Aluminum Watercraft Project- Sealing Compartments

[PX Machines]

Hi Everyone,

I have a project to build an all aluminum offshore watercraft 30ft, and after completely welding all exterior surfaces I need to seal off internal compartments, but would rather not seam weld everything. If I stitch weld everything internally, would it be acceptable to use some type of aluminum bonding to seal the rest of the seams? And then epoxy paint over everything? Any suggestions are greatly appreciated! Thanks, in advance!!

[kmorin]

PX,
I'll offer my experience and make suggestions about the question of fully sealed xverse welds on water tight bulkheads.

First the short answer is: weld them solid on one side (AT LEAST) and back stitch the other; at a minimum.

Long answer is: There are several considerations that have been seen in aluminum boat fabrication that may not be considered in your idea of stitching and "gluing" ? I'll go through some of the points for your consideration.

Primarily a structural item; in regards evenly and uniformly transferring loads from a skin/hull/"bottom&topsides" load to a transverse framing (not just ring frames) bulkheads form key structural element in offshore hulls. By stitch welding the main bulkheads instead of seam welding them (both sides welded continuously is not unheard of) you leave areas of thermal expansion/contraction that are not 'met' by the bulkhead to hull connection; So.... the metal movement is not uniform and the ends of the welds can begin to express fatigue into the hull plates- like happens with trailer rash at high speeds on the highway.

Also, you leave the collision/impact joint subject to failure points (especially with a bow or collision bulkhead) and

[Secondarily] you leave the obvious 'flooded state' containment at risk of communication of bilge water between chambers where sealed bulkheads confine any catastrophic flooding to the chamber where the hull was compromised/holed/breached.

Any sealant material/goop/glue/calk is subject to trapping moisture & forming a corrosion cell forming an intersection around the hull! Welds, properly prepped and performed do not.

Any sealant material/goop/glue/calk adds nothing to making the hull a single solid metal 'honeycomb' of fully fused integral shapes interlocking into a rigid, reinforced shape with the ability, built-in; to confine the bilges from one another- equivalent to a weld's fusion of the parent metal.

So, the strength isn't the same with calk, the seal has a life expectancy with calk (where renewing the calk might be a real rodeo!! ??) and the potential for corrosion cells exists with calk but not fully welded.

So how does a properly fit bulkhead happen? The bulkhead is fit, either from the plans/NC cutting or by fitting methods after the hull is tacked up. (or in the case of NC cut hulls or traditional tack up w frames; before the hull panels are tacked up.) Either way- the fit shouldn't allow more than a sheet of paper (few thousand's of an inch) gap. Then the edge of sealed bulk heads are beveled either one side fully or both sides partially to allow a continuous weld on one or both sides.

Forward of the Master Station, where the forward face of a bulkhead touches the hull panel continuously, there is no point to bevel- those bulkheads where the after edge is slightly forming a 'natural bevel' should be beveled slightly on the after side - like the collision bulkhead or forward cabin bulkhead.

At the Master Station or the after bulkheads, the hull shapes of most planing boats, in Plan View, are very nearly parallel so a choice can be made about the side to bevel (or both sides) and the welding should proceed as follows.

The bulkheads should be marked along their outer edges on the beveled side for tacks. The tacks should be marked so they are arranged at the beginning and end of all welds - the bulkheads' continuous welds should be chained/stitched/put down in 6-8" stringer type welds with MIG.

The weld type is important so as not to warp the hull panel any more than needed and where the seal is achieved by a narrow "stringer bead" that just fills the bevel and wets both the hull panel and the bulkhead fully. Puddled/patterned/whipped welds are inappropriate for this weld joint. The MIG torch should be continuous moved and the voltage very high with high wire speed for extremely fast travel; down hand welding is completely acceptable and much more desirable than uphill.

Each area of the hull where a xverse frame or bulkhead, or longitudinal will be welded needs to have the mill scale removed with a wire wheel (SS only, wire size 0.014") of small 4" dia. OR a Scotchbrite (tm) type buffing pad so there is no mill scale under the added material. Critical to solid weld fusion and to hull longitudinal's integrity to the hull panels' strength.

Once tacked in; all tacks need to be dressed, so they're 1) uniform in cross section; 2) hollow 'inside' the bevel; 3) still fused to both the bulkhead and the hull panel- and not cracked due to dressing too deep. Best tool for this I've found is the die grinder with 1/4" tear drop bit. These chips will blow out of the joint and vacuum up easier than sanding particles and hard discs/ flap sanders/ other grinder type solutions can leave dust that will contaminate the roof face of high MIG welds.

The fist set of welds should happen at or near the long seams of the hull; keel, chine(s) and sheer. If the hull has more long seams they should be welded with the same priority as these listed longitudinal seams. This 'tensions the bulkhead' and the hull at previously reinforced long seams. These initial welds should SPAN the long seam- not end or start on the seam welds. So a bead would be planned to start 3" above the chine, cross the chine weld, and end 3" down the bottom hull panel (in a single chine planing hull).

Once the the bulkheads are located, tacked, tacks dressed and all seams air blasted for dust/sanding particles; the entire seam should be wire wheeled again using the same 4" wire wheel with 0.014" wire (only). IN the corners where a 4" inch wheel can't clean- an end brush on a die grinder (still SS) works to get the tack weld's soot and dross out of the corners. NOTE: no tacks should be within 2" of any hard corner, seam or longitudinal welds.

After each series of welds are added- the ends that will be 'tied too' should be gouged using a high speed carbide (only) die grinder bit and the 'tooth brushed' in preparation for the next series of welds.

The pattern of adding the welds two both sides begins with the decision to use a single bevel- one side continuous and opposing side stitched, or a double sided bevel with both sides continuous?

IF you use the first scheme then add welds in series symmetrically about the bulkhead- any stitches on one side are followed by those backing up the other side. This way, all the stitched side welds are done after about 40-50% of the opposite welds are in. The stitch sided welds are all 'wrapped' while the continuous side welds are not.

IF you use the second scheme of a double sided fully welded bulkhead; then do one side fully and repeat on the other side. None of the welds would 'wrap' starts and stops as both ends would be gouged clean using the carbide burr before the next pass was added to tie in.

I realize you didn't ask to have "War and Peace" length description of bulkhead welding but.... that's what you got. Hull longs (all types), hull panel transverse seams, and the location in the hull all play into the decisions to weld in bulkheads.

The worst case is you 'butt' the bulkhead to the hull plate/sheet (inside) and add some 1/2" wide, whipped puddle between the two as though it were a fillet in a tank baffle!!! That will warp the hull. Beveling the bulkhead is very important to resulting in a fair hull.

No matter how you weld you'll have "print through" where the bead's heat will bulge the hull plate outside the hull- even if you stitch. So running fast little stringers, under a beveled edge, down hand at high voltage and wire feed is the only way to get a bulkhead sealed properly with a print through you can sand off - if needed?

Cheers,
Kevin Morin
Kenai, AK

[kmorin]

PX,
after posting the bulkhead remarks I'll follow with the actually correct method of installing ring frames/transverse frames/partial bulkheads that won't be welded continuously.

First remark is; what do your plans call for in regard fully welded xverse bulkheads? I'd follow the plans?

Now to xverse elements of all kinds that will be 'stitched' or added by welding to the hull panels' inside surfaces but not with continuous welds.

All these elements are like a T fillet weld to the hull panels/sheet/plate. SO... each of them should be beveled where each weld will be added. That is; if a stitch pattern is 4" forward side, skip 4" and 4" on the after side (as an example only) then the bevels for each weld should be where the eventual welds will be put down.

By putting a weld 'under the edge' weld distortion is radically reduced. Weld contraction- or the 'hungry horse sided' welded boat comes from allowing the face of the weld between the hull and the xverse member to be wide. The wider the weld face- regardless of the root face dimensions- the greater the contraction ACROSS this face. This is not proportional- but nearly exponential, where a 3/16" weld face doesn't pull up the two adjacent surface 3x and a 1/4" weld face pulls up 4x... that would be linear relationship.

Instead a 3/16" weld face pulls less than 1/3 the pull or face contraction of 1/4" face with beads. In the above remarks that means the 1/4" face weld pull 12x! Imagine what your hull will look like if you have 1/2", whipped beads w/o beveled recesses!!

Stitch welds should have wrapped ends- starts and stops should not be gouged or cratered and this means testing on the bench! Proof is in the weld test.

This followup post is to remark that beveling for 100% of all transverse framing welds is key to a fair hull. This implies planning and extreme welder qualification. I've had entire hull's warped irreparably by welders who were to stupid (yes; I used the term correctly, derogatorily, and accurately) to do what they were told! So if you can't get the exact welds needed out of your welder; get another welder!!! Once welded incorrectly; there's no way to recover the hull shape, but if done correctly your build can be attractive and build your rep.

Cheers,
Kevin Morin
Kenai, AK

[PX Machines]

Mr. Morin,

Thank you so much for such an in-depth explanation of how you approach this!!! No I didn't expect war and peace length but I did spend quite a bit of time dissecting your well thought out response and sincerely do appreciate such a thorough breakdown of how to approach this! I did look for your profile on linked in to send you an invite but could not find you! I do understand there are significant material property differences between 5083, 5052, 6061 and as you say "goop" :) I guess I have a hard time putting all the "goop" in the same bucket :) I mean in NON ALLOY boats its goop that bonds the glass together and in aircraft riveting and gluing is what hold the aircraft together. For this reason I have a hard time believing that using aircraft epoxies in this application wont work well, I just dont know what the correct implementation is. I would think that material bonds, thermal changes, mechanical stresses would be at least somewhat similar to aviation applications and it keeps these machines in the air for decades. Yes, one local marine company using combination did have a gluing delamination issue which contributed to the sinking one of their boats, but in all the pictures I saw there were not near enough spars, ribs and stringers to mitigate moment of inertia flex across the entire span of the boat hull in all 3 axis. It just makes sense to me that some things can be welded and others put together using structural adhesive complementary to each other?? I'm sure there are many things I have not taken into consideration, and do not want to disrespect your many years of experience either. I continue to welcome input for sure and when we are done I will post up some pictures of our watercraft.

Very Best,
Pete

[kmorin]

PX,
Kevin is fine! as the saying goes "Mr. Morin was my father". Anyway, glad you got some additional perspective from my remarks. I'd like to reflect a bit on using applied elastomeric products to a hull and contrast that to aluminum aircraft.

I'm sure you've noticed the leading welded boat manufacturers- and builders- where the difference is the latter usually build one off for a client while the former build full time and have distribution like an automobile manufacturer; don't use the technique you're describing? That is welded boats, even riveted boats, aren't glued together but the riveted boats do rely on a rubber/plastic gasket in the riveted seams in some instances.

So, I'll just say; best practice is to weld in welded boats where many locations can be stitched or chained and obtain the required structural integrity required in 'that' location. However, relying on non-weld fusion seals for hull chambers is roughly equivalent to tacking a metal tank together and using fuel proof calk to 'seal' the tank. Might work? But isn't widely practiced- not even on airplanes!

Now to metal airplanes' features that can't be related in any way to welded boats. Air is much less dense than water; so much less dense that object, like bullets, can travel through it at high speeds- supersonic speeds even. But it still remains less dense than water even at super high speeds.

Water 'hardens' the faster you go on it, or through it. For example the 50cal Barrett rifle's huge, super fast bullet can travel through air with incredible momentum. But, when that same bullet enters water it disintegrates within inches of impact due to the Reynold's Number of the water shifting to become extremely high ("hardening" the water) in response the high speed of the bullet. The bullet reacts to water like concrete- not like air.

While an airplane may travel at high speeds in the air- it doesn't in the water. So sealing a seam with goop in a riveted airplane is not equivalent to doing so in water due to planing on water imparting much more impact force on the boat hull compared to the wildest turbulent in air. Airplanes are structurally sound as long as the skin remains in the original shape- think egg shell. However boats have to stand an impact countless times higher and still keep the skin/shape/shell/hull surface intact.

Now comes the big difference that I think renders your equivalency concept as inapplicable: Boat hulls have to be able to accept/absorb/accommodate an impact onto a rock, piling, grounding that is a point load of the entire mass of the boat and keep on going or remain integral and watertight. The same impact on the same area in an airplane would poke a hole in the thin skin.

To make a real world comparison- take a pick or an axe and swing overhand with your best 'shot' on a 3/16" x 12" x 12" piece of 5086 alloy aluminum on a 2x4" wooden frame. The axe bounces back up and no matter how much you repeat it will take a long time to do more than dent the hull panel sample.

Now, repeat the exercise with a 0.040" thick pc of 2000 or 3000 clad aluminum used in wing and fuselage skin? The comparison between the two results is because of how each material needs to perform in impact. The airplane is never suppose to see it, and the boat is supposed to experience it 100% of its useful like.

Areas around bulkheads that are stitched - may provided adequate strength but if there is a collision(?) a fully welded bulkhead will deform uniformly with the surrounding hull panel and remain fully sealed, full structural integrity and continue to seal the various bilge areas of the hull from flooding.

Generally, welded aluminum hull's in the 30' class are usually 0.187" (3/16") or thicker where panels below the waterline might be as heavy as 0.31" or 5/16". This is intended to allow an underwater collision without holing the boat.

An airplane is never designed for any collision of this type, and the slightest 'ding' in the skin/surface will 'ground' a plane and render it unsafe for use- where as, in a welded aluminum boat we expect to take that kind of event for the life of the boat.

Even though the two types of vehicles are both made of alloys of a similar base metal- aluminum- the alloys are so different, the scantlings are so divergent, and the design and structural requirements of the two sets of vehicles are so separate: I can't consider elements of one necessarily useful for the other.

What the two do have in common is fuel tanks. In both they're 100% continuous welded, air pressure tested and done with same alloys, techniques and design premises so; except for the scantlings of the two series of tanks, identical. But the tank alloy is not the framing/skin/or surface of the airplane and tanks not sealed with goop(s).

If you're working from plans, we'd enjoy seeing them, and following along with your build.

If this is your first welded boat? I'd suggest before you begin that you consider building a 20's open skiff as a 'warm-up'. I suggest 20'er as the smallest to build to avoid foam, level flotation regulation silliness and generally be a good size for a beginning skiff. Won't (shouldn't) take more than a hundred man-hours, in most designs, and would help give a feeling of what's what in a welded hull build. Also, as skiff project allows a crew to tune up their roles as foreman, fitter, welder, helper... etc. a 20' open utility skiff with clean lines and well placed, uniform and proportional welds usually sells well in any location so you'd recover the expense of warming up for a 30'er.

Cheers,
Kevin Morin
Kenai, AK

[PX Machines]

Kevin, thanks so much. You make some very good points and may have saved me some headaches or even worse! One thing I realized is the watercraft we are building where the bulkheads need to go could be seam welded without too much concern about of aesthetics. I am going to sift through your responses in detail and just need to do some faring work where the surfaces are shown on the machine. THANKS SO MUCH

Are you on Linked In?

[gandrfab]

May I suggest some reading.

boatbooks.PNG (551.3 KiB) Viewed 6907 times

[PX Machines]

Absolutely, thank you! Ive just ordered a couple of these books. Guess I should have done that first!!!

[kmorin]

PX,
All these books do have some good information; however, reading them does imply you should plan to build a 'plain jane' skiff before laying the keel(s) on a 30' offshore design!!!

The first dozen or so skiffs, like the first few power boats over 28' are really "educational" for the new builder in welded aluminum. The list of mistakes I made working up to these milestones is so long it would take a couple phone book thick books just to list. Let alone explain the reasons to avoid and "How to Avoid" them... (So) a word to the cautious businessman- warming up with a simple open skiff is a very small investment to insure you're ready to take on a bigger project. Especially if the books mentioned are new to your experience!

Cheers,
Kevin Morin
Kenai, AK

[gandrfab]

I would have to look to know for sure. If I remember correctly Ernest Sims talks about riveting.

[PX Machines]

I would have to look to know for sure. If I remember correctly Ernest Sims talks about riveting.

Most of the machines I have built with thinner aluminum skins were using pressed in flush pem studs where the skins were fastened to the weldment underneath and bonded in between with aircraft 3M or loctite structural adhesive. Worked great, but then not nearly the vibration of offshore planing boat...