The Atlantic Generating Station

Recent announcements and news stories about a Russian project to build a floating and essentially portable nuclear power plant have been variously tabbed with the heading “new.” The idea of a floating, mobile nuclear plant (which is not self-propelled and not a ship) is indeed not new—the nuclear barge STURGIS, itself a converted Liberty Ship, served as a power source for the Panama Canal for many years, beginning back in 1967. The new Russian plants bring extra excitement because they are classed, properly, in the now-popular small modular reactor plant category, having been based on true seagoing designs. This, of course, hints at the fact that their output will not approach that of any of the large, conventional nuclear plants familiar today.

For the historian, the question might then come to mind as to what the largest nuclear plants ever seriously considered for construction to such a design were. The answer is very simply this:  Full size, commercial nuclear plants in the normal (>1000 MWe) range common today. While the plants weren’t to have been fully mobile in the same sense, they were to have been barge–mounted and would have remained floating while in operation.

In the late 1960s, Public Service Electric & Gas (New Jersey) began to invest very heavily in nuclear energy. The company bought a major investment in Philadelphia Electric’s Peach Bottom expansion, and also began to order units of its own. In 1966, PSE&G ordered Salem Unit 1, followed in 1967 by Salem Unit 2 (both from Westinghouse);  in 1969, PSE&G awarded a contract to General Electric for its Newbold Island nuclear station (two units), which eventually would be cancelled for siting reasons; however, with that cancellation, simultaneously the project was moved next to Salem to be built as Hope Creek. According to PSE&G literature of the period, because of increasing worry about the thermal effects (waste heat) of nuclear plants, it decided to make its next order for a nuclear plant a bold, radical step; it decided to contract with Westinghouse to construct nuclear plants essentially at sea, in a man-made structure and mounted on floating barges.

The site eventually chosen after some consideration and study was as shown here in an original advertising illustration from a PSE&G brochure on the project. The caption reads: “The proposed offshore site is 2.8 miles out in the ocean, off Little Egg Inlet, and approximately 12 miles north of Atlantic City.” The location chosen kept the nuclear station out of major shipping lanes.

The site itself would have been prepared (with a breakwater surrounding it) including two moored, side-by-side nuclear power plants, separate from each other but identical and which would have been mounted on gigantic rectangular barge structures. According to the PSE&G brochure, “The Atlantic Generating Station,” the construction process would have been as follows:

“The breakwater will be the largest and strongest structure ever built in the ocean. First, concrete caissons will be floated to the site, sunk, and filled with sand and gravel. Next, thousands of tons of rock will be brought by barge to create the artificial reef, within which the plants will be moored. The mound facing of the reef will consist of sand, gravel, and stones topped by an armor of interlocking pre-cast concrete units called ‘dolos.’ A typical large dolos weighs 42 tons and measures 20 by 20 feet. Approximately 70,000 of these dolosse, in various sizes, will be placed on the breakwater.” 

The structure and the plants were designed to survive 43 foot waves, sustained (continuous) hurricane winds of 156 miles per hour and tornado winds of 300 MPH.

Above, cross-section view of the installation as planned. (Our apologies for the slight imperfections in some of the illustrations, which are contained in vintage materials, are not always printed perfectly, and which are often printed across the center staple fold.) Below, an artists’  illustration of the plant, whose official name was in fact the Atlantic Generating Station, from the air.

The two plants that were to become the Atlantic Generating Station (AGS) were first announced in 1971, according to WASH 1174-71, but were not named at that time, nor was a location specified. In September 1972, according to the Atomic Industrial Forum (now the Nuclear Energy Insitute) report “Historical Profile of U.S. Nuclear Power Development,” 1985, the two plants were officially ordered from Westinghouse as Atlantic-1 and -2. (The reactor plants were to have been 1150-MWe four-loop PWRs.) The plants were to have been built at a wholly new dedicated facility in Jacksonville, Florida, as a part of a joint Westinghouse–Tenneco operation known as “Offshore Power Systems.”

PSE&G had printed, in its public relations materials of the time, that it intended to rapidly increase its nuclear generating assets. From a 1976 brochure on Hope Creek: “To prepare for the coming ‘electric economy’ when electricity will play an even greater role in our daily lives, PSE&G is relying on nuclear energy. From now until the end of this century, all new major generating units will be nuclear. It is our intention to phase out our oil and coal burning plants and eventually have approximately 50 percent or more of our electric capacity in the nuclear stations we share with other utilities. This nuclear capacity will provide approximately 75% of our energy needs by 1990.”

To that end, in November, 1973, PSE&G ordered two further nuclear units of the same type as ordered previously as Atlantic-1 and -2. While the AIF document previously mentioned does not give a plant name or site for these, a later Energy Information Administration/Department of Energy document identifies these plants as Atlantic-3 and -4.

The nuclear plants would have been mounted on barges approximately 400 feet square. The draft of the nuclear plant barge structures (that is, the depth to which they extended underwater) would have been roughly 30 feet; the breakwater/reef enclosure would have had a further 10 feet of clearance under the plants for water flow. In an interesting nod toward today’s AP1000 plant and its modular construction, PSE&G said about the AGS units that the shipyard fabrication plan “allows for assembly line production techniques, as well as standardization of design and licensing procedures—which will result in reduced costs and planning lead times.” Heavy underwater cables, instead of high tension towers, would have connected the plants to the grid. A shore base would have been built, to shuttle workers to and from the AGS and to station repair parts, consumables, and any other requirements for the nuclear station several miles out to sea.

Above:  “Artist’s conception depicts how the plant will appear on a clear day to a person standing on the nearest beach.” The illustration is meant to dispel the fears that the plant would be an eyesore.

Of course, we all know today how this overall plan played out. PSE&G did not experience nearly the expected growth in electric power demand that it had predicted. While a 1976 PSE&G brochure on Hope Creek also prominently features the Atlantic Generating Station, a 1977 brochure on Salem does not mention it at all. In 1978, PSE&G cancelled all four units ordered for its offshore nuclear power station program, and the AGS project died immediately. (Work did continue on the other plants mentioned earlier, but not even all of these were finished; work on Hope Creek-2 lagged, and that plant was finally cancelled in 1981, leaving Hope Creek-1 a single unit.)

As we can see, a large amount of challenging engineering and construction would have been required to complete the Atlantic Generating Station. One wonders if such a project could survive today’s regulatory environment—to say nothing of clearing approval by a utility’s ownership when the extra cost of constructing the artificial reef type breakwater and shore-based support infrastructure is considered. The best guess for both right off the bat is “probably not,” meaning that the Atlantic Generating Station was probably the closest we’ll ever get to a full-scale commercial nuclear plant situated well off shore.

Sources of information and illustrations: Various original PSE&G brochures—”The Atlantic Generating Station” (undated), “Hope Creek Generating Station” (8/76), “The Salem Generating Station” (4/77), “PSE&G: Nuclear Energy” (6/85).  Also, “The Nuclear Industry 1971″–WASH 1174-71, U.S. Atomic Energy Commission. “Historical Profile of U.S. Nuclear Power Development,” Atomic Industrial Forum 1985. “Nuclear Plant Cancellations:  Causes, Costs and Consequences,” US EIA/DOE 1983.  All materials in Will Davis’ library.

For more on this topic, particularly the plant construction end of the project, see Rod Adams’ article from 1996 on Atomic Insights.


Will Davis is a consultant to, and writer for, the American Nuclear Society. In addition to this, Davis is on the Board of Directors of PopAtomic Studios, is a contributing author for Fuel Cycle Week, and also writes his own blog Atomic Power Review. Davis is a former US Navy Reactor Operator, qualified on S8G and S5W plants.

About Will Davis

Will Davis is the Communications Director for the N/S Savannah Association, Inc. where he also serves as historian, newsletter editor and member of the board of directors. Davis has recently been engaged by the Global America Business Institute as a consultant. He is also a consultant to, and writer for, the American Nuclear Society; an active ANS member, he is serving on the ANS Communications Committee 2013–2016. In addition, he is a contributing author for Fuel Cycle Week, and writes his own popular blog Atomic Power Review. Davis is a former US Navy reactor operator, qualified on S8G and S5W plants.

22 thoughts on “The Atlantic Generating Station

  1. Engineer-Poet

    Instead they [Madison] choose to offer one of the best solar feed in tariffs in the country.

    You state this like it’s some sort of success.  What’s the expected yield from this FIT, and at what price to the economy?  Similar efforts in Spain destroyed several jobs for each one created by the FIT.  That makes it a boondoggle even worse than fuel ethanol.

    that NEVER delivered on it’s promise of too cheap to meter.

    There’s an oft-repeated lie.  Nobody ever promised power too cheap to meter (though at less than 1¢/kWh, the cost of uranium hardly makes it worth it); it was something a utility exec said AFTER he was asked to speculate about the future.  Unmetered power might make sense anyway; you’d buy your juice with something like a co-op membership and get 1 kW per share 24/7.  If you didn’t want to do that, you’d buy from other sources or trade with co-op members on the spot market.  Some co-op member getting juice at 5¢/kWh, charging his EV and running his Ice Bear A/C at night and selling to greenies on hot, cloudy days would laugh all the way to the bank.

    a nuclear accident can turn into a country killer, centuries of harm.

    The babushkas of Chernobyl show the lie of “centuries of harm”.  We’ll never see anything fractionally that bad again, because nobody is ever going to build another RMBK with or without a containment.

    Playing with the most dangerous thing on earth

    That would be something like diptheria toxin, which is both natural and biodegradable.

    You’re fervent and zealous in your ignorance.  This is a bad combination.

  2. Will Davis Post author

    The accident rate for nuclear generating stations is far less than one per decade. That’s pretty clear. Commercial nuclear reactor accidents — and I mean accidents, not unusual events or any other level of notice or alert — have totaled three, ever: Three Mile Island, Chernobyl, and Fukushima Daiichi. Averaged out per decade since the first Soviet nuclear plant went on line in 1954 and the first US nuclear plant went on line in 1957, we see that it’s not one per decade. And direct deaths from TMI and Fukushima Daiichi as a result of the accidents equals zero.

    That’s a pretty low risk, even for people on site. You’re at higher risk every time you step into your car. Or bath tub.

    There are no “successfully covered up” nuclear accidents- at least in the West. Assertions to the contrary are baseless. It seems that conspiracy theory is central to today’s anti-nuclear movement; it has supplanted or replaced environmental speculation (now that the actual minimal environmental impact of nuclear plants is known) as one of the top drivers of anti-nuclear vitriol. I wonder if replacing environmental speculation based fearmongering (as was done in the late 60’s) with false assertion based fearmongering (as is done now) really lends weight and credit to the anti-nuclear movement, or instead pushes the movement to the fringes.

    Remember – solar isn’t what caused the Kewaunee shutdown. Shutdown of fossil plants plus merit based dispatching makes Kewaunee less economical to operate daily. Failure of the owner to be able to incorporate the plant into a larger network of plants to help reduce costs sealed its fate. Solar has nothing to do with it.

  3. stock

    Uh huh, 99 serious accidents of nuclear (plus the ones that were successfully covered up, which they try to cover all of them up until forced to fess up).

    No, sorry little minded one thrashing insults. I have a master from U of Mich in material science and mech eng, I understand nuke pretty well. If you could take humans out of the equation, and mother nature, nuke could be run at reasonable risk, like the 1 accident every 1000 years “they predict”, Instead it is more than one VERY serious accident a decade.

    Nuke fails again and again. We have much better options, and nuke does not even begin to compete on economic grounds, sheesh, even the clunkers like Kewaunee are being shut down by the strong solar program in Madison of the frozen tundra.

  4. James Greenidge

    Stock, I’ll be non-PC and state my views don’t represent ANS or anyone else, but unless you’re just a troll, you’re not only a health/safety hypocrite in the face of nuclear power’s worldwide 60 years safety/mortality/property damage record vs other power sources, but a zero risk hypochondriac at that. You better shrug off your skewed green friends and smell the roses of fact. baby. When oil and gas and coal live up to nuclear’s internationally certified safety and mortality record then you’ll have a reason to squawk about how horribly awful Andromeda Strain bad nuclear energy is even though there’re things under your kitchen sink that’ll put you away far faster than radiation could, but till then, it’s ignorant — yes, ignorant, points of view and votes and FUD spewing as yours that will condemn hundreds of millions not to “if” “maybe” “might happen” nightmares of radiation harm that’s had 60 years worth of chances of being proved, but to historic and provable maladies and wholesale lethal accidents and widespread pollution by fossil fuel use continuing to this current day. Millions of respiratory and pollution patients in hospitals today due fact — not fiction or fear-spun fantasies. To continue that course’s _real world_ hundred year proven curse of fossil fuel use is what is really technically, rationally and logically “insane”. Unless you feel doing that just to assuage your vaporware nuclear fears is worth it.

    James Greenidge
    Queens NY

  5. Pingback: Weekly Digest for January 7th » NA-YGN Southeast Region

  6. stock

    Sure but unlike that boat accident, a nuclear accident can turn into a country killer, centuries of harm. “Your” boat accident, which was you decision to participate in, hardly affects me at all, however, your nuke accident can increase disease of many sorts, cancers, and damage my DNA. And that nuke accident occurs without my consent or willing participation.

    Playing with the most dangerous thing on earth in order to boil water, when we can make electricity directly from my only approved nuclear reactor (the sun) is nothing short of insane and arrogant.

  7. Will Davis Post author

    Humans are involved in things that injure and kill thousands of people every year – like automobile accidents, boating accidents, airplane accidents, mining accidents, electrocutions and falls in tubs. And much more. Risk, and perception of comparative risk, is a reality lost on most people today who will not accept the “risk” of a nuclear plant near by which will never kill or injure anyone, but who will gladly jump in a car one to five times a day, go out on a lake or the ocean in a boat, or fly in a plane.

  8. stock

    The problem with nuke is not the science, and the technology. The problem is that humans are involved. And that is a problem that cannot be solved.

  9. James Greenidge

    Stock, isn’t it bizzaro world that a power source that has killed less people and caused far less physical and environmental damage in its global existence even including its worst rare days in nearly 60 years than just one year of _one_ of its its fossil competitors? Only bad education and philosophical axes to grind spiced with Hollywood scarefests and ego-driven FUD hucksters can turn one from making a positive no-brainer choice here. BTW, one of the nuclear blogs examined the “meter” mantra with surprising results — and why anyway would one snub a clean safe quiet energy source based on meter pie? If anything, like people often cough up a lot more for better quality anything like bottled water over city water, wouldn’t paying a little bit more for a clean environment be worth it?

    James Greenidge
    Queens NY

  10. stock

    Kewaunee was shut down because Madison refused to renew a contract to buy power from Kewaunee. Instead they choose to offer one of the best solar feed in tariffs in the country.

    The economies of scale argument is a joke. Its an excuse to try to cover up the basic fact of a failing industry, that NEVER delivered on it’s promise of too cheap to meter.

  11. Will Davis Post author

    Kewaunee is in a peculiar case wherein shutdown of fossil plants (due to EPA regs) and merit-based dispatching, coupled with low natural gas price and lack of association with other plants (‘economy of scale’) has placed Kewaunee at the low end of profitability. This exact situation may be related to scattered other isolated plants in the U.S. (and indeed other parts of the world) but in no way represents the conditions found over the entire nuclear fleet.

    We have four large commercial plants under construction right now, at Vogtle and Summer. There seems to be good chance we’ll get two or four more in Florida underway within the decade. TVA is completing one unfinished large commercial plant, and when that’s fueled, it will begin completion on two more unfinished plants. The DOE has awarded funds for one group’s SMR development and is expected to make another award in the first quarter of this year. While there will be older plants that will close, the fact is that people do retire and create job openings. For these reasons, there’s no reason to tell students that nuclear energy is a dead duck. That’s simply a fallacy, or a wish.

    Keep in mind when you talk about money that NOT receiving any Federal loan guarantees isn’t a stopping point for either Vogtle or Summer. They’ll proceed whether or not they ever do get any loan guarantees. In the regulated markets where these plants are located, nuclear makes sense in the long term, which is why these plants are being built.

  12. Will Davis Post author

    It’s hard to defend one hundered percent of the several (former) Soviet nuclear programs, but it’s important to note the exceedingly good safety record of their nuclear powered icebreakers over the last 30 years – and the power plant to be used in the floating concept borrows from those, too. I believe their new floating concept is a whole lot more like the Sturgis operationally than the AGS in that it won’t be three miles out to sea but dock side or near shore.

    I do believe that the Russians have done a great deal to make their commercial nuclear plant operation safer since Chernobyl.

    Having said that, the “environmentalist whackos” as you call them were one major reason this plant design was selected by PSE&G – it moved the plant’s heat plume well off shore, which was a major point of contention environmentally at the time. This avoided the construction of (also proposed) ‘nuclear parks’ where up to a dozen plants would have been built on one large site, with a resultant high heat load locally on whatever river or ocean frontage was employed, or else large concentrated heat / drift from cooling towers. (The original papers used for this report don’t go into great detail on what the impact at the plant site was, but consideration at the time seems to have been a ‘lesser of two evils’ affair.)

    Do I personally think that the offshore plants of this design were a great idea? I’m not thoroughly convinced; I’d have to do a lot more reading on the concept, particularly the design and testing of the barge structures themselves and the resistance of the breakwater. Silting under the plants seems, off the top of my head, to be a major consideration I’d look for a response to just from my Navy experience. I do think enough of the idea to be willing to investigate them more (and some people have already, in response to this article, sent me some long out-of-print documents) instead of just dismissing them out of hand. What certainly seems clear to me right now is that our energy demand in the present time, expected demand in the future and environmental considerations as we see them now will not drive anyone to have to consider this sort of design for the foreseeable future.

    And let me close by adding this – historical articles printed here are not necessarily endorsements of concepts. They’re representations of what was, or in this case what almost was. And, as far as the AGS is concerned, a lot of money and time were spent on the concept (make sure to follow the link to Rod Adams’ site for much more) — it was not just a sketch on the back of a napkin. As such, I present the information I find “as is” without trying to judge it in most of the historical articles I write, in the interest of objectivity. The ‘comments’ section is a whole different affair, at least to my way of thinking.

  13. stock

    Nuke is going to be very challenged going forward. The public is on guard against nuclear. The aging fleet of plants is becoming to non economical to run, such as Kewaunee in Wisconsin closing down due to lack of profitability. The New Vogtle plants will cost $14B they say, but overruns of up to 250% are still in the 1 sigma band of reality. Vogtle has a 1.666% annual rate of return on investment, which just shows that bring nuke up to modern safety standards simply makes it to expensive to amortize.

    These factors make me advise aspiring engineers to do something else, like solar, wind, energy storage, battery systems which all have strong futures.

  14. Dennis Mosebey

    Never would have been built even then, Environmentalist whackos would have gone nuts. As for the Russian floating nukes–I saw an article once where a Russian official was quoted as saying “We will rely on our experience in our naval reactor program.” Look up Russian Sub Disasters on the Internet—-scares the crap out of me even thinking of the Russians building one offshore. Heck they cannot even run their land based ones safely–aka Chernobyl.

    Bad Idea now and bad idea then. Forget it.

  15. dan deighan

    my son 30 years old works for navy as mech. eng hvac. he is getting masters. neighbor tells me of challenging rewarding career as” licensed test eng.” he works on nuclear plants all over country. he started with navy who put him in 6 mo. school. he also has mech. eng. any suggestions for son, career advancement, more challenging job?personable, very smart.

  16. stock

    What could possibly be more insane than putting a civilian power reactor on a boat. Wow.

  17. Will Davis Post author

    Exellent links, Pete- thank you! I have a couple of these large drawings myself, but not for AGS. They’re spectacular. I did not know that UNM had that collection – that’s a very valuable link!

  18. Will Davis Post author

    As you may have seen, James, the reason that PSE&G decided to build offshore was primarily due to waste heat – that is, the need to either use a river or ocean water, or build cooling towers. Given PSE&G’s rather highly concentrated service area, it seems to me fairly obvious that eventually it would run out of locations for new nuclear plants on shore (considering its plan for vast nuclear expansion at the time.) The site out to sea solves the problem of impact on rivers or coastal shore line due to heat. Certainly, there would have been different considerations about radioactive release in an accident scenario with the plant out to sea, but that was not the driving factor in the site selection.

  19. Chris Pragman

    While the Atlantic proposal was impressively audacious in it’s own right, it’s only a slight advancement from the artificial island that Salem and Hope Creek sit on today. That site is also audacious! The cofferdams created to stabilize the site are enormous. The site originally planned for something like 6 or 8 units.

  20. James Greenidge

    Happy New Year All!

    I have highly ambivalent feelings about this sea-based nuclear plant concept. While I support the basic idea, it inadvertently piques the public perception that nuclear power is so perniciously hazardous and fatally fragile that it must isolated from inhabited locations as much and far as possible. How would this notion spin on current and future INLAND plants? It might be far better for the positive acceptance and perception of nuclear power to build them underground to deal with accidental release contamination and local aesthetic issues, either digging from scratch or even exploiting existing environs like salt caverns or abandoned underground facilitates. Yes, maybe digging under might be a little more expensive (?), but regretfully, going the off-shore site route isn’t going to give nuclear energy a better friendlier good-neighbor image, IMHO. Think leper colony.

    James Greenidge
    Queens NY

Leave a Reply

Your email address will not be published.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>