[idea] Concrete Modular Submersible Structures

[BATAAN]

When I was 13 or so I drew up a great one using an aircraft drop tank and an electric motor.... At that time I didn't understand pressure more than vaguely but I loved the idea of driving around underwater in my own Tom Swift Electric Submersible, still do.

[JRMacGregor]

Theodose

You have received some good advice already above.

There is no doubt that concrete can be used as a pressure resistant material underwater. Many, many concrete oil platforms are located in the North Sea in water depths of around 150m. Some of these are pressure resistant, and have been in service since the 1970s.

1) Famous design, many examples = Condeep, by Norwegian Contractors (Beryl, Brent, Statfjord, Forties, Draugen etc)

2) France also has its own design from Doris - as applied at Ekofisk, Ninian Central, Total MCP01 etc.

Of course there have also been accidents, despite all this experience and thousands of manhours of engineering - like the sinking of the Sliepner platform during construction when a tank cell in the bottom of the structure imploded when the platform was deep submerged in a fjord for pressure testing. The subsequent impact with the seabed was registered on the Richter scale. This was because of an error in design/analysis of the structure in that location.

The cyclic stress issue mentioned above is also important. Maybe you can keep ALL of your submarine structure permanently in compression ?

At the Marseille dry dock during mid 1990s I observed the construction of a huge concrete barge to be used to support 20,000 tonnes of offshore oil equipment at the N'Kossa field offshore West Africa. This hull was not subject to much external pressure but was subject to cyclic stress due to the ocean swell (long low waves offshore Africa).

As well as the normal rebar, this concrete structure was "kept" in compression by tensioned steel wires running internally in steel tubes in the structure. The weight of these wires was a significant part of the total barge weight.

Several technical papers have been published on the design of the N'Kossa barge, including details on the concrete strength/specification. Since the main contractor was Bougyues (French), I expect you will be able to find many more of these papers.

The Norwegians have also built two huge floating oil platforms in concrete - Troll semi-sub and a tension leg platform. But these had more steel weight in them than a pure steel equivalent, and were therefore heavy/huge - leading to huge and expensive mooring systems. So not justified on a cost basis - perhaps on an employment basis.

As for concrete in submarines........my memory tells me there is quite an extensive and serious paper published in the Transactions of SNAME on the design of large concrete submarines - from 1970s.

[Théodose]

I didn't expect to receive so many contributions in a such short amount of time... But that makes me more eager to continue !
I'll try to answer your different questions as exhaustively as possible.

- mydauphin : Thanks you for the detailed advice, I'll keep them in mind if I can make small-scale tests in the future. On a side note, the 5000 psi you mentioned correspond to 35 MPa, the maximum admissible stress for the concrete I used as a base case. But a more resistant mix can be used, especially with non-metallic fiber reinforcement since it looks like a promising technology to get more homogeneous mechanical properties and reduced corrosion issues.

- rwatson, bataan : An intermediate bulkhead will always be placed between two adjacent tubes, for safety and structural reasons. The inner "collar" (not sure of the right term in English) at each end of a tube and bulkhead has at least 24 reinforced holes for strong bolts, in order to keep the assembly sturdy. So 4 to 6 long rods could be passed along the entire length of the structure through these holes and tensioned at the end-caps. Plus these rods could also serve as fixation points for the internal frames panels holding the equipment racks : double-purpose !

- JRMacGregor : Thanks you very much for all these references, I'll try to find and share the corresponding papers. It won't easy though, since the university I'm working doesn't have a Mechanical Construction department. Assessing the cycling stresses is of course my main concern in this effort to extend from static to dynamic applications, not forcibly submarines.

The most simple scenario I can think of would be repeated dives from the surface to the service depth, with a pressure variation of 5 bars. So the azimuthal stress in the tube would be varying between 0 and S = dP*R/e = 0.5*1.125/0.1 = 5.625 MPa.
Now I need to find a paper similar to this one about fiber-reinforced HPC but about a 35/45 MPa concrete and then check on the Wohler curve the admissible number of cycles. That constitutes the upper limit of fatigue resistance for this structure and I'd be happy if this figure exceed 4000, as it represents about 10 years of service with a daily full dive. But that's still in the oligocyclic regime (<50000 cycles) and I think that correctly-prepared concrete can withstand more than that.

[Submarine Tom]

I have no doubt that "properly prepared concrete can withstand" this and more.

I encourage your theoretical research but at some point, you're going to have to build one and hang it on an automatic winch that raises and lowers it to service depth. Either until failure or for a number of service cycles and then simply lower it to failure/implosion. I don't mean to be overly keen on destructive testing but once the theory is done, I think it the prudent way to go.

-Tom

[Submarine Tom]

That is a costly proposition. Perhaps the way to go here is pressurized, mixed gasses.

Remove the cycling all together. One has to play with gasses anyway at depth: CO2 monitoring (scrubbers), O2 monitoring, etc.

The entire gas system could be automatic with fail-safes that simply return you to the surface when/if there is a malfunction.

Sure beats the hell out of imploding! That would spoil everyone's day.

-Tom

[Théodose]

I know that this destructive test has been recommended in the other threads concerning the concrete submarine. I just can't realise it in a foreseeable future because I need to finish my PhD first and I don't have any kind of equipment or workshop/test area at my disposal, since I live in the centre of a large non-maritime city.

On an other subject, I found traces of the technical papers proposed by JRMacGregor (ironically, they were mentioned in one of Elmer's many mirror sites...) and I filled the needed forms at the university science library to retrieve them. I'll share them on this tread once they'll arrive.
- Valenchon, Nagel, Viallon, Belbeoch, Rouillon: The NKOSSA concrete oil production barge.
OMAE 1995 - Copenhagen - 14th International conference - June 18-22 1995.
- SMITH, D A : FEASIBILITY STUDY FOR CONCRETE SUBMARINE. Society of Naval Architects and Marine Engineers, March 1975

Another interesting paper found during this bibliography search is attached with this message : it quantifies the durability of fiber-reinforced concrete under cyclic compressive loading, exactly the kind of data I needed !
Well almost exactly : it concerns a slightly stronger concrete (50 Mpa of maximum strength instead of 35 MPa for my base case) reinforced with 30-60mm long steel fibres instead of re-bar and with no pre-stressing, but that's close enough to get some useful insight.
The Wohler diagrams on page 3 show that this kind of concrete can withstand at least 10,000 (logN=4) compressions cycles rising to 60% of the maximum strength, largely enough to meet the durability criterion I mentioned in my previous message...

So it should be possible for a concrete submarine to withstand repeated dives at a service depth of 50m. But one must be careful when extrapolating the regression equations present in the table page 4 : the level of stress S I calculated is very low (11-16% of maximum strength) while the compression cycles in the article go from 10% to S=60-90%...

it concerns a slightly stronger concrete (50 Mpa of maximum strength instead of 35 MPa for my base case

Slightly stronger is a bit understated.

Up to C 25/30 every dumb nickel can produce concrete.
From C 30/37 on, you need special equipment and a laboratory to get a proper product.
So, when already C 45/55 is required, you can as easy go straight for C 80/95 (formerly B95) to start with the right compression strength. (of the concrete, not the structure)

Regards
Richard