Bill said : “one of those lessons is the incredibly low allowable stress levels that we must design with..and live with..in an aluminum ship structure”

Excellent post.

If I may repost some of the comments I made back in June 2009 :

——————————————————————
Another problem with LCS-1 is that interactions between the steel hull and the aluminium superstructure are not without major *challenges*, e.g. :

1) Aluminium is difficult to join to steel structures (you need to use either explosion bonding or biweldable strips).

2) Aluminium can lead to galvanic corrosion with steel.

3) Aluminium has a coefficient of thermal expansion almost double to that of steel, which may cause distortions with temperature variations in service.

These *challenges* are not just hypothetical : for instance, in 1991, when USS Princeton detonated an acoustic mine under the ship’s quarterdeck (the blast detonating another mine 300 yards off the starboard beam), a 6-inch crack opened in the Princeton’s aluminium superstructure running up one side and down the other, with more than 10% of the superstructure separating from the main deck.

Mea Culpa: six months is 4300 hours…finger-itis. It’s still a lot of hours on a frame, and if you multiply that out 20 years..

Scott, don’t forget the wonderful affect that high velocity salt air (rushing through various ventilation inlets) has on aluminum.

And I bet that crack was just aft of the main mast, or just forward of it…

Doc, given what I do for a living, which is ship repair/construction, I will tell you that the life cycle cost for aluminum is much higher than steel. Period. Dot. End of sentence.

Doc, do you have any idea as to what it costs to work with aluminum? The special restrictions applied by NAVSEA quality assurance? NDT requirments?

Here we go:

In the “critical zones” (which are the areas most affected by stress in a ship, the mid-body), these restrictions take affect. In the fitting process, plasma arc is not allowed for cutting purposes with respect to the final cut…which means if you plasma cut (the most efficient means, and also the most common)you have to cut off a further one inch of material…with a saw of some sort. You are not allowed to use any sort of lubricant on the blade, as this will affect the NDT process. There are restrictions on other tools, including the welding processes, the brushes you can use, the methods of contouring and beveling the weld joints. A NAVSEA check point for fit up also has to take place, and the weld joint must be “floated”, with no tack welds in the weld joint. That means a LOT of temporary attachment welds, which by the way also have to have NDT. The weld site has to have weather and wind tight containment. The weld gas must be tested twice a day to insure proper dew point. Once all of this is done, welding can now take place.

Now the hard part starts. Every inch of every pass must stay completely free of any sort of debris, so there’s a lot of cleaning that goes on during the weld operation. Once the weld process is complete, the weld must then be further contoured so that at no point in the weld joint can any dye penetrant be trapped. If you can imagine the places on the underside where all manner of structural elements are in place, you also have to apply the same weld restrictions. Think LOTS of grinding…lots. And all with tools that are anything but high production. To further add to the welders misery, he has to contend with the existing conditions in the parent metal.

Now that the weld is ready to present to QA and NAVSEA, the really ugly part starts. Every inch of weld gets dye penetrant tested…every inch. Butt welds (where two plates come together) are also tested with ultrasound to detect interior porosity. No matter how good the welder is, this process has never in my experience passed the first time. There’s always repairs.

Now steel? The same insert will get a visual inspection and one or two UT shots. Now you know why aluminum is so damn expensive to work with.

When you talk about aluminum, keep in your mind that it’s a damn sponge. Most metals are porous, but aluminum is a sponge. Oil, salt, hydraulic fluid, any of the different things you find on a ship will get impregnated into the parent metal.

Bottom line is that the damn stuff is way more expensive that working with steel…especially when all you can buy is ASTM-B-928 (take a guess how many mills make this material AND will supply a certificate of compliance to the buyer?).

Now, just to further complicate the issue :

1) The high thermal conductivity of aluminium means that fire boundary needs to be contained by boundary cooling, otherwise fire will propagate throughout the ship.

2) Boundary cooling traditionally requires fairly large fire fighting teams, which makes it vulnerable to temporary / permanent loss of personnel.

3) How is this supposed to work with the skeletic manning currently envisioned on both LCS designs ? Automatic fire suppression ? Mmmhhhh….

Next problem with aluminium is one that, strangely, doesn’t receive as much attention as fire-resistance : aluminum tends to produce splinters when hit.

Again, the problem is not just hypothetical, as Norman Friedman explained in his description of the USS Worden incident (hit by a Shrike missile off Vietnam in 1972) :

“Her aluminium superstructure acted to multiply the fragments produced by the missile; every pellet from the missile produced two or more in the superstructure, so that instead of a shield it became a deadly instrument in its own right”

(Norman Friedman, Modern Warship Design and Development, page 168).

You could off course somewhat mitigate the problem with anti-fragment materials, but then again, the impact in terms of weight and cost is not insignificant.

“aluminum alloys have a low resitance to temperature, with a softening point @ 200°C and a melting point @ 600°C, meaning aluminum structure will suffer structural collapse much faster than steel structures, ceteris paribus.”

Let’s put that in sailor terms. It means that the all aluminum bulkheads transmit the heat to the ladders, WTD hinges, hatch covers and decks very efficiently, so they “soften” well before the compartment ambient temp reaches 200- 250 degrees. This happens very quickly and relatively far from the site of the fire.
Decks sag under the weight of equipment resting on it,bulkheads sag and rupture, spilling flammable liquids, ladders lengthen, sag, burn flesh touching them and give way at loads well below the weight of a man. Water tight door hinges sag and give way under the weight of the door and are no longer gas or water tight.
Flammables touching bulkead or decks flash into fires distant from the original one.

Deathtrap.