[The Following is an excerpt from a letter to Allen Meece]

I have a new suggestion for further reducing the VBP's weight. We can change the shape of the balloons used from cylindrical to the more common conical type which is proven its utility for low operpressures.

The maximum spacing between lift cells is only slightly less than that for an array of vertical, cylindrical lift cells. The radii of the cells must be greater for the same number of cells using conical cells. However, the tapering of the bottom of the lift cells allows them to be moored very close together without great compression or reduction in volume. There is significantly less total open space between the cells.

Most of the space between the balloons must be shielded by the shroud so that the crew can work there. This means than open space between the balloons represents additional cross sectional area for drag to act on the platform.

A lift cell array of conical balloons has about half the frontal surface area of one using the same number of cylindrical balloons. This is a significant reduction in platform drag.

A staggered arrangement of the lift cells is useful because it props the lift cells up against being pushed around by drag force. Keeping inflated balloons in contact lets the friction of their contact act to keep them steady. And by properly arranging the cut of the balloon fabric, we can allow orderly expansion of the balloons during the ascent without their own expansion forcing them to rub against one another. This staggered array is no different in this regard than others we've examined.

As for motion of the balloons from side to side (as from tacking or transverse waves in the array), a group of balloons moored at a single point experiences less abrasion between the balloons than a group of balloons in contact but moored at separate points.

The lift cells will tend to rotate about their mooring point. Two lift cells moored at the same point will rotate in unison, ideally with no relative motion. This is especially true if they begin motion in contact over a large surface area, as greater contact area means greater friction to resist motion relative to each other.

If moored at two separate points, the motion of balloon A need not affect balloon B until A moves over far enough that it first contacts B. This means that the angular momentum of two lift cells moored at the same point is twice as much as a single lift cell. Thus, a single lift cell rotates twice as fast in response to the same impulse.

If balloon A keeps moving, it will start B rotating as well. Unfortunately, this pushing will cause A to rub against B is their cross sections don't change. In B's reference frame, balloon A taps it at point 1 and drags along its surface to point 2, imparting motion.

So, in addition to reducing forward surface area, an array of conical lift cells is less prone to rupture due to abrasion by large oscillations in the array because it reduces relative motion.

Another important advantage to this configuration is that it can be expanded. If we suddenly find ourselves in need of 50% or so more lift, we can raise additional balloons and hydrogen to the platform and deploy them symmetrically along the sides.

The balloons would have to be deployed outside the shroud (not shown here), but hopefully by that point the platform will be high enough to bear the extra drag, and if the array is at full inflation they won't roll around too much. The angle is a bit steep to expect no compression, but it shouldn't be more than the array can handle.