Home-  Sailboat-  Submarines-  ROVs-  Metal Working-  Other Stuff -  About Us

Why Build a Sub
   Submarine 101
   Wet Sub
   Planing Wet Sub
   Dry Ambient Sub
   Submarine Yacht
Design Evolution
Cutting Aluminum
Building the Hull
Kort Nozzles
Battery & Chargers
Engine & Jet Drive
   Engine Tune-Up
   Hull Cooler
   Wet Manifold
   Engine Mounts
   Engine Box
   Jet Pump Valve
Helm Controls
Ballast Sled
Landing Gear
Trim Tanks
Wiring Harnes
Tow Truck
Rules of Thumb



Building the Hull


The aluminum was arrived!  Just over 1 ton of various 3/16 inch thick sheets, two of which are 6 x 20 feet and weigh over 300 pounds, plus angle for reinforcement.  My order came from MetalSalesUSA in Louisiana, even though I could have purchased locally, availability and the price including the shipping were lower when ordering from a company that routinely supplies boat builders.  I latter priced some more material from MetalSalesUSA and the prices were much higher that other suppliers, so the I would always recommend shopping around.

All of the sheets arrived strapped to a frame work of 3 - 4x4's and planking so I gladly splurged and rented a fork lift for $150 including operator and helper. Once the skid was off the truck the fork lift was able to slide it down our narrow drive way and into the backyard.  Individual sheets can be dragged off of the stack and onto 2x4' where they can be cut before moving them into the assembly area.

All of the material is 5000 series aluminum designed for marine use.  Two of the 4x20 ft sheets are 5086 H116 which is a harder and more expensive alloy that the 5052 H32.  There were intended to be used for the bottom hull an only by luck did this actually happen, because my fist mistake was to forget that two of the sheets where 5086 H116.  Lucky they actually got used cut into the correct pieces.  I learned that you identify the sheets based on a color code painted onto the end edge. 

It turns out that using 5086 H116 for the bottom hull only made bending it to form the bow more difficult. In hind sight I would likely dispense with using 5086 and only use the cheaper 5052.  That would not be the case if I were building a river boat and indented to frequently bounce off rocks, but as a submarine my worst nightmare is a surface vessel that ignores the dive flag that will be surfaced before I surface my sub, and proceeds to run his keel across the top of the cabin.  Now that I have had a chance to work with 3/16 inch 5052 I feel fairly confident that I will win any battle with a fiberglass hull.


Aluminum Order

Harvey LA    
  Sheet Each Cost
4 .188x48x240 5052 H32 319.00 1276.00
2 .188x48x240 5086 H116 401.00 802.00
2 .188x72x240 5052 H32 478.00 956.00
1 .125x48x192 5052 H32 212.00 212.00
  1x1x3/16x25ft 5456 28.00 140.00
  1.5x1.5x1/4x25ft 5086 53.00 106.00
30 .030, 1 pound spools 4.56 136.80

Material Total




Local Shipping


Forklift Rental


Grand  Total


Aluminum Alloy Comparison


  5052 H32 5086 H116 6063 T6
Component Wt. % Wt. % Wt. %
Al 97.3 95.4 98.9
Cr 0.15 - 0.35 0.05 - 0.25 Max 0.1
Cu Max 0.1 Max 0.1 Max 0.1
Fe Max 0.4 Max 0.5 Max 0.35
Mg 2.2 - 2.8 3.5 - 4.5 0.45 - 0.9
Mn Max 0.1 0.2 - 0.7 Max 0.1
Si Max 0.25 Max 0.4 0.2 - 0.6
Zn Max 0.1 Max 0.25 Max 0.1
Ti 0 0 Max 0.1
Density 0.0968 lb/in³ 0.0961 lb/in³ 0.0975 lb/in³
Tensile Strength, Ultimate 33400 psi 42100 psi 34800 psi
Tensile Strength, Yield 28300 psi 29700 psi 31200 psi
Elongation at Break 12 % 12 % 12 %
Shear Strength 20300 psi 25400 psi 21800 psi















Laying out the pattern on the aluminum sheet.
Laying out the paper pattern.

Laying Out the Pattern

Before the sheet is marked it is dragged out of the stack and onto 2x4's so it is laying flat.  The 2x4's will also act as saw horses.  The blade of the skill saw is set so that it only protrudes about 1/4 inch thought the sheet, so there is no problem cutting over the 2x4s. Since I need two copies of most pieces and the aluminum is the same on both sides, I can stack up 2 sheets and cut both parts at the same time.  I did not have the guts to do this with the large sheets because I wanted to limit the damage if I made a mistake.  The second option is the use the first piece cut as a pattern to mark the second piece, and that works nicely.

The CAD drawings created using TurboCAD where plotted onto paper by a company that normally plots patterns for making sails.  I tried transposing the design just using the measurements for the drawing, but the curves are very time consuming making the $40 spent on the plotting a great investment.  The plans are not complete nor did I have every drafted part plotted.  I only did the parts with curves and almost every thing that formed the framing for the cabin.  The CAD drawing and the patterns are also not intended to be the final shape.  I'll make small adjustments and changes that I did not care for or foresee to make in the CAD drawings.  So some of the patterns, such as the rear bulkhead of the cabin will be more for a reference check and for actual size.

Having never seen the parts of a boat laid out it was not possible to see how the pattern would form a bow, so I first tested the pattern on a on sheet of masionite and formed it over a plywood frame.  This helped provide the required confidence to made the fist cut. An equal amount of anxiety must be over come to start the first weld.  The two bottom hull sheets are now tacked together and its starting to look like a boat.

I considered a building frame
to turn the hull so that welding
could be done face down.
Plywood cradle supporting bottom sheets.
Plywood cradle cut to the 12
degree deadrise of the bottom.

Hull Cradle

I considered building a set of round frames that set atop rollers so that the hull could be rolled while it was under construction, but decided to use simpler plywood blocking and just leave much of the welding until the boat can be dragged out onto the lawn and rolled over.  The plywood frames are cut with a 12 degree angle for the dead raise and then blocked and shimmed to proper alignment.  After the sheets were laid on the blocking the blocks were leveled and the sheets were welded together with several 2 inch welds.  Temporary cross bracing was used to help pull the sheets together and provide extra support.

The plan is to set all 6 of the longitudinal frames in place that come forward into the cabin. Those will then support a winch that will bend the bow into shape.

When the the first large side panel went in, the boat stared to take shape and the shop started to look much smaller.  Just 4, 1 to 2 inch welds were enough to hold the sheet in place. 

Temporary bracing is added in order to maintain the shape of the structure before additional welds are added.  A cradle with a 12 degree angle was made in order to easily assure that the sides are properly aliened with the bottom before the temporary braces are tack welded in place.


Bottom sheets in the cradle and tacked together.
(1) First sheets for the bottom of
the hull tacked together and
temporally braced.



Weld down the bottom.
Side hull set in place.
(3) One of the hull sides is sitting
in place before welding.
A frame is placed 2 inches inside the side of the hull.
(4) Hull sides on, bracing added,
and one frame in place 2 inches
from the hull side.
Strongback attached to hold the side straight.
(5) Strongback attached before
the webbing is stitched in place.
Center frames are added.
(6) Center frames are added.
Lots of clamps are temporary
braces hold everything in place
and allow for adjustments to be
made before everything can be
tack welded in place.


Detail of the clamps.
(7) Detail of the clamps.
Webbing tacked into place.
(8) Webbing tacked into place.

(9) .and folded over.

Welding Stuff Together

Any welds that will not be covered up by the next piece will not be completed until final welding or just before they are enclosed in order to reduce the amount of overall distortion to the frame. The idea is that the addition of additional structural components will reduce some of the warping caused by welding.  Because the frame is held together will relatively few welds any serious distortion from welding on new parts will break some existing welds with a loud BANG!  That your cue to find the weld that just broke and to do what you can to reduce the stress your causing, such as letting the part cool down.

(2) So the bottom center and the inside of the sides of the hull are completely welded before the inside frame are added.  The sides are welded to the bottom hull with 2" welds with 4" skips,  and the bottom is a solid weld but completed in 3 to 6 inch segments.  It is important to move about the structure and add only two or three welds in any one area in order to prevent heat buildup and reduce deforming the structure.

(4) There is only 1, 5/8 inches between the frame which are 5 inches deep, so welding inbetween the frames is not possible.

(5) A temporary strongbacks were added to each side hull to assure alignment while the webbing is being stitched in place. Once the webbing joins the side hull to the frame the two pieces become very stiff.

(6) The center frames came next.  (7) Clamps and temporary cross bracing are the order of the day.  With enough clamps even large pieces can be completely dry fit. 

(8) (9) The webbing between the frames is tacked into place and then bent to the shape of the frame.  Heating with a propane torch in addition to hammer persuasion is necessary for the sharper bends.













(1) Bend, weld a bit and repeat.

(2) Notch a scrap, hook it on,
pull, clamp it in place and weld.

(3) Quickly fabricated bending

(4) Get the winch, hook it on and
finish it off.

Bending the Hull to the Frames

(1) With the center frame in place I figured it was time to bend the bottom hull up into shape.  The first part of the bend can be done by hand and a couple of jack stands will hold the bend while the hull is tacked to the frames.

(2) A 2x4 is then used to further bend the hull up to the frame and a few more welds are added to again hold it in place. (3)Then notched pieces of sheet are used to hook onto the end of the hull sheet so it can be clamped into place against the frame while additional welds are added.

This is where I wish I had not designed the bow quite so steep, but I wanted to maximize the size of the bottom hatches and to be able to bring the ballast sled as far forward as possible.  The problem is that a 3/16 inch sheet of Aluminum that is 2 feet wide does not bend easily unless the radius of the bend is about 5 feet.  (13) The last part of the radius my hull is about a 1 foot radius, so out came a winch to pull the hull into shape. Even the winch could not obtain the needed curve, but a couple of pieces of sheet with notches cut in them make quite nice bending tools.

Once hull sheet was in place, I learned that I should have waited until the rest of the cabin framing was in place.   The problem is that the 3/16 sheet became a large spring and was forcing the center frame down and out of position by about 1/2 inch.  Part of the problem is that the hull sheeting is  5086 H116 and so it is even stiffer that the 5052 H32 used for the rest of the boat.  Additional bracing was needed to force the center frame back into position and my attention has now turned to completing the cabin framing in order to hold the center frame in place.

(1) 1st mate tries the cabin out.

(1) A 2 ton jack, makes it fit!

(3) Backside of cabin bulkhead,
showing the groove cut for
bending the back rest into it.

(4) Pushed the frame back to
where it is suppose to be, and fit
in the upper, side of the cabin.

(5) Header, front window frames
and deck panels all tacked in.

(6) The header over the forward
windows has 1/2 holes drilled for

(7) Bench seat and hatch

(8) A 1 1/2 inch pipe was cut in
half to reinforce foot rest frame.

(9) Hatch frames are two pieces
each that will latter hold 3/4 inch



(10) Carl cleaning up my welds.

Cabin Assembly

(1) The cabin bulk head is in.  I have a pattern for this piece created from the CAD drawings, but from this point on the parts are subject to changes where the CAD drawings were not detailed, and where it was just impossible for me to figure out exactly  how the parts would fit together, even when looking at a 3-D model.  Perhaps it is because this is the first time that I built something with more that a 2-D drawing on paper. So the construction now is more like building a house; something I understand.  The plans and patterns give direction but the detail in left to the builder.  This will drive most engineer types nuts, but the carpenters of the world will feel right at home.

(2) If you need to push something into place, there is nothing like some 4x4's and a 2 ton bottle jack.  The cabin's rear bulk head has a bend so that it slopes back like a chair.  Every cubic inch of air that can be squeezed out of the cabin is that much less lead ballast that has to be carried around.  If you think this cabin looks small now, wait till its fully enclosed.  (3) You can put a sharp bend in 3/16 inch aluminum with a break press or by cutting a groove where you want the bend.  There is not much stress on this bend, so I cut a groove with 3 passes of the skill saw removing about half of the material.  Even with that it took my wife and I jumping up and down on the piece to get the bend started, and the hydraulic bottle jack to finish it off.  The neighbors got a kick out of watching two 40 something's bounce up and down in the driveway like kids on a bed, but it was also a great confidence booster to see just how strong this material is.

(4) The welding on the outer frames caused them to bend inward about 3/8 of an inch, so bracing was required to force the frame back to where is was suppose to be. Then the cabin's upper side frames could be fit in place.  Since the upper section meets the lower section at a 12 degree angle, the webbing between the upper frames is 1 1/2 inches as opposed to the webbing in the lower frames they rest on which is 1 5/8 inches.  That difference and some grinding is enough so that the joint between the upper and lower sections is horizontal.  This was done just to make to joint a little easier.

(5) The upper cabin framing is in place, and the bracing has been removed. The header across the front windows is now supporting the center frame in its correct position so the bottle jack has also been removed. (6) One half inch flood and vent holes are drilled at the appropriate locations so that the inside of the frame will flood when submerged.  This will help to decrease displacement and ballast weight requirements.

December 15, 2004 and after a 4 week break, and closing in the shop with a few more tarps in order to keep the cold out, I'm back to work.  (7) The cabin's bench seat is in after more bending on the lower hull, to form the bow.  I sloped the seat forward just a bit so water will drain onto the hatch when the boat is trimmed. Since the bottom hatches will be the lowest points I am planning on drilling a small hole through the hatch and installing a spring loaded valve that can be opened to purge water from the floor of the cabin.  The forward edge of the bench, the center and side frames, and a flat piece under the heel rest form the hatch opening.  (8) The heel rest is made from an 1 1/2" pipe that was cut in half. An additional flat bar under the heel rest and turn up on edge (see photo) was latter removed as the heel rest does not require a stiffener and additional structural framing to protect from impacts would be provide by the lower hatch frame.

(9) The bottom hatch frames are going together.   They are constructed from 3/16, 1 1/2" 6061 T6 angle, and form 2 frames, one which fits inside the other. They will be held together with stainless steel screws and they will sandwich a 3/4 inch sheet of cast acrylic.  The space between the frames is actually 1 inch in order to allow for the acrylic to  be bedded in a 1/8 inch layer of silicone on each side. 

(10) The Boy", aka Carl, my son, was home from the Army for Christmas Break so he got a taste of the grinding work that lays ahead.  Having done enough grinding that I know it's the part I truly dislike, I plan to utilize him on between his Combat Medic training and his likely deployment to Afghanistan this Summer.

Acrylic View Ports

Deciding on 3/4 inch acrylic was not an easy task. The submerged speed applies only a small force.  If we hit 3 knots I'd be surprised, but in a 6 ton sub in a lake with 20 foot of visibility, I'd be nuts to want to go any faster.  The acrylic will be 3/4" but that is intended to stop the submerged log from coming through it.  The greatest normal force will be on the lower foot rest sections when they hit a wave on the surface where we might be moving at 20 mph. The foot rest are about 7 x 22 inches and I don't remember what the impact load was, but I looked like it would be sufficient. :)  Since the cabin is ambient the other big loading force is on the upper view ports because the air trapped inside wants to get out.  It' kind of like an aquarium in reverse -- take an acrylic aquarium, empty out the water, turn it upside down, put it in a room full of water, and fill it with air.  Hence I got my best information about sizing the acrylic from aquarium builders.

(1) Lower hull extensions.

(2) Bending and tacking.

(3) Close-up

(4) An angle is temporarily
tacked to the side of the hull to
support the top of the hull
extension until it is welded in


(5) Scrap aluminum is botched
together to make a temporary
way to pull the side of the hull
out to meet the top of the hull


(6) Another brace to pull the side
of the hull extension out.

Lower Hull Extensions - Pontoons

(1) I guess I'll call them pontoons for lack of a better term.  Perhaps running boards is better?  They are really extensions from the lower hull, and if I did the math right they will be partially awash when the boat sitting still in the water.  They provide additional displacement to reduce the draft, additional surface area in order to help the boat plane, and serve as a swimming deck.  They are not extended above the water line so that they do not needlessly increasing the boats cross section.  Limiting the cross section will improve the top speed while submerged.

(2) (3) The pontoon is only 12 inches wide so bending it up was not nearly as challenging as the bow.  The plan was about 1/4 inch off on this piece between the bottom and the inside hull.  A 3/16 inch piece of scrap was used to fill the gap and small stitch like welds that bridged the gap were added until they were about 1/2 inch apart and a continuous weld could laid down.

(4) Inside welding was completed and then a 3/4 inch piece of angle was added in order to provide a temporary brace for the top. Once the top was in place the angle could be twisted from the rear and the tack welds broken.  This ended up working just fine, but since the rear of the boat is only 10 inches from the back wall of my shop, I will have to wait before I can recover my angle.

(5) (6) Having scrap aluminum laying around makes it easy to quickly knock together the perfect brace and pry bar for each special requirement. The top of the pontoon had to be pulled in against the side of the hull and then the side of the pontoon had to be pulled out to meet the edge of the top.

The bottom outside edge was then ground smooth, in preparation for a continuous weld that will eventually cover the seam.  If you don't mind a little danger you can reduce the grinding requirements by using your circular saw and router free-hand.  I trimmed about 3/16 from the outside edge of the pontoon's bow and beveled the bottom edge of the entire pontoon.  Before you do this, you should know that this is how people end up with 8.5 fingers. 




(1) Frame behind the cabin
bowed after straightening the
center frame.

(2) A relief cut in the frames
allows them to straighten out.

(3) Bottom T-frames added.

(4) Ballast sled that will hold
2000 pounds of batteries and
lead, and be moved forward and
backwards on three rails that run
from the cabin back to the engine

(5) Ballast sled and lid installed
with a lead screw for moving it.

Framing the Main Section of the Hull

(1) "For every action there is an equal and opposite reaction."  In this case it pulling the hull up to form the bow, bowed the two center frames out of place.  (2) A relief cut relieved the pressure but cutting frames once they are in place is a bit of a problem.  The circular saw will cut about 2 1/2 inches, then a reciprocating saw can get all but about the last 1/2 inch and a couple of holes will weaken the remaining material and allow the saw kerfs to close up.  Welding the joint back together will further pull the joint together, but be sure to place a temporary plate behind the butt joint in order to prevent the weld from burning through.

(3) Thirty T-frames on 10 inch centers have now  been welded in place starting from the bow back to where the engine mounts should begin.  Each frame is 1.5 inches tall, with a 1 inch tee from the same 3/16 sheet used throughout boat.  Rat holes were precut in each frame to allow for drainage down the keel.  A good trick here is to cut yourself a couple of sticks that space the frames properly and support them at 90 degrees to the bottom.  Each frame was intentionally cut a 1/16 or so long and ground to its final length for a snug fit. 

The three longitudinal frames will each have an angle welded on top the the 90 degree corner pointing up.  These angles will form the rails for the corresponding V-grooved wheels that I cut on the lathe that will be build onto the ballast sled. 

(4) The ballast sled will carry about 1 ton in batteries and lead forward and aft as needed to trim the boat for diving and surface running respectively.  Waldo (aka Carl my son) is sticking is head up through what will be the cabin's top hatch where he is still grinding away on my sorry looking welds.

(5) The rails are parallel to within 1/16 on an inch but I expect to warp the hull about 1/4 of an inch due to the 2000 pound load on the sled.  For that reason the sled will have two casters with 5 wheels each on each of the 3 rails, for a total of 30 wheels. Each wheel will have about 1/16 of an inch play on its axel so it can float from side to side, and rubber pads between the casters and the rest of the sled will provide additional flex allowing for the load to be distributed amongst several of the 30 wheels at any given time.  Details are here: "Ballast Sled".


Top Hatch

Well... it's been a month since anything has been done on the hull.  I have been learning how to cast aluminum in order to start making some of the smaller parts, like the trolley carriages for the ballast sled. So since its been raining like the days of Noah I started work on the top hatch. 

(1) After many hours of designing the fast acting water tight hatch, I finally dry fit the top and cut the opening out.  The physics of an ambient sub dictate that the hatch open inward, because once submerged all of the air pressure inside the cabin is pushing upward. Just think of it as a large bubble trapped in an upside down cup.  The air trapped inside wants to head toward the surface so it is pushing upward on the compartment.  (2) When I played around with an inward opening hatch inside of my cardboard mockup, I convinced myself that there was not enough room inside the cabin for my wife and I as well as the hatch. Fortunately a second test with my wife an I in the real cabin proved that its easy to lean a bit forward and allow the hatch to drop down and back behind our heads.  It's a relief to know that the hatch will easily seal and is very unlikely to structurally fail.  The new hatch is still quick opening but it now only needs four dogs or just enough to compress the seal so that it is air tight.  Details of the hatch parts and O-rings are here.

(1) Outward opening top hatch
in its closed and opened position.
This design was abandoned in
favor of a safer inward opening

(2) Inward opening hatch with
external handle. This was
abandoned as there was no
room for the outside handle
when the hatch was opened.

(3) Final design for the inward
opening hatch once it was
discovered that there was no
way to accommodate the
external handle.

(4) Bottom side of top hatch.


(5) Top side of bottom hatch
showing welds used to correct
for the warp caused by welds on
the other side.

(6) Second sheet on the top of
the hatch brings the hatch flush
with the top of the cabin.

(7) Hatch Closed

(8) Hatch Opened

Once submerged the seal will further compress and opening the hatch while submerged will be impossible unless the cabin is completely flooded.  Flooding the cabin will only be done in an emergency, and unless the sub is already on the bottom flooding the cabin will cause a significant lost the buoyancy and instigate a rapid decent to the bottom so getting out quick will be important and I don't want to be fooling around with more than one leaver.  If the situation has deteriorated to this point is also implies that there is something preventing you from exiting the cabin through either of the two bottom hatches, like the ocean floor, lake bottom, or a net.  The bottom hatches  are the preferred emergency exit path since they can be opened while there is still air in the cabin and before the sub is rocketing downward. 

Oops ...After all the calculations and Carl doing a beautiful job turning the parts for the top hatch we played around with fitting it into place and learned that I did not account for the outside handle when I decided that the hatch would slide inside the hull and rest against the back wall of the cabin.  With the outside handle the hatch would be 2 1/2 inches from the back wall and way too cramped.  (3) After debating creating a recess in the back wall to accept the handle, we finally decided to ditch the handle altogether.  The only reason for it is to allow for rescue access if the occupants are unable to open it.  It's not a real loss, the bottom hatches can still be opened from the outside which is a safer entry route anyway, and its one less thing to leak air or snag an entanglement. So, now we have the 3rd and hopefully last version of the top hatch.

(4) The top hatch is now starting to come together.  Carl machined the spindle parts and the center hub along with inside bracing and guide for the dogs are in place.  The inside bracing was welded the the hatch thought a set of 5/8 inch holes drilled through each brace and the outside sheet was done the same way.  It is not recommended in the metal boat building books to weld sheet material together like this because of the problems caused when oxidation starts to form between the sheets, so I guess I'll see how bad it gets because I have to have a thin hatch in order to accommodate it when it comes inside the cabin.  I thought that I might be able to get by with only adding reinforcement to the inside but that was not to be the case.  Even carefully allowing the parts to cool between each weld did not prevent the hatch from bowing inward about 1/4 of an inch.  (6) The addition of a sheet on the outside does have the added advantage of filling the recess so that the top of the hatch is now flush with the top of the sub.  (5) The top sheet was welded on with the same technique as the internal braces, however the welds were not enough to correct the inward bow, so I made a couple of 1/8 inch deep saw cuts across the sheet and welded them closed.  That did the trick!  As the welds cooled they drew the sides of the cut closer together and pulled the bow out of the hatch.  In fact it was a bit too much but a couple 1 1/2 inch welds (without cutting a notch) on the other side corrected it back.

(5) (7) (8) The hatch has two notched pins that will slide on rails in order to allow the hatch to slid back over our head and hang against the rear cabin wall.  I since have removed the notches and only have the pins in order to reduce the binding.  Since the hatch only hangs on its' rails by two points there is nothing to keep it aliened as it is moved.

RTV for a Hatch Gasket

(1) Wax paper on one surface.

(2) Set seat the parts together

(3) Separate the parts after 30
minutes and allow the gasket
to dry for 24 hours.

(5) Peal off the wax paper

(6) Trim away the excess RTV.

(7) Rasp to roughen the surface.

(8) The rougher surface
improves the gaskets ability to
resist pealing up, and trimming it
at an angle and back from the
edge helps as well.

The two part RTV that is intended for making molds, and which I sometimes use for potting does not make a good gasket because it does not stick well to metal.  The RTV I am testing here, is the stuff you get at the auto parts store.  It comes in small tubes as well as calking gun size tubes which sell for around $15.

Most of the time you just put some of this between to parts and bolt them together. But my neighbor Travis recommended this method for making a reusable RTV gasket. Reusable because the hatch is repeatedly opened and closed.  You use wax paper between the two parts so the RTV only sticks to one side. (1) I used a spray adhesive to help tack the wax paper in place, the laid down a bead of RTV. (2) Then

genially pressed the parts together. (3) After 30 minutes the parts can be separated allowing the RTV to cure faster. Do not try to removed the wax paper. (4) Let the RTV dry 24 hours first, and then the wax paper will peal off without lifting the gasket.  (6) Finally, trim away the excess RTV.

(7) RTV is a soft material and it does not take much abrasion to peel up the edge. So I roughened the aluminum up with a rasp and made another test piece. (8) This did help but it also helped to trim the gasket back from the edge of the aluminum and trim it at an angle. Both make it more difficult to peel up. When it did peel away from the aluminum it stretched and did not easily tear and it would fall back in place where it would still be usable.








(1) Cabin with a portion of the
deck that has been wire brushed.

(2) Randi brushing the hull and
building upper body strength.

Surface Finishing

(1) Carl is modeling the cabin which now has the bottom hatches installed.  We have also decided on finishing the surface of the boat by simply brushing the surface with a cup brush and Randi my other assistant has volunteered for the job.  Running a 5 pound angle grinder will help build that upper body strength she wants.  The small patch on the front deck is a sample of the brushed look.  It will also be a great way of hiding small scratches. 

The inside of the cabin will  be painted mat black order to reduce glare and also to insure an air tight seal on all of the welds.

(2) Randi,  my other assistant, volunteered for the job for the job of brushing the external surfaces of the boat with a th a cup brush and a 5 pound angle grinder.  That will help build that upper body strength she wants too. 





(1) Berkeley 12JC-A jet pump.

(2) Jet pump test fit into the hull.

(3) Jet pump intake welded in.

(4) Framing for the rear deck.

Aft End Hull Work

It's May, 2005 and it has been a while since I was in the back of the boat.  (1) I am installing the intake for the jet pump. It feels good to be back working on the hull, even though cutting a hole in the bottom is not comforting. 

(2) (3) After test fitting the jet pump on the intake I welded in the intake. I could have bolted the intake in place but I figure I'll have no need to remove it and the welds should help it take the additional forces that will be applied through the gear box that is bolted onto the pump.

As mentioned before the details of how each item will be completed was not part of the original plans.  Doing so would have restricted the design and not allowed for me to apply knowledge that I gain along the way, so in keeping with the organic design process I marked out where the gear box and engine would go and the stood in the back of the boat and decided the best place to put frames for the rear deck that would provide convenient access to the engine and ballast sled.

(4) The rear cabin deck, frames and transom were all installed in one day with Carl's help.  It's nice to have him for a couple of days before he leaves for Iraq. I took advantage of the muscle put on him courtesy of the US Army to get the boat framed up and out onto the yard where I could completed the bottom welding and get it on it's trailer.

(1) Boat meet the light of day.

(2) Boat dolly.

(3) Rolled up on its side for
welding the bottom.

(4) Improvised crane to roll it
over onto the other side for more
finish welding.

Roll Out

(1) The boat finally meets the light of day.  (2) I welded a steel dolly with 4, 10 inch tires and leveraged the boat out of it's plywood construction cradle  and onto the dolly.  It was then easy to roll it out of the shop an out onto the lawn.

There was over 2000 pounds of aluminum purchased for the boat, but it's hard to say what actually ended up in the hull. I'd venture to guess its about 1000 pounds.  (2) Once on the lawn we levered it up an pulled the dolly out and then rolled it up onto its side and we blocked it there so the bottom could get it's final welding.

Here is a lesson: If you want to make a small shop look huge, just remove the 20 foot boat!  I had planned on rolling the boat back into the shop on it trailer, but that will not happen because the trailer it's too wide for either set of doors.  So now I am planning on building a temporary shelter to house the boat in the drive way just outside the shop.

With Carl off to Iraq, rolling the hull over onto it's other side just got a lot harder and slower.  I improvised a crane out of the 4x4's that were once the pallet the aluminum was delivered on, Kay bolted on a worm gear winch.  I welded on a couple of hand holds that served as lifting points onto the sides.  Even then we had to use an engine lift to help raise it, but it worked fine for lowing it back down.

(1) 6000 lb capacity trailer

(2) Reinforced fenders

(3) Randi winches the sub from
its' dolly and onto the trailer.

Trainee Takes Injury in Stride!
Actually she just bumped into a sharp corner.

Trailer Modifications

I considered building my own trailer but after looking around a little I soon found that I could buy a new trailer cheaper than I could build one from new parts.  (1) I got a Loadmaster Sportline designed to haul 6000 pounds from Pirate Marine out of Ohio, and with delivery the cost was less than $3000.  A used trailer would have saved money but finding a used aluminum trailer in Oklahoma is almost impossible.

I lowered the trailer bunks about 7 inches by cutting down the bunks and cutting down their brackets.  The idea is to get the boat to sit as low as possible on the trailer so that it floats off without having to back so far down the boat ramp that there is no more boat ramp. The boat will actually fit between the fenders of the trailer with less than an inch on each side, but I will reinforce the fenders add add a bumper rail if needed.  They will make nice steps once reinforced too.  Since the side will be covered by the fenders there is some finish welding to be completed before I drag the boat from its dolly and onto the trailer.


(2) Randi helped me reinforce the fenders today. They are now enclosed and each have 5 additional bolts holding them to the main I beam.  This makes the sturdy enough to stand on and protects the hull from objects kicked up into the wheel wells.

(3) After completing the finish welding the sub was levered and winched from its dolly and onto the trailer.

We are enjoying some beautiful weather for this 2005 Memorial Day, and I hope that next year about this time we will have the sub in the water. (Note: It is now 2008, Memorial Day and we now plan to be in the water by August, however we have built and house, healed Carl from wounds sustained in Iraq, married the kids, and I am now a grand dad, so hence the delay.)

With Carl gone I have been forced to recruit more labor from the family labor pool which means Randi and Kay.  Some days it has to be viewed as a long term investment, because it takes a lot of time when your help is starting from almost zero experience but we are making fast progress up the learning curve.  We did experience one training accident.  Kay went and found a sharp corner on the trailer with her leg but she is a real trooper.  I could not ask for a better pair of helpers or a better wife and daughter.




Kay demonstrates the A-Frame.

The pipe used to build the
A-Frame is 1/4 inch thick,
2 1/2 inch ID steel. The span
between the legs is 8 ft span.
And it can lift 2000 lbs.


A-Frame corner bracket detail.

..who says my dog is a chicken?


I have an engine lift but that just wont lift the engine high enough to clear the side of the boat when it's sitting on the trailer so I built an A-frame from 1/4 inch thick, 2 1/2 inch ID steel pipe.  The legs are 13 feet long and the cross piece is 8 feet long.  To hold it together at the corners I used 2 1/2 inch OD pipe.  Each corner is built from 3, 16 inch pieces welded together with angle for reinforcement.  The legs and cross piece just slip over the corner bracket and only 1 bolt is used the secure the cross piece so it does not slide out.  Each leg must weigh 60 pounds, so standing the thing up is a bit of a task, but it needs no additional bracing. The steel cost $75 from the scrap yard and  Harbor Freight had their 3 Ton chain hoist on sell for $65 dollars.

For a test run we hoisted the boat off the trailer so we could alien and secure the bunks.  We used the engine lift on the transom just to keep the boat balanced, but the hoist and A frame worked great.  I could not see any bow in the cross member.  We have since made a habit of lifting the sub with the engine which is over 2000 lbs.  This is best done using two host so that the load is closer to the supports and not in the center.


Engine Installation and Other Aft End Work.

I worked on the sub two days this weekend but can't really show much for it. I had planed on getting the engine mounts made but one distraction led to another and I only managed to get the brackets for the engine partially made. Seems like everything always takes more time than I ever expect it to.  It was a pleasant weekend, none the less.  We did get a lot of the new hull parts brushed and I even got a couple of hours work out of Kay.

(1) Fitting the jet pump nozzle to
the transom.


(2) Openings for air tanks were
cut through the side hull
extensions, and the engine is
rolled out for a trial fitting.

(3) Hatch cover bolted in place.

(4) Engine hanging from the hoist
so it can easily be test fit.

(4) Engine mounts bolted in.

(6) The engine on it's mounts .

(1) We can finally get a good photo of the transom.  The jet pump came with a flange but it's to bulky are requires the control cables to be routed where they interfere with our custom clean out port valve. So instead we are just using a coupled of flanges that compress a neoprene gasket against the pump.

(2)   Openings for air tanks were cut through the side hull extensions. A 300 cubic foot air cylinder will be housed in each side to provide air when we are not using the hookah compressor on the surface. The valves for these tanks will be accessed hatch openings to the inside of the hull.  This will also allow the side extensions to flood and drain.

(3) Kay fitted hatch covers for the air tank openings.  The hatch plate is held down with 1/4 inch bolts that are tapped into the hull. A neoprene gasket will be used to seal the hatch.

(4) The engine was suspended from the A-Frame so it could be lowered into position and measurements taken for building the engine mounts.  (5) The engine mounts made from 2 inch square tubing and they are bolted into longitudinals on the hull bottom. The photo also shows the dip in the center framing that accommodates the standard oil pan on the engine.  The center frame will be a tunnel through the engine compartment that will allow bilge water to drain from the front.

(6) The sitting on the mounts.  Next comes the lower half of the engine compartment that will keep the engine dry.



More Aft End Work

(1) There are bolts sticking through the transom that are holding on the outside flange around the jet pump, and those same bolt need to hold the flange on the inside.  To mark where those holes need to go on the inside I coated the flange with a thin layer of molding clay, then positioned it and pressed it against the bolts.  I have also used the trick of dabbing a little grease or calk onto the end of the bolt, but I think the clay works much nicer.

(2) The drain hole in the transom can not be in the bottom of the V because of the jet pump so water and trash kept collecting there to form a little swamp.  The problem was easily solved with epoxy mixed with an equal measure of talc. The boat was trimmed and then the mixture poured in to the space previously occupied by the swamp water.

(1) Modeling clay used to mark
where the bolt holes should be
drilled in the inside flange.



(2) Epoxy used to fill the bottom
V of the hull up to the drain hole.

(3) Modeling clay dike retains
epoxy around acorn nuts.

(4) Epoxy used to seal an seam
on the inside of the tail fin that
can not be reached for welding.

(5) Tail fin welded to the rear
deck acts as a snorkel and allow
the diesel engine to run while the
boat is mostly submerged.


(6) Aft inspection ports

(7) While the boat is off the
trailer, Kay set about cleaning up
the welds an brushing the side.

(8) If the seam was welded from
both sides then the external weld
is ground flush.
(9) Left - One of the forward
support legs extends from the
bottom of the hull.  It is driven
by a linier actuator inside that
is capable of lifting 400 pounds.

(10) Hand rails.

(11) Air supply port.

(12) Air supply port detail.

Having gone through a dozen of those 5 Minute Epoxy syringes I decided I should indulge my love for the stuff so I purchased a 1 1/2 gallon kit of the stuff from and now I have placed my second order.  Adding talc is a great way to bulk up the epoxy and get more for your money. Wood flower, aka: fine saw dust also works.  If your belt sander has a collection bag then you already have a ready source for wood flower.

The forward and aft trim tanks are formed by simply closing off the top section of the front and aft end of the hull.  These partitions must be removable so that the machinery behind them can be accessed.  They also have to be air tight.  The welded seams have been sealed with a coat of epoxy to seal any pin holes.  And the partitions will have a neoprene gasket against the frames and will be bolted in with 1/4 inch acorn nuts on the back side of the panel to prevent leaks.

(3) Welding aluminum nuts to a 1/8 inch sheet is a skill beyond my ability so the nuts are glued to the back side of the partitions with epoxy. Modeling clay is again put to use to make a dike to retain the epoxy.  The surface of the aluminum was first cleaned with acetone and then roughened with a sanding disk to give the epoxy a better surface to grab onto.  The acorn nuts are temporarily held in place with bolts from the opposite side. There may be a problem with expansion breaking the epoxy off but that will remain to be seen.  I smacked the nut on a test piece with a hammer and all it did was crack the epoxy so it should be more that sufficient.

(4) (5) While on the topic of epoxy I have also used it to seal up a joint on the inside of the tail/snorkel. The tail fin is larger than required for just providing air to the engine so it is partitioned so that the back portion will free flood separate for the actual air intake path.

The partition was welded to one side of the snorkel before the halves were put together but the other side of the partition was only spot welded in place through holes drilled in outside sheet.  In hind site it would have been best to weld flat strip onto the partition to make it into a T so that the it would have exposed a solid plate through the hole instead of just the 3/16 inch edge. 

So, as for the epoxy, I just poured some mix down the un-welded seam and when it look like it was sealed up I added a little clay dike on each end and poured another mix into the joint to make sure it was good and sealed.

As for the the tail fin; it is constructed from a 2-1/2 inch schedule 40 pipe that was cut almost down the middle and then the edges were grooved 1/8 inch deep to make a lap joint with the 1/8 inch sheet used for the sides.  The 1/8 inch sheet was simply pinched together in the rear and welded up.

(6) I have I have also added a couple of view ports on the rear deck that will allow inspection of the rear trim tank. Once the partition for the trim tank is added, this will be an easy way check the equipment inside. The view ports each measure about 5 by 7 inches, and will have a 1/4 piece of acrylic bolted on from the inside with a 1/4 neoprene gasket.  The 1/4 inch bolts are treaded to the 3/16 inch hull sheet and were wrapped with Teflon tape to seal them from leaking air from the trim tank.

(7) (8) The weather is warming up have we are back to working on the hull in that one hour between getting home from work and sunset. Kay has taken on running the grinder in order to make my welds look their best.  The external welds are just smoothed out if the seam is only welded on one side, and they are finished flush if the seam was welded on both sides

(9) The support legs or landing gear are driven by electric linier actuators. The details of their construction can be found on the Landing Gear page.

The landing gear will support the hull only when it is underwater. They allow the hull to come to rest on the bottom while maintaining sufficient clearance for the lower hatches to be opened.

(10) More hand holds go in as a part of the push-pull cable system that is part of the Emergency Surface System. These will also make it a lot safer to ride on the surface while sitting on the cabin top.

(11) (12) The air supply system is now in place and that included an air supply port on the back side of the forward trim tank. This port allows surface supplied air from a hookah system to be connected to the sub so that air from the high pressure system is not required.  It also provides two ports for air lines with scuba regulators that allow for the crew to exit the sub while submerged.

View Ports

Finally ready to start thinking about installing the acrylic view ports.  I've been worried about suitable mechanical fasteners but the boating industry secures their view ports in off-shore boats with glue, and the mechanical fasteners are minor if they are used at all.  Life-Seal is a popular choice, but it is made specifically for fiberglass. Sikaflex is another popular choice, their site is  It creates a bond so strong that destroying the acrylic is almost a given if you want to remove the view port.

Sikaflex is sold for marine as well as construction, but with different names.  The marine stuff is Sikaflex 295UV for the adhesive and Sika Cleaner 226 and SikaPrimer 209 for cleaning and priming the metal and acrylic surfaces. Sikaflex 295UV datasheet list the tensile strength at a whopping 450 psi. You can buy these from: or Merritt Supply, but check eBay too.  Sikaflex  295UV is $13.29 a tube, SikaPrimer 209 is $49.36 and Sika Cleaner 226 is $12.37.  There will likely be a hazards material fee.

You may be able to find "Sikaflex Constructions Sealant" and "Sikaflex 260 Primer" from a building supply and it is virtually the same stuff.

Masking tape off the edge of the acrylic.
Lightly sand the exposed edge with wet/dry 320-400 grit sandpaper.
Clean frame and acrylic with Sika Cleaner 226 or acetone and wait 10 minutes.
Paint on a thin coat of SikaPrimer to the frame, the bonding surface of the acrylic and the edges of the acrylic and wait 30 minutes.
Using the supplied V-Cut nozzle, apply a generous bead of Sikaflex adhesive to the bond surface 1/4 inch away from the edge.
Apply 3/16 inch thick, soft rubber or plastic spacers every 10-12 inches next to the bead.
Set the acrylic in place and gently firmly press against the spacers.
Clean up any squeeze out with mineral spirits.
Tool the Sikaflex smooth with a plastic spatula.
Once Sikaflex has firmed up, backfill the edge gap with additional Sikaflex adhesive.
Remove masking tape from acrylic once Sikaflex has setup..

The acrylic came from a local supplier. A 4' x 8' x 3/4" clear cast acrylic sheet was $417.79 including tax.


Materials  Epoxy Paint CM 15 with long potlife and lots of flex for sealing pin holes while darkening the cabin. $65/gal.  Epoxy and RTV
Acrylic - Plastic Fabricators, Tulsa. 4' x 8' x 3/4" - $390 -- Ask for Becky
Acrylic - Odd sizes: Check with Arrow Shield, Dallas 713-462-8021