As time went on, we were given a derelict 10" Atlas lathe in neglected
condition.  After rebuilding it over a summer vacation, it turned out to
be better than we thought.  With care, precision work could be done.
   At the same time, we acquired a used Sears AC welder for a good price.
Over the summer vacation, we decided to do something major with the new
"machine shop".
   It had to be a submarine. Everybody builds boats, swamp buggies, airboats
and even airplanes, so that wouldn't do.  It had to be something more
challenging.  At the time we didn't realize how many individuals were
building subs.  Print-media articles began appearing after we build ours.
   As a physics teacher, I already understood the science of subs.  Ours
would be as small as possible so it would be as light as possible. It would
be atmospheric pressure so you have no physiological problems. For propulsion,
pedalling was chosen since we wanted to tow it most of the time, then pedal
it into exact position. It would be dry since (1) you have to move your legs
to pedal and it's a lot harder if the hull is full of water and (2) it's the
same weight wet or dry, since the thing has to sink.  It would be towable by
an average Florida salt-water boat.  It would use a common gas for buoyancy,
either nitrogen or carbon dioxide.  Carbon dioxide was chosen because it's a
liquid in the tank, so you have a lot more of it. We then realized this design
is equivalent of SCUBA. With SCUBA, you only need to fill your air tanks.
With our sub, you only need to fill the carbon dioxide tank, which can be done
at any fire-equipment supplier.
   We ran a hull-thickness program to determine what sort of steel we're
looking for.  It was decided that a maximum depth of 350 feet would be
sufficient (crush depth 600 feet).  Our program showed that 3/16" steel would
be sufficient.  This thickness seemed intuitively to be awfully thin and so we
got the help of an engineer and he confirmed it.  Just for insurance the hull
has a reinforcing band inside at the weakest area and the ballast tanks in
front and back reinforce those areas.
   NOTE: It's about the same work to build a much thicker hull and you
         could see a much greater volume of ocean if you chose to do that.
   Looking around for a steel tank to use, we found that propane tanks are
3/16" and are very affordable.  We made a scale cardboard mockup from a bath
tissue roll, sat on the floor, measured all dimensions, measured where feet,
knees and eyes would be located and discovered that it was all feasible.
So the project began with the purchase of a propane tank.
   We then decided on the layout.  The feet in front would turn a right-
angle drive using bike cranks and pedals.  The front seal would be a common
ceramic water pump type.  The ballast chambers would be 1/8" steel painted
inside and out with epoxy boat paint. There would be no plexiglas hemispheres,
since they cost thousands.  It would use flat plexiglas close to the face. The
view is adequate.  Ports are 10" in diameter and 1 & 1/4" thick.
   We marked the tank diagonally, as the mockup indicated, using an aluminum
straightedge and a line-filament bulb to cast a sharp shadow. We cut the
steel using an industrial saber saw with bimetal blades. The insides and
outsides were cleaned with a 4" grinder using sanding disks and wire brushes.
   The stink of the ethyl mercaptan, put in tanks so you can "smell the
gas", which is odorless, is atrocious.  Which is how  we gained the derogatory
title, the "Skunk Works".  After the initial sting of insult, we decided to
adopted the title. The name is associated with Kelly Johnsons's section at
Lockheed where the "sound barrier" was broken and several super-secret
military planes were developed. Solvents used at a factory nearby earned them
the nickname, which was in turn borrowed from the popular Al Capp cartoon
"L'il Abner", referring to the smell of the Dogpatch moonshine still.
   The tank was welded in an L shape before the hatch end was cut open to
lessen any distortion.  A common pipe section was used to mount the pedal
drive to the sub hull and front ballast pyramid. Stainless thru-hulls allow
for steering and to drop the steel emergency weight underneath the sub.  The
rear wheels are attached to the drop weight.  The front wheels are on an axle
bolted to the original tank feet.  The domed hatch was cut off and a rolled 1"
steel ring with a groove cut into it was welded to the tank.  The hatch fits
inside the groove, where there is a poured-urethane gasket.  The epoxy-painted
hatch edge is the mating surface. The hatch rim gives the hull extra strength.
There is a 3/8" x 6" rolled reinforcing band inside the lower hull half. There
is a common bronze flood valve to let water in, relieve the pressure and allow
the operator to get out while submerged if necessary. Releasing the drop-
weight alone should allow a quick return to the surface.  In the worst-case
scenario, such as running out of CO2, drop the weight, float to the surface,
open the hatch, jump out and let the sub sink if it begins to roll over. If
not, wait for the tender crew to retrieve you.
  There are two valves from the carbon dioxide cylinder to the front and rear
ballast chambers, so you can adjust the float angle on the surface. One valve
lets the buoyancy gas out of both chambers to sink.  Neutral buoyancy while
submerged must always be set with lead weights internally. You can't use the
buoyancy gas for neutralizing because a rising bubble gets bigger and bigger
and comes up faster and faster, and a sinking bubble does just the opposite.
   The control surface can be mounted vertically to be a rudder if it's
being pedalled. Or it can be turned horizontally to be a diving plane if
being towed. Lead-weight-shifting allows some pitch control when submerged.
   The biggest problems turned out to be breath moisture, lessened by exhaling
thru closet dehumidifier crystals in a canister. And we had severe difficulty
in towing the sub.  It really needs to be lifted out of the water, but then
you get into a lot of complication.  A pontoon tender would be ideal, since
there's only a foot of freeboard when the sub is on the surface as it is now.
  The QBASIC program calculates hull strength and dive depth for steel. Our
sub can go to 600+ feet crush depth.  It weighs 1200 lbs.  We have a special
trailer for the sub.  The time down can be twenty minutes with hard pedalling
or 45 minutes with no effort.  Of course, you can come up as fast as you like
and you can go back down as fast as you like. We have a crude signalling
system. Striking a steel plate in the water with a hammer tells the sub to
come up.  The sub ringing back means "get out of the way, I'm coming up".
                   A BETTER SUB DESIGN
   Our original sub was designed on paper and with a cardboard model.  A lot
of thought went into playing devil's advocate to find problems and correct
them during that stage, before construction began.
   This Better Sub design is the result of some unforeseen problems and better
design details. The ports and the hatch are better.   This sub could dive
deeper.  It uses electric power and so saves knee space. The prop is shrouded.
  The biggest changes are in the operator position and ballast tanks.  The
operator is prone like a SCUBA diver. The dive plane and rudder are controlled
by foot. The ballast tanks pivot to be parallel or perpendicular to the hull.
On the surface, the ballast tanks (floats) are horizontal while the sub is
vertical.  The weight-drop control, motor control and ballast tank angle are
hand-operated.  The operator is standing when the sub is at the surface.
  The batteries are a significant internal, fixed ballast. The sub is neutral
when submerged and all ballast gas is purged so that internal lead weights are
the ballast.  The emergency drop weight would only need to be about 100 lbs.
  Manipulators and snorkels are bad ideas.  So are electrical thru-hulls
for external lights.  Use a flashlight from inside.  Attach a diver's depth
gauge just outside a port.  Oxygen replacement can be used for long dives.
You must have an oxygen meter, preferably two.  Carbon dioxide scrubbers use
caustic materials. Closet moisture absorbers will remove uncomfortable breath
moisture. A fan will be necessary for the last two, which will draw battery
power. A barometer and your eardrums will tell you when pressure is not right.
A much better idea is to surface more frequently and just flush out the stale
air with fresh air as quickly and as many times as you care to.