Toroid Tanks

guest blogger john hare

The VTVL vs HTHL vs VTHL arguments will probably never end. Many people simply cannot admit that different methods fit different requirements. That I prefer HTHL myself is not a compelling argument for the method. It does mean that I will look for ways to support my preference.

The number one advantage of VTVL vehicles is light airframes. I may have found a method of getting a runway vehicle close to the attainable mass ratio of vertical vehicles. Or an air launched vehicle with a better ratio than wings on a rocket. It all starts with the tanks in spacecraft. Launch vehicles have been described as propellant tanks with auxiliary equipment. The two types of successful tanks are spherical and cylindrical. The multi lobe tank was one of the causes of the X33 failure.

It is my understanding that a torus tank would have very slightly worse mass per volume compared to the normal spheres and cylinders. It offers the possibility of advantages that more than make up the difference in terms of over all vehicle mass, construction cost, reentry shape, and payload volume. The inflatable raft offers a dramatic view of the layout and strength of a toroid based structure. The billions of auto tires on the road demonstrate the same. Picture either one based on an inflated sphere or cylinder. Other than a hand full of special cases, it won’t work.

Though zee Griffenshaft is an extreme example of long and thin problems, most ELVs use considerable mass for intertank structure and payload fairings, not to mention the attention that must be paid to keeping the whole vehicle rigid enough for the job. While these things are understood and handled, they cost. Any VTVL RLVs will also have mass in those places.  Torus tanks offer the possibility of eliminating the inter-tank structure and reducing the payload fairing structure. This is because the structure of the two tanks can be concentric and stiff in three dimensions. The donut hole is the payload bay with relatively roomy volume. 

The sketch here is HTHL mode. Yank the aero surfaces and use it for VTVL for the same result. Less parasite mass, if my numbers are correct of course. In the very small sizes the tanks can be irrigation pipe formed into a circle to get a more rigid structure from small diameter COTS parts. In the larger sizes it isn’t necessary to have 10 meter diameter cylindrical tank tooling. 5 meters coiled to a torus can do the same job. Stack similar donuts for an even lighter VTVL structure with no serious inter-tank structure or payload shroud structure.

In 2000-2001, a consultant ran a disk plane simulation on Xplane and got a lift/drag ratio of 8. What I am suggesting here would have a different layout than he used and would have  more drag. The lifting body program of the 1960s had L/D ranging from 2.5 to 4. This lifting arrangement would most likely split the difference to get 5 or 6. This would be terrible for an aircraft, acceptable for a spacecraft though. For a spacecraft, we should be willing to sacrifice efficient aerodynamics to achieve low dry mass as long as the vehicle is capable of doing the job of getting on the ground safely. The German and American disk planes of the 1940s had poor efficiency due to the very low aspect ratio, and this one will be even worse. The only thing that could possibly make this layout acceptable is the possibility of a very strong airframe that is very light. The lightness is because the tanks provide virtually all the vehicle strength with most of the other structure being skin with stiffeners.

I don’t really like the wings on a rocket school of spacecraft. The objections of the VTVL people are somewhat accurate for most of them. Many have all the structural problems of the totem poles, added to all the structural problems of an airplane. The lifting bodies to me are nearly insane. Terrible flight characteristics and terrible mass problems for no visible gain. The pure VTVL lacks cross range options for the most part, even if the landings are getting well defined.

On reentry the vehicle could come in belly first for a very fluffy profile. At alpha 90, it is hard to picture anything short of a balloot that would have higher drag. The fluffy reentry profile allows simpler lower tech TPS that many people suggest for vehicles that are mostly tank. The big problem I see is keeping it stable at high alpha during reentry. Fortunately there are people that know how to solve that problem.

This idea seems obvious enough that there have probably been studies about it. That I have not seen them doesn’t mean they don’t exist. If there is a good reason this won’t work, it was likely written up in the 1940s or 1950s.

The following two tabs change content below.
johnhare

johnhare

I do construction for a living and aerospace as an occasional hobby. I am an inventor and a bit of an entrepreneur. I've been self employed since the 1980s and working in concrete since the 1970s. When I grow up, I want to work with rockets and spacecraft. I did a stupid rocket trick a few decades back and decided not to try another hot fire without adult supervision. Haven't located much of that as we are all big kids when working with our passions.
johnhare

Latest posts by johnhare (see all)

This entry was posted in Uncategorized. Bookmark the permalink.

23 Responses to Toroid Tanks

  1. Tim says:

    Found this at Encyclopedia Astronautica
    http://www.astronautix.com/craftfam/lenicles.htm
    The front page has links to a patent and a paper on a “space bomber” that I remember being related. They’re under Newest/Aditional changes.

  2. john hare says:

    Tim,
    Interesting link especially the Bono big lifter. They seem to feel that the lenticular shape is structurally superior by itself without regard to the tanks. I’m not convinced that the flying saucer shape is all that good, simply that it my be acceptable. All the lifting bodies seemed to need very large verticle fins for stability, sometimes to the point that they might as well have used wings.

  3. Adam Greenwood says:

    In your diagram what do the solid red and green circles represent?

  4. john hare says:

    Adam,

    Doh, can’t believe I forgot to label that. Green LOX and red fuel. The two on the left standard cylinder or sphere tanks and the two on the right the torus based lenticular layout.

  5. Eric Collins says:

    John,
    Interesting idea. I think there may be another consideration which you did not mention, and that is thermodynamics. Irregardless of the structural characteristics of toroidal vs. spherical tanks, the more tank surface area you put in contact with a given volume of fuel is going to require additional cryrogenic systems to keep your LOX (and LH2 if used) boil-off under control.
    From an aerodynamics perspective, I cant see how the ‘fluffy’ profile of this vehicle is a net plus. Even if your airframe does not require additional structural elements, you’ve more than made up for it with additional tank wall mass. Unless you can provide significant overall fuel savings with this layout, the vehicle is going to end up displacing more volume, and thus encounter additional atmospheric drag during ascent.

  6. john hare says:

    Eric,

    I did miss the boil off problem. This concept would clearly take a hit if LH2 was used. Possibly LOX as well. It would need serious investigation before going for even a preliminary design study. Good catch, thanks.

    The vehicle would be fluffy coming back through reentry. Spreading the thermal load over as much area as possible to reduce the heat per unit area. 90 degree difference in flight direction and load paths.
    Going out would be edge on like a wing for minimal drag. Tank wall mass is supposed to be only slightly heavier than an equivilent cylinder. The tank structure is a bent cylinder that eliminates dome ends. Unless you see something I don’t, tank mass is almost identical to a very long cylinder.

    Where do you see the extra wall mass requirements?

  7. tsenf says:

    John,

    Given Eric’s comment, perhaps this approach be particularly suited for O2/Methane given the (relatively) similar cryogenic temps: -150°C / -161.6°C, respectively.

  8. Paul Breed says:

    I think getting the tanks fabricated will be tricky.
    I don’t believe they make mandrel benders in that size.
    (For any value of size apropriate from orbiatl rocket)
    A Fabric bladder tank for non cryogenic combination like H2O2 and RP1…. maybe.

  9. Eric Collins says:

    Ok. I guess you are right. I ran the formulas for volume and area of a torus against that for a cylinder with hemispherical end caps.

    Vt = (pi r^2)(2 pi R)
    Vc = (pi r^2) h + (4/3)(pi r^3)
    At = (2 pi r)(2 pi R)
    Ac = (2 pi r) h + (4 pi r^2)

    Assuming that the cylinder and the torus have the same interior radius, r, then for a given volume of fuel, you get:

    Vt = Vc
    (pi r^2)(2 pi R) = (pi r^2) h + (4/3)(pi r^2) r
    (2 pi R) = h + (4/3) r

    The surface areas are then:

    At ~ Ac
    (2 pi r)(2 pi R) ~ (2 pi r) h + 4 pi r^2
    (2 pi R) ~ h + 2 r

    using the relation obtained from equating the two volumes, we arrive at the comparison

    h + (4/3) r < h + 2 r

    Thus At r is the radial distance to the center of the toroidal tank.

    Another concern could possibly be with stability. The further away you put the tanks from the vehicle’s center of gravity, the more likely any shifting of fuel or asymmetric aerodynamic loading is going to cause non-trivial moments to be imparted on the vehicle.

    I’m not trying to be difficult. I’m just trying to raise some concerns that come to mind as I try to ponder the characteristics of this particular configuration.

  10. Stu says:

    I agree with Eric, sloshing seems like it would be a very large issue with toroidal tanks. They tanks may need to be subdivided into smaller sectors to keep propellants from shifting. At that point, you would be back to a configuration with basically spherical/cylindrical tanks around the edge of your flight vehicle.

  11. Eric Collins says:

    Wow, something really got clobbered on my previous post. Ok, let’s try that again.

    Thus: At is less than Ac

    So, I guess my main concern is mostly with the amount of cross-sectional area that will be presented during ascent. For a cylinder, you have a CSA of about (pi r^2). For the lenticular configuration (i.e. edge-on torus), you would have at least (pi r^2) + (2 r)(2 R), where R (always greater than r) is the radial distance to the center of the toroidal tank.

    (reminder to self: look up the html codes for less than and greater than)

  12. john hare says:

    Tsenf,

    Given Eric’s comment, perhaps this approach be particularly suited for O2/Methane given the (relatively) similar cryogenic temps: -150°C / -161.6°C, respectively.<<<<

    It is possible that you would want similar temps. Either your way or Pauls, needs study.

  13. john hare says:

    Paul,

    I think getting the tanks fabricated will be tricky.
    I don’t believe they make mandrel benders in that size.
    (For any value of size apropriate from orbiatl rocket)
    A Fabric bladder tank for non cryogenic combination like H2O2 and RP1…. maybe.<<<<<<

    The issue could well be if this concept offers enough advantage to build specific tooling for them. Even if is is proven to have a small technical advantage, the financial implications of building dedicated tooling may make it a non starter.

  14. john hare says:

    Eric,
    Ok. I guess you are right. I ran the formulas for volume and area of a torus against that for a cylinder with hemispherical end caps.<<<<<

    I appreciate you running the numbers. My capabilities are variable in that field. Long story.

    Another concern could possibly be with stability. The further away you put the tanks from the vehicle’s center of gravity, the more likely any shifting of fuel or asymmetric aerodynamic loading is going to cause non-trivial moments to be imparted on the vehicle.<<<<

    This is clearly an issue to be addressed. My thoughts run to a series of low pressure bulkheads and trim pumps. Or at least one low pressure bulkhead in the 6 oclock of each torus. Somebody has probably simulated this somewhere. It would be nice to know if there is a problem and how big if there is. Required fixes could easily cancel the advantages I think this offers.

    I’m not trying to be difficult. I’m just trying to raise some concerns that come to mind as I try to ponder the characteristics of this particular configuration.<<<<<

    Tear it up. If you can shoot an idea down at this stage, it needs it. If you can score a clear kill, you could easily save somebody time and money. I would offer prizes for first honest definitive kill if I could afford it.

    I accept that there will be somewhat greater ascent drag than a clean cylinder, with possibly some offset for an aerodynamic lift flight profile. This would be less important for an air launch method with the initial burn at 30,000 feet or so.

  15. PeterH says:

    Have you considered a ‘toroid’ of other than circular shape. Maybe a triangle of 3 cylinders with toroid sections joining them. Strength of the toroid with some aerodynamic section options.

    Given concentric toroids, I’d be inclined to give the inner toroid a larger minor radius to better fit the lenticular shape. Alternately, bulkheads could divide a single toroid into multiple tanks.

    One potential for an upper stage is the engines firing broadside to the lenticular shape when aerodynamic drag isn’t much of a factor.

  16. john hare says:

    PeterH,

    Joining cylinders is well understood without need to rope toroid sections in. If you are going to use cylinders for lift, placing a series of them lengthwise in a wing can get a decent structural shape.

    My base size preference is 18 foot disk with 18 inch outer fuel tank and 24 inch inner LOX. Leaves an 11 foot diameter cockpit/cargo bay in a fairly small vehicle. The single larger toroid with bulkheads gives better structure and payload bay.

    The engines firing broadside is certainly possible for the upper stage as a massive expansion ratio aerospike that could use the same engines for landing vertically. Something like this was mentioned in the saddlespike comments.

    I am sacrificing what I consider optimum in these posts in favor of as much clarity as I am capable of. It is claimed that my ideas are crazy enough that I need to be conservative here.

  17. Roger Strong says:

    Sorry if this is silly, but it might lead to something worthwhile. I have no engineering background.

    Suppose you were to cross your design with a Sea Dragon – the same torus tanks, but MUCH larger and ocean launched and recovered.

    I understand that Skylab was originally to be launched “wet” – the lab itself to be filled with fuel on the way up. Could torus tanks be inhabited the same way once in orbit, with the whole thing spun for (low, perhaps Mars-like) artificial gravity?

    Alternately, would torus tanks (and the larger design) give your design any advantage for use as reusable orbital fuel depot?

    A couple years ago Jon Goff described how fuel depots need a way to keep the propellants settled so that you can pump them. Other than adding lots of complexity, spinning the depot for artificial gravity was mentioned. “The problem this creates is that it is really hard to dock with such a station at any point other than along the centerline, so that might limit the number of vehicles that can visit the depot at any given time (unless you despin the inner section, but that introduces other issues)…”

    A torus shape puts most of the volume farther away from the center of mass than spherical and cylindrical tanks – less spin is needed. The torus makes it easier to de-spin the inner section to allow more vehicles. If I understand correctly, the disc shape could be kept edge-on to the sun AND the direction of travel, to minimize both boil-off and drag.

    Being reusable you wouldn’t need to launch propellant to it – you’d bring it home when empty. The next launch could put it in a different orbit. (OK, it’s more of a fuel truck than a fuel depot.) Granted, the type of fuel you serve is probably limited to what you use for launch.

  18. john hare says:

    Roger,

    “Sorry if this is silly, but it might lead to something worthwhile. I have no engineering background.”

    My education in this field is self inflicted.

    “Suppose you were to cross your design with a Sea Dragon – the same torus tanks, but MUCH larger and ocean launched and recovered.”

    The concept is not about a particular size. It is about a design possibility that might be useful. While I think this might work in Sea Dragon sizes, I am more comfortable with smaller vehicles with high flight rate and lower development cost.

    “I understand that Skylab was originally to be launched “wet” – the lab itself to be filled with fuel on the way up. Could torus tanks be inhabited the same way once in orbit, with the whole thing spun for (low, perhaps Mars-like) artificial gravity?”

    If you did a Sea Dragon class vehicle SSTO with dense fuels, there would be on the order of 15,000 tons of fuel, and 15,000 cubic meters of tankage. If you took Peters’ suggestion of a single torus with bulkheads and did a twenty foot minor radius for working room, outer major diameter would be over 500 feet. My BOTE suggests it might work, though it would be close to the 3 rpm considered the limit for people tolerant of motion sickness. *

    “Alternately, would torus tanks (and the larger design) give your design any advantage for use as reusable orbital fuel depot?”

    Eric brought up the point of surface area per volume enclosed affecting the insulation requirements. While it would probably work, a sphere would be better if it is cryo fuels.

    “A couple years ago Jon Goff described how fuel depots need a way to keep the propellants settled so that you can pump them. Other than adding lots of complexity, spinning the depot for artificial gravity was mentioned. “The problem this creates is that it is really hard to dock with such a station at any point other than along the centerline, so that might limit the number of vehicles that can visit the depot at any given time (unless you despin the inner section, but that introduces other issues)…”

    A torus shape puts most of the volume farther away from the center of mass than spherical and cylindrical tanks – less spin is needed. The torus makes it easier to de-spin the inner section to allow more vehicles. If I understand correctly, the disc shape could be kept edge-on to the sun AND the direction of travel, to minimize both boil-off and drag.”

    I started to answer negatively because the fuel would tend to distribute itself randomly around the whole perimeter in an unpredictable manner and have slosh problems. If the torus was spun slightly off center though, the fuel would tend to accumulate in specific sections of the tank. Interesting use possibility.

    “Being reusable you wouldn’t need to launch propellant to it – you’d bring it home when empty. The next launch could put it in a different orbit. (OK, it’s more of a fuel truck than a fuel depot.) Granted, the type of fuel you serve is probably limited to what you use for launch.”

    In the truly large sizes you are suggesting, I don’t know. My mindset is more toward the small stuff for financial reasons. A ten ton hauler that launches once a week places 500 tons of propellant in orbit per year. And when you have one smaller airframe fully worked out, building more of the same is much easier.

    Transferring fuel in orbit is the next one I’m working on. I see it a bit different than most people do.

    * note to self check numbers on self launching mini spin station torus

  19. Roger Strong says:

    Thanks!

    A follow-up question, in an area even further above my head….

    As Jonathan Goff noted, (“Additional Thoughts on the SA’08 Propellant Depot Panel”) a number of people are suggesting that a LOX-only depot makes a lot of sense.

    LOX is paramagnetic. Instead of spinning the torus, could you use a toroidal magnetic field to keep it away from the tanks walls to reduce heat transfer? (And perhaps to move it around as needed?)

  20. Jonathan Goff Jonathan Goff says:

    Roger,
    Yeah, there are lots of people who’ve studied “magnetic propellant positioning” for LOX. It has a lot of potential.

    ~Jon

  21. john hare says:

    Roger,

    While I still don’t think Sea Dragon class vehicles make economic sense, your thought of a self launching station with Mars level spin gravity does. The outer diameter could be a little under 250 feet. This is a huge project that just might be worthwhile. The first stage could be inside the donut hole with the station torus carrying the fuel as the second stage. The problem that I am aware of is that there have been serious objections raised to reusing a propellant tank as living and working quarters.

    Jon is the propellant depot expert, I’m just trying to rock the boat. I think the only advantage to a torus for a depot would be if that was what was available. An inflatable Bigelow tank might make more sense for a dedicated tank launch.

  22. Randy Campbell says:

    John;

    The “reasoning” behind the lenticular reentry vehicles was that they were found to be dynamiclly stable at high (hypersonic) mach numbers and in the case of one of the smaller versions, (the Apollo version: http://www.astronautix.com/craft/apocular.htm) had a hypersonic L/D of over 4.4!
    (As an aside with that L/D, some external burning to reduce drag and maintain speed during atmospheric skips during skip-glide you can have a world circling RTLS at only Mach 10 :o)

    A thought is also dividing the aft end up into engine modules similar to those suggested for the Advent spacecraft as lighter cheaper version of the linear aerospike engine.

    Randy

  23. john hare says:

    Randy,

    While I believe that the lenticular units can eventually get decent aerodynamic performance, the potential mass efficiency is more interesting at the moment.

Leave a Reply

Your email address will not be published. Required fields are marked *