New Drydock

It is a rare occasion when ComNavOps gets to sincerely recognize and praise a US Navy accomplishment but such is the case with the recent delivery of a new dry dock intended to support Columbia class ballistic missile submarines.
 

The new dry dock, designated “Atlas” was announced as received by General Dynamics Electric Boat’s primary shipyard in Groton, Connecticut … [1]
 
Atlas was constructed by Bollinger Shipyards in Louisiana and completed the ~2,100 mile (3,380km) journey to Groton on January 3rd. …
 
As detailed by GD Electric Boat, the dry dock stands at 618 feet long, 90 feet tall, and maintains a width of 140 feet.[1]

Dry Dock Atlas

As a reminder, the first Columbia class submarine began construction in May 2019 and will be delivered sometime around 2030.  This is a discouragingly poor construction performance but at least the dry dock is ready.  As a point of comparison, the first modern, tear-drop shaped submarine, the USS Albacore, a truly revolutionary vessel, was laid down in Mar 1952, launched in Aug 1953, and commissioned Dec 1953.  Adding nuclear power to the modern submarine, in the form of the USS Skipjack, SSN-585, resulted in the vessel being laid down in May 1956 and launched in May 1958 with commissioning in Apr 1959.  Clearly, over a decade to build the Columbia class is disappointing compared to what we used to do routinely.
 
Submarine construction woes aside, the delivery of the dry dock is a rare moment of accomplishment that deserves recognition.  Good job, Bollinger.
 
 
_____________________________

 
[1]Naval News website, “Electric Boat receives new floating dry dock to support Columbia-class submarines”, Ethan Gossrow, 7-Jan-2026,
https://www.navalnews.com/naval-news/2026/01/electric-boat-receives-new-floating-dry-dock-to-support-columbia-class-submarines/

Ohio Versus Columbia Cost

Just for fun, let’s do a quick check of the Columbia SSBN program costs and see how they compare to the Ohio class.  The costs should be about the same after adjustment for inflation, right?  The Columbia is just a modern repeat of the Ohio - same function, same basic sub so the costs should be the same.  Well, let’s see …

 

As a quick review and in order to have a basis for comparison, here’s a brief table of specifications and cost for the Ohio and Columbia classes.

 

 

	Ohio	Columbia
Length, ft	560	560
Beam, ft	42	43
Displacement, tons submerged	18,750	21,140
Missile Tubes	24	16
Inflation Adjusted Cost	$3.3B a (FY2021)	$9.15B b (FY2021)

 

 

^a Ohio:  $2B (around 1997) for final sub of class (1) = $3.3B (FY2021)

^b $109.8B total class procurement cost for 12 submarines = $9.15B average cost (FY2021) (2)

 

From this table, a couple things jump out:

 

Size.  We see that the Columbia is, essentially, a repeat of the Ohio class as regards basic dimensions so the costs ought to be comparable, allowing for inflation.  In fact, we see that the Columbia has 8 fewer missile tubes so the sub should be significantly smaller/shorter since the missile tube section is the largest section of the sub. 

 

Looking at the cutaway drawing of the Ohio class below, we see that 8 fewer missile tubes (4 on each side, so 4 tubes in the profile drawing) represents around 11 m (33 ft).  From that, one would expect that the new Columbia would be around 33 ft shorter than the Ohio and yet they’re the same size.  Since no new functions have been added that we’re aware of, what’s occupying the 33 ft of ‘empty’ space?

 

Ohio Class Cutaway Drawing

 

Similarly, nuclear reactors have gotten steadily smaller.  Here’s the best data I could find:

 

Ohio - The Ohio class uses 1x S8G pressurized water reactor.  The S8G reactor compartment for the Ohio submarines is 42 feet (13 m) in diameter, 55 feet (17 m) long and weighs 2,750 tons.(3)

 

Columbia - The Columbia class uses 1x S1B pressurized water reactor of unknown dimensions.  A concept drawing of the Columbia class suggests that the reactor compartment is around 30-35 ft long.  While this agrees with the reasoning that the reactor compartment should be smaller than the Ohio’s, a concept drawing is a highly suspect source of information and the information should be treated accordingly.

 

So, comparing reactor compartment lengths,

 

 

	Ohio	Columbia
Reactor Compartment Length, ft	55	30-35

 

 

Since no new functions have been added that we’re aware of, what’s occupying the 20-25 ft of saved space from the smaller Columbia reactor?

 

Adding the 33 ft of saved length from the reduced number of missile tubes to the 20-25 ft of saved length from the smaller reactor, we see what should be a reduced overall length of 53-58 ft for the Columbia.  Inexplicably, it seems that the Columbia is actually slighter larger than the Ohio and has a significantly greater displacement!  How can a sub with 8 fewer of the massive missile tubes and a smaller reactor be larger and have a greater displacement?

 

Specifications.  As we’ve already mentioned, the Columbia has been down-spec’ed compared to the original Ohios:  33% reduction in missile tubes and a smaller reactor.  The missile reduction is puzzling given that the original threat that drove the Ohio design not only hasn’t decreased, it’s increased as China has significantly more military capacity and potential than the Soviet Union did.

 

Cost.  As we’ve just seen, the Columbia is, essentially a repeat of the Ohio class but with significantly fewer missile tubes and a smaller reactor.  Setting the internal contents aside, the two subs have nearly identical dimensions and one would reasonably expect them to cost the same after adjustment for inflation and yet this is not even remotely the case.  Why?  Why should a down-spec’ed modern Ohio be nearly three times the cost of the original?

 

 

 

 

Speculation

 

Why has the cost nearly tripled for what ought to be a significantly smaller and cheaper submarine?  Well, we have no actual idea so here’s some speculation:

 

Numbers – The Navy has not only reduced the number of missile tubes per submarine but also the total number of subs with a reduction from 24 Ohios to just 12 Columbias.  It would be reasonable to assume that cutting the total build in half would increase production costs.  Of course, economy of scale is rarely seen in Navy ship programs so this is a highly debatable assumption.  Still, simply spreading a shipyard’s overhead costs over fewer ships is guaranteed to increase purchase costs (see, “Shipbuilding Costs – Impact of Low Volume”).

 

Berthing – Presumably, the Navy is planning to man the subs with mixed gender crews and the need to duplicate facilities undoubtedly increases costs.  Still, we’re trying to explain a nearly $6B cost increase so there must be more to it than gender related amenities and duplication.  The need to duplicate facilities may, however, explain some of the lost ‘empty’ space that we can’t account for.

 

Comfort – We’ve seen an across the board increase in creature comforts for crews as the Navy has moved to nearly year long deployments (see, ”Crew Comfort”).  Despite the idiocy of turning WARships into cruise ships, this may account for some of the missing ‘empty’ space and cost. 

 

Supra-Inflation – We’ve seen cost growth over and above the rate of inflation for almost every Navy shipbuilding program and the magnitude of the increases for some of them, such as the Ford class, have defied belief.  This may be more of that same unexplained supra-inflation.

 

Bid Inflation – The Navy has clearly demonstrated to industry that the Navy’s construction plans are etched in water and not to be relied on.  In fact, with Navy shipbuilding programs, it is almost a guarantee that the numbers will be cut before the program is over, thus negatively impacting industry profits.  Without a doubt, industry is aware of this and builds in some profit margin cushion to compensate for the expected decrease in numbers.  How the Navy could decrease the SSBN build below 12, which is already reduced from 24, is unknown and yet history suggests that it is a very real possibility.  For those who might think that a leg of the nuclear/strategic triad would be immune to reductions, one has only to consider the example of the Air Force B-2 nuclear/strategic bomber which was planned to build 132 and then was reduced to 21.

 

Gold Plating – I have zero evidence that the practice of unnecessary over-spec’ing, referred to as ‘gold plating’, is taking place in the Columbia program but, given the Navy’s constant tendency to do this, it would be surprising if this was not taking place to some, likely significant, degree.

 

 

Conclusion

 

We have no concrete conclusion that explains why the Columbia is 53-58 ft longer and nearly $6B more expensive than we can account for – only speculation.   Without a doubt, some or all of our speculation is correct although we cannot quantify the magnitude of any of it and none of it seems sufficient to explain the $6B increase.  Regardless of the actual reasons for the staggering cost increase, the Navy’s continually demonstrated inability to produce reasonably priced new ships is a monumental concern and is inexorably shrinking the fleet to a point of combat-ineffectiveness.  The latest evidence of this is the Navy’s turn to small, weak, unmanned ships to replace Aegis cruisers and Burke destroyers in an effort to keep ship counts up despite the resulting decrease in combat power.  The Navy simply must get control of its shipbuilding programs and learn to produce affordable, combat-effective ships or we will find ourselves without an effective Navy.

 

 

_________________________________

 

(1)https://web.archive.org/web/20090426041620/http://www.dailypress.com/news/dp-local_subs_0424apr24,0,5810806.story

 

(2)Congressional Research Service, “Navy Columbia (SSBN-826) Class Ballistic Missile Submarine Program: Background and Issues for Congress”, May 2020,

https://crsreports.congress.gov/product/pdf/R/R41129/179

 

(3)Wikipedia, “S8G reactor”,

https://en.wikipedia.org/wiki/S8G_reactor