# How does Orbex save structural weight by using coaxial tanks?

British aerospace company Orbex is developing a small orbital rocket called Prime, which uses propane and liquid oxygen in two coaxial tanks.

One key aspect of propane is that it remains liquid at cryogenic temperatures. That enabled a “coaxial tank” design for Prime where a central tube of propane is surrounded by an outer tank of liquid oxygen, creating structural mass savings in the rocket.

How are inner and outer tank structurally linked together, and where come weight savings from?

To add food of thought, I came across these pieces of text found here

"Prime launchers are up to 30% lighter and 20% more efficient than any other vehicle in the small launcher category, packing more power per cubic litre than many heavy launchers."

"If only the outer LOX tank needed to bear structural loads and the inner propane tank just needed nothing more than to physically separate that fuel from the LOX with no thought to insulation, might that account for the 30% mass savings?"

"If the inner tank was even inflatable like a bladder or balloon within the LOX, might a single pressurization system suffice for both propellants?"

"Flexible cryogen bag tanks are apparently a thing, so tension anchoring the thing in the LOx tank is certainly doable."

• Thermal insulation (if any) is needed for the outer oxygen tank only. Bending forces are handled by those outer tank walls mostly. Forces caused by tank pressurization are taken by the outer walls only. Thus inner tank walls may be thinner and lighter. – Uwe Aug 18 '19 at 23:18
• The "tank pressurization are taken by the outer walls only" requires some operational restrictions that can be pretty tough to live with. You'd have to fill and pressurize both tanks together, and similarly drain them together if you didn't launch. – Bob Jacobsen Aug 20 '19 at 0:34

Firstly, there could well be potential mass saving for this design:

• Less tubing needed to get propellants to engines.

• greater surface area to volume ratio of overall tanks, but probably not by much as overall shape determines the surface area to volume ratio and the only additional issue would be the bulkhead, but this is mostly cryogenic-to-cryogenic, so not much saving. Further LOX boils at -183C,Propane freezes at -188C, while there is a bit of overlap, and pressures play a role it likely this will need some form of insulation, with more surface area than the bulk head separation.

• Only one tank to mount. With only one outer structural tank, directly above the engines, the structural mountings etc might be simpler, though this is offset by the increased pressure of having a taller tank under high g.

However, its worth noting:

• They don't specifically claim the tank design is the cause of the 30% mass savings (and the "up to" makes the claim pretty meaningless).

• Unless what they are comparing it to is not sensible, I highly doubt the tank design alone could be responsible for anywhere near a 30% dry mass saving.

• This is their own marketing material, so take it all with a pinch of salt. Though this could just be prejudice against the guys who brought you the "cubic litre" [sic]...

@Uwe mentioned important facts already in a comment. The most important one, I am rather sure, is the surface to volume ratio:

Given a cylinder of radius $$r$$ and height $$h$$, its surface to volume ratio is $$2( r +h )/(rh)$$.

If the radius or the height or both of them are increased, the ratio drops. This means you need less wall structure to house a certain volume as opposed to adding another separate tank. This is where the Orbex design is advantageous. As Uwe mentioned, the inner wall of the coaxial design does not carry many loads and can be very lightweight.

In the classical design, where there are separate tanks for fuel and oxidizer, each of them has its own ratio and there is none such “synergetic“ effect.

Of course, actual tanks are not cylindrical, but this argument holds qualitatively for any shape. The rule "bigger means comparatively less surface" is very general. Many examples can be found in engineering and in nature.

Coming to think about it, this idea combines the best of a mono-propellant consisting of pre-mixed fuel and oxidizer and safe and stable separate fuel tanks. The pre-mixed infernal stuff might actually even save volume due to the volume reduction effect of mixtures. But that wouldn't make up for its risk.

• I don't understand your calculation. Simplify Orbex to 1m diameter, 10m tall. With stacked tanks, you provide three ends plus a 10m *2pi*1m outer wall. With concentric tanks, the outer skin is still exactly the same size, you lose the middle end, and you add an inner skin that's about 2/3 the area of the outer skin. – Bob Jacobsen Aug 20 '19 at 0:37
• @BobJacobsen The radius is included in the calculation to make the answer more general. You (and Christoph in his comment to the question) are correct that in this specific case, the only thing that might be saved is a bulkhead, maybe two, depending on the design. The calculation remains valid. If you have any specific question about my calculation, I'd be happy to answer it. – Everyday Astronaut Aug 20 '19 at 10:19
• Let me rephrase, as I can't edit my comment: How can the surface to volume ratio be the most important thing when it doesn't change? – Bob Jacobsen Aug 20 '19 at 15:26
• @BobJacobsen it obviously changes if you save one or two bulkheads. But yes, after re-thinking it, I am unsure if this is really the most important thing in this specific case. – Everyday Astronaut Aug 23 '19 at 6:03