Oh man, am I gonna throw a wrench in the works. I started looking up bubble accelerations, and… it looks like everyone’s wrong. Apparently, rising particles can accelerate faster than g, depending not only on the difference in density between the particle and surrounding fluid, but also the shape of the rising particle.
Here’s Exhibit A, Particle force balance equation used in FLUENT (commercial fluid dynamics analysis software): The equation of motion used in FLUENT is
du[sub]p[/sub]/dt = F[sub]D[/sub](u - u[sub]p[/sub]) + g([symbol]r[/symbol] - [symbol]r[/symbol][sub]p[/sub])/[symbol]r[/symbol][sub]p[/sub] + F[sub]x[/sub]
where the subscript “p” refers to the particle in question, so:
u[sub]p[/sub] = velocity of the particle,
u = velocity of the surrounding fluid,
du[sub]p[/sub]/dt = acceleration of the particle,
F[sub]D[/sub] = drag force coefficient,
g = gravity,
[symbol]r[/symbol] = fluid density,
[symbol]r[/symbol][sub]p[/sub] = particle density, and
F[sub]x[/sub] = a virtual force due to the “virtual mass” of the fluid; for a sphere,
F[sub]x[/sub] = 1/2([symbol]r[/symbol]/[symbol]r[/symbol][sub]p[/sub])*d/dt(u - u[sub]p[/sub])
What does this all mean? Well, in still fluid (u = 0), and for low velocities, like on takeoff (u[sub]p[/sub] = 0), the EOM simplifies to:
du[sub]p[/sub]/dt = g([symbol]r[/symbol] - [symbol]r[/symbol][sub]p[/sub])/[symbol]r[/symbol][sub]p[/sub] - 1/2([symbol]r[/symbol]/[symbol]r[/symbol][sub]p[/sub])*du[sub]p[/sub]/dt
or
du[sub]p[/sub]/dt = g([symbol]r[/symbol] - [symbol]r[/symbol][sub]p[/sub])/([symbol]r[/symbol][sub]p[/sub] + 1/2[symbol]r[/symbol])
And that means, for very light spherical particles ([symbol]r[/symbol][sub]p[/sub] ~= 0), that the maximum acceleration = 2g. Moreover, this max acceleration depends on the shape of the particle (or balloon, if you will). This wasn’t what I was expecting, but it does make some intuitive sense to me: a long slender balloon will be “moving” less mass around at a given time than a lumpy-shaped balloon.
And, for those of you who want even more background, I present Exhibit B, Textbook deriving equations of motion for bubbles. If you scroll down to section 5.10 (equations of motion), you’ll see the following:
Noting the change in nomenclature, this agrees with the FLUENT page quoted above. This conclusion is supported by extensive derivation and citation, so I tend to think it’s pretty definitive.
However, I must admit that the subject matter in the derivation is more than a little out of my area of expertise, so I can’t comment on it’s validity directly. I do, though, work with a guy who got his PhD working on cavition bubbles, so I’ll ask him for input on Monday.