[wcoenen]

I was wondering about the same thing. It doesn't seem to be explained on the company site.

The only thing I could come up with is to have neutral buoyancy air blatters that can expand/contract. These would be inside the volume of buoyant gas but connected to the outside air. Then pump the buoyant gas into a pressure tank to reduce buoyancy, or release it from the tank to increase buoyancy.

[nuccy]

Yes, but the size of the blatter would be enourmous for 250 claimed tonnes: 1 m^3 of air weights 1.21 kg, so if to reach neutral buoyancy you use helium (0.18 kg / m^3) or vacuum (0 kg / m^3, best case), then to counteract lifting/releasing of just one tonne (1000kg) you need to offset 833 m^3 with vacuum and 970 m^3 with helium. Having blatter inside and having such high expension requirements means that the vehicle effectively needs two shells, which will add weight and complexity. Stretchable main "ballon" might be simpler.

[wcoenen]

That makes sense. It would indeed be simpler to make the buoyancy volume variable in that case.

I'm not sure though that fabrics exist which are both stretchable and can hold on well to small molecules like helium. (e.g. mylar can be stretched only 4% at most.) Maybe some type of "accordion" type connection between the bottom and top halves of the saucer-shaped balloon would work.