Underwater gliders propel themselves through the water in an interesting way. They empty their tanks to float up, fill them to sink, repeat, going up and down rhythmically. But instead of just going up and down on the spot like idiots, they use wings to create forward movement from this up-down motion. This PDF says they "use lift on wings to convert this vertical velocity into forward motion". The concept is illustrated well in this video.

enter image description here

This is much more energy-efficient than using propeller or jet-pump propulsion. "seagliders operate with a high efficiency because the buoyancy engine only operates twice per cycle, once at the top and bottom of each yo, resulting in a low average propulsive power" and the PDF already linked says "high efficiency is more easily achieved in the simpler hydrodynamic environment of wings". This is desirable for a solarpunk world I am building, where energy-efficiency is important.

That PDF again: "A high operating speed would, of course, be desirable, but this conflicts with the primary goals of low cost, small size and long duration." – supposing I didn't care about "low-cost" (it's a non-monetary world), and "small size" (I'm talking about a vessel big enough for a crew of dozens to live on), in that case would it be possible to have a large, reasonably fast underwater glider? (When I say 'reasonably fast'; existing models typically get about 1km/h, and that's ridiculous for a manned mission. 30km/h would be great, but 10 or even 7 would be fine. Energy-efficiency is more important to solarpunks than speed.)

My question is could this method of locomotion work on such a big, manned submarine? Would it fail to create enough thrust? Would it make the ride unacceptably bumpy?

The paper with DOI 10.1177/1729881420920534 says, in essence, that big rise-and-falls, say 1000m, will provide the most efficiency per unit distance travelled, and you'd go 10km forward for every yo. This would barely be noticed by people on board. That means about a 6° angle (tan6° is about 1/10)

Blended wings (where the whole body is a wing) are discussed here and in the paper with DOI 10.1080/17445302.2019.1611989 , which seems to assume 0.8m/s (i.e. 2.88km/h)

Note1: Stealth is unimportant; people always bring up stealth when submarines are being discussed.

Note2: It will have an auxiliary active engine (hydrogen-fuelled) for emergencies, sprinting, surface propulsion, etc.

Thank you.

  • 1
    $\begingroup$ @AlexP Not updrafts. The device in the linked video generates up-down motion by pumping ballast. It converts the up-down to forward motion using fins. $\endgroup$
    – Boba Fit
    Commented Feb 2, 2023 at 20:49
  • 1
    $\begingroup$ "Stealth is unimportant" begs the question, why a sub? Running underwater slows you down. Subs go faster on the surface. So why are you underwater? You need a reason to be down here. That reason limits what is feasible to do. $\endgroup$ Commented Feb 2, 2023 at 21:47
  • 1
    $\begingroup$ How does it beg the question? I didn't assume the conclusion in the premise. $\endgroup$
    – wokopa
    Commented Feb 2, 2023 at 21:51
  • 2
    $\begingroup$ They could do fun Cousteau stuff, chat with their whale buddies. Why did The Beatles live in a submarine? $\endgroup$
    – wokopa
    Commented Feb 2, 2023 at 22:00
  • 3
    $\begingroup$ @wokopa to sell albums. $\endgroup$ Commented Feb 2, 2023 at 22:01

3 Answers 3


Ok, after a few more hours research, I think I'm ready to answer my own question.

Yes, and bigger is actually better

The best paper on this is Jenkins, S. A., & D’Spain, G. (2016). Autonomous Underwater Gliders. Springer Handbook of Ocean Engineering, 301–322. doi:10.1007/978-3-319-16649-0_12

It says in the abstract/summary, as one of the main points:

  • "Increases in efficiency can be obtained by: 1. Increasing the loaded mass (with larger buoyancy engines) and increasing the overall size of the glider, which increases the glider’s speed and maintain sufficiently high Reynold’s num- bers to avoid the drag crisis."

That's music to my ears. It's saying the bigger they are, the faster and more efficient they are. Somebody mentioned the Liberdade class gliders, which outperform the competition and are also bigger.

In the section on 'Size effects', the same paper goes on to say:

  • "The minimum specific energy consumption from(12.13) decreases as V =4 b , i. e., it decreases with increasing net buoyancy. Because net buoyancy is somefraction nb of the total vehicle volume V0 , where Vb D nbV0 , bigger buoyancy-driven vehicles generally are more transport efficient. Actually, surveys of natural and man-made fliers by [12.12] and [12.14] demonstrate that specific energy consumption monotonically decreases across 12 orders of magnitude of size increase. This size advantage is accentuated in an underwater glider because the buoyancy volume factor, nb , increases with increasing vehicle volume approximately as nb 1:2 105V7=6 0 , due to economies of scale in packing efficiency [12.7]. Larger nb permits higher glide speeds (speed increases as the square root of the increase in net buoyancy) and higher wing section Reynolds numbers, which results in higher wing section lift-to-drag ratios"

More music to my ears! Big ones are not only more efficient, they are faster too! And this effect endures over 12 orders of magnitude, so yes we could scale up the man-portable drones that now exist to something that houses dozens of crew.

Some details:

The Liberdade vehicles are 1400lb, or, in freedom units, 0.6356 tonnes. A 4000-tonne vehicle is 6293 times the size. (It may seems I'm confusing volume with mass here, but that's valid because all submarines have the same density, i.e. the density of seawater)

A blended wing design (the wing is the body) is best. At first I was resistant to this because it's too exotic, but the more I think about it, the better it seems. It's not that much more high-tech to build a wing than a tube; it doesn't require the complex fly-by-wire stuff of airborne flying wings. Cigar-shaped submarines are pretty boring & ugly let's face it. It is closer to nature, in particular manta rays, but even whales have tapered foil-shaped bodies. The living quarters are not narrow and that's more comfortable.

enter image description here

The "speed increases as the square root of the increase in net buoyancy", which is to say if you double the size of the ballast tanks, you increase speed 41%. If you quadruple the size of the ballast tanks, you double speed. Obviously that cuts into space you could have otherwise used for living space, hydrogen fuel, equipment. We can go faster if we lose our ping-pong table.

This diagram shows the relationship of glider speed to ballast tank size (the four arcs) and angle.

enter image description here

35° is the fastest angle, which will be uncomfortable. 4° is the most efficient. The tradeoffs between a comfortable flat angle, speed, and efficiency are something the crew can decide as they steer. They can sleep in hanging beds while the boat dives at steeper angles:

enter image description here

It's not a constant up-and-down bouncy turbulence. It's smooth-but-slanted and stays the same for a fairly long time.

A second reference supporting the scale effects is: https://escholarship.org/uc/item/1c28t6bb – The Table E.1 shows that for each class, the bigger versions are better. Chapter 10 is entirely devoted to discussing the effects of size on gliders, what the question was about.

I have a detailed design for a 25-man, 4000-tonne blended with submarine worked out now. I won't bore you with it unless one person asks me.

  • $\begingroup$ The last calculation is wrong. You need to sub in 1 in place of the 6293 and compare the ratio of results. You're comparing 0.569 to 0.0034, so it's 160 times faster, which as a matter of maths makes sense given that it's almost a simple square root relationship with 6293 as an input ratio, but as a matter of real life performance seems unlikely. I'm assuming that the other expressions are correct uncritically. $\endgroup$
    – user86462
    Commented Feb 4, 2023 at 10:09
  • $\begingroup$ I'm just gonna delete the last calculation. If someone wants to read the section on 'Size effects' in the paper with DOI 10.1007/978-3-319-16649-0_12 and calculate how much speed you'd gain from a 6293× increase, that would be appreciated/upvoted. $\endgroup$
    – wokopa
    Commented Feb 4, 2023 at 12:55
  • $\begingroup$ I'm always happy when people figure out solutions to their own questions and share them here. Wish I could upvote twice for the freedom units. $\endgroup$ Commented Feb 4, 2023 at 17:56

Yes, that would work. This could easily be done with oversized ballast tanks. I have no doubt that current submarines take advantage of this from time to time.

More likely, they would have numerous piston chambers that sucked water in at the surface and pushed it out as jets at the bottom of the ship's sawtooth path.

Issues I can think of:

  • the entire floor would be pitching forward and back with every shift. Maybe you'd put the bunks and control center on gimbles.
  • you'd incur an immense amount of wear and tear as the hull was constantly being pressurized and depressurized.
  • You'd need a significant depth to operate.
  • You'd still need normal propulsion to get into and out of port.

The speed would be based on the size of the ballast tanks. Larger tanks = more buoyancy = faster gliding.

  • $\begingroup$ These are all good points, I think I can call them all annoyances rather than dealbreakers. 1) The pitch would be about 6° and change less than once per hour. 2) I had already decided that these vehicles would be made of composites, not metal, so that it could be made flexible so that the sea pressure would be within the fatigue limit. This mentions a sandwich structure for sub hulls which could be flex outside, strong inside. $\endgroup$
    – wokopa
    Commented Feb 2, 2023 at 22:17
  • 1
    $\begingroup$ @wokopa, The 6 degree pitching would be less than most sea-going boats experience due to surface turbulence, so I wouldn't worry about it. I think you got this. $\endgroup$ Commented Feb 2, 2023 at 22:19
  • 1
    $\begingroup$ Stuff falling over every now and then, I'm happy with that as part of the worldbuilding. The sort of everyday annoyance that adds flavour. The 30° dives could be timed when most of the crew are sleeping. $\endgroup$
    – wokopa
    Commented Feb 2, 2023 at 23:13
  • 2
    $\begingroup$ @AlexP the surface of a plane's fuselage is not excluded while calculating aerodynamics, here the issue is 770x worse, so no, tilting the sub's hull is mandatory. $\endgroup$
    – Vesper
    Commented Feb 3, 2023 at 5:56
  • 1
    $\begingroup$ @AncientGiantPottedPlant, What you're describing sounds more like "cargo ship" kind of "speed isn't important." My impression was that OP meant something more like the "not actually in a hurry" kind. $\endgroup$ Commented Feb 5, 2023 at 3:37

Thinking on this - I think the problem you are going to hit is one of displacement and momentum firstly and a pretty major safety issue secondly.

As you get larger, you displace more water, you can only increase your buoyancy so much for a given volume and displacement - which provides an upper limit to how much up-thrust you can generate. Will that be enough to overcome the momentum of the ship? I don't have access to complicated Fluid Dynamics software - but my gut tells me that at a certain point, it's easier for the water to flow around the wing, than it is to generate a forward motion.

The second issue is that without an engine - if you have any problem with your Ballast system - you are probably dead.

I'm thinking of the case of the USS San Francisco - SSN-711 when she booped an underwater mountain running at Flank speed. IIRC, because of the damage to her front ballast tanks, she had to keep her nose up using the rear diveplanes and the Propulsion from her engine.

If you have no engine, and you have an issue with buoyancy - your only option is to throw the chicken switches and hope that gets you to the surface, you have no other means of propulsion of getting to the surface. And once you've done an Emergency Blow - that's it.

So, for small Unmanned subs - I think it's possible, but for larger Manned subs - I think you are going to have issues with overcoming the momentum on the boat and simply put - such a design would be massively unsafe.

  • 3
    $\begingroup$ Ooo, chicken switches. It makes sense that they would exist, but I hadn't heard of them. $\endgroup$ Commented Feb 2, 2023 at 21:49
  • $\begingroup$ "Major safety issue" is what makes it fun! $\endgroup$ Commented Feb 4, 2023 at 17:58

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .