I don’t know much about US-China diplomacy or the finer points of espionage, but after 10 years working at Loon I do know something about high-altitude ballooning.

Types of high-altitude balloons (HABs)

There are three basic types:

• Latex: a latex balloon filled with helium, i.e. your basic workhorse weather balloon. As they rise the external air pressure drops and the balloon expands, and usually after a few hours they get so big they pop. It’s possible to under-fill them (intentionally or not) such that they reach a steady float altitude before bursting at which point they might last a day or two before coming down, but these aren’t used for long-duration flights or for heavy payloads because of the strain on the material.
  
• Zero-pressure: an inverted bag of helium or other lift gas with a hole at the bottom. As you go up and the external air pressure drops the lift gas will poor out the hole, so the internal pressure is always the same as atmospheric pressure (thus zero pressure differential). These can go for a night and a day, but without a way to replenish your lift gas (like with a hot-air balloon) they’ll typically come down shortly after your first sunset as the lift gas cools.
  
• Super-pressure: sealed mylar or other non-elastic balloon where the lift gas is higher pressure than the atmosphere at float altitude. Think of your more expensive mylar party balloon that’s tight with helium when it first comes from the store. These balloons can last for a long time because you don’t lose any lift gas, and as long as you have enough superpressure your float altitude won’t change with temperature. These balloons are harder to manufacture because they need to withstand higher forces, but they’re the standard balloon for long-term missions.

Given its size, shape, and the size of its payload I assume the Chinese balloon was a standard helium superpressure.

How do they maneuver?

One thing that caught my attention is that the Pentagon claimed that the balloon was “maneuverable”, though they wouldn’t go into any more specifics than that other than to say the balloon had “changed its course”. The Chinese Foreign Minister similarly stated that the balloon had “limited self-steering capability”, but again we’re left to speculate the extent of that control.

Balloons are basically giant sails, so for the most part they go with the winds. You can make fine adjustments with propellors (think blimps), but for the most part you steer by changing altitude to catch the winds you want. Altitude is just a function of mass and volume, so by changing one or both you can change altitude and thus direction. That gives you a handful of variants — there are some hybrid approaches and we played with a few wild ideas in the early days of Loon, but these are the main methods I know of:

• Fire and forget: With good wind forecasts you can run forward simulations to figure out roughly where a balloon will go at a given altitude, and to some extent you can control where you’ll go by varying your total mass, amount of lift gas, launch time and launch location. Error will accumulate over time (and wind forecasts in the stratosphere aren’t all that great to begin with) so this isn’t supper accurate, but it can be OK if you just want to hit a general area.
  
• Ballast dropper / helium vent: A ballast dropper is a little device that sits on your payload and drops a controlled amount of ballast (e.g. lead shot), either autonomously or by radio command. They’re simple and robust, but can only be used a limited number of times and can only be used to go higher. Their counterpart is a vent which lets out a controlled amount of lift gas, bringing you lower at the cost of longevity.
  
• Montgolfiere (hot-air balloons): These are zero-pressure balloons that go up by heating a lift gas, and down by venting. Typical hot-air balloons have a burner at the base for heating the air, but so-called solar Montgolfiere balloons use the sun to heat the lift gas during the day and infrared from the Earth to heat them at night. The French Space Agency have circumnavigated the globe using solar Montgolfiere, and JPL has looked at using them to explore the surface of Mars and Titan.
  
• Air ballast (ballonet): Loon’s altitude control system worked by pumping air into a ballonet (small balloon inside the main superpressure balloon). Since the volume of the superpressure air bag is fixed pumping air raises the mass and the balloon goes down; opening vents releases the air and the balloons go back up. (Pro tip: we eventually figured out that a reverse-ballonet, with the lift gas on the inside and air pumped to the outer balloon, worked just as well and reduced the leak rate.)
  
• Propellors: Propellors won’t give you as much maneuverability as on an airplane, but they can still give some additional control when the wind isn’t going your way. The downside is they tend to be complex, power-hungry, and require shaped balloons to have much impact.

From the images I’ve seen of the Chinese balloon looks like a big spherical superpressure balloon. The envelope at all aerodynamic so I seriously doubt it has propellors, and though the images are low-resolution it looks like the balloon is sealed at the bottom so I doubt they have a ballonet either (though I can’t completely rule them out). We might learn more when they fish the payload out of the ocean, but my guess is that the balloon was equipped with just a simple ballast dropper and/or lift-gas vent. That honestly works fine for short-to-medium flights so long as you pack them with enough ballast and gas.

How well can they maneuver?

Depends on where you are and time of year. Sometimes the winds are blowing in different directions at different altitudes and you can literally do loop-the-loops all day long, but other times you have no choice where you go. As a general rule maneuverability is better in the tropics than the temperate latitudes, and better in the spring and fall than winter or summer. Right now maneuverability in the Northern hemisphere is pretty bad — the winds are all coming from the west, so you’re pretty much stuck going east though you might be able to choose between northeast and southeast if you’re lucky. Of course, if your intention is to go from China to the Continental US then current winds make it pretty easy.

Is China’s explanation plausible?

China claims that this was a civilian research balloon gone astray, and that sort of thing certainly happens. In my ten years at Loon we had a handful of “Flying Dutchman” flights where the components that were supposed to cut the balloon down failed. One early failure took out the entire communications system, and as luck would have it it was also one of our most robust envelopes to date. We wound up tracking the balloon through a combination of simulations and following its progress via UFO sitings on local newscasts. That one wound up being our first trans-continental flight, finally touching down in Montreal. A couple years later we lost control of another balloon when both the primary and backup cutdown systems failed, and since that balloon was supposed to just be a quick test it didn’t have any altitude control system installed. We wound up tracking it as it went across the US, across the Atlantic and eventually passed through the UAE and, yes, even briefly entered China. So yes, it happens.

That said, I’m skeptical that this was a civilian mission. Once Loon balloons got past the Android-phone-in-a-Styrofoam-cooler stage we always mounted a transponder on them and coordinated with the regional air traffic control, and the handful of times Loon lost control of a balloon we made sure to alert every country it entered. Apparently whoever was in charge of this balloon did none of these, even though the payload appears comparable in size. The payload also had a couple of big solar cells, which you only need if you’re planning a multi-day mission (otherwise you can just pack extra batteries). With current winds it takes about 2.5 days to cross all of mainland China, so while it’s possible the cutdowns failed on what was planned to be a short mission it seems more likely that they planned to go into the US from the start.

Hopefully we’ll learn more once they collect the debris from off the Carolina coast, assuming any of what they find becomes public.

[Migrated from Google+]

For 2 out of every 3 people on earth, a fast, affordable Internet connection is still out of reach.  Today we are unveiling our latest moonshot from Google [x] – Project Loon: balloon-powered internet access. Learn more: goo.gl/3VUHC

[Migrated from Google+]