Thursday, September 25, 2008

Fuel Inerting

While discussing the recent Qantas B747 incident with some very knowledgeable pilots I learned something that bears more research, about oxygen supply and fuel inerting technology. I'll start with some background and an overview of oxygen supply options in aircraft, before I finally get to the cool stuff that I learned.

The simplest way handle the problem of ensuring people on board can breathe is for the airplane to have an unpressurized cabin and non-turbocharged engines and to be operated only at altitudes where the engines and everyone on board has access to adequate oxygen in the outside air. Pretty much everyone who learns to fly does so in an airplane this simple. It works well. The laws of physics and of the government work in concert to forbid climb beyond a safe altitude.

The next method can be retrofitted into any aircraft, and that is to carry an oxygen cylinder and masks. Gaseous oxygen is compressed into the cylinder and the flow regulated by a valve. There are various mask and flowmeter technologies to ensure comfort and efficient use of the oxygen available, and you've read about some of the ones I've used. You can also have a similar oxygen system built in, which just means that the oxygen cylinder is stowed in a custom compartment, as opposed to in the cabin. Individual occupants attach their masks to outlets throughout the cabin. It has the advantage of being integrated with cockpit systems and seeming more sophisticated to the passengers, but the disadvantages that you can't easily take the cylinder somewhere to be filled and that you're married to masks that fit the outlets that were manufactured into your airplane at the time that Galaga was the pinnacle of videogame technology.

Remember that there IS oxygen at any altitude where airplanes fly, and it is present in the same proportions everywhere. You probably already know that air is 21% oxygen, 78% nitrogen and 1% other gases. we only actually need the oxygen to breathe, the rest could be replaced with something else harmless. So that proportion of oxygen stays the same as you go up, it's just that the air at altitude is at a lower pressure, so there is less of it in the same volume. One lungful of air at 18,000' has half as many oxygen molecules as the same lungful at sea level. If there were a way to cram more of that thin air into your lungs, you would have access to just as much oxygen as at sea level. And there is: pressurization. Under normal circumstances, most high altitude aircraft pressurize the fuselage to a breathable atmosphere. It's important to realize that this isn't a matter of sealing in ground level air and keeping it from escaping. The air in the airline cabin is constantly being replaced with air from the outside, it's just that it's compressed and cooled on the way in. A significant portion of a turbine engine is dedicated to compressing air, and there's enough left over, called bleed air to power some aircraft systems and pressurize the cabin. The airplane is fitted with outflow valves to prevent overpressurization. You can sometimes hear them making horrible whistling sounds. This explains why a small hole in the fuselage doesn't cause movie-style airliner disasters: there are already holes in the fuselage to let the air out. Adding another one might make the pressurization system work a little harder, but escaping air is replaced.

Sometimes, like on the Qantas flight I mentioned, something happens to create a bigger hole in the fuselage. The pressurization system can no longer keep up with the outflow, and the airplane depressurizes. The checklist in such a case calls pilots to don their own oxygen masks and commence an emergency descent. That means they cause the airplane to descend just as fast as it structurally can. Kind of a fun one to practice! Pressurized aircraft are required to have a supplemental oxygen system in place to allow everyone to breathe during the time it takes to descend from cruise altitude to ten thousand feet, where there is breathable pressure. In an airliner, pilot emergency oxygen is always supplied from oxygen bottles, as their masks are positive pressure, designed not just to increase the partial pressure of oxygen, but to exclude noxious substances like smoke. That's because pilots have to perform useful functions in an emergency, while passengers just have to stay alive. It's fair, really.

This is all extra scary for unknowing passengers because they witness the depressurization event, probably accompanied by a loud noise. The masks drop, which despite all the safety briefings still awaken some human fright instinct based on snakes dropping out of trees. The drop in cabin pressure is such that moisture in the air may condense into mist, which looks like smoke, and then the airplane plummets. Try to keep in mind, if this happens to you, that the plummeting is deliberate and for your own good. The masks really aren't designed to muffle your screams in order to maintain the concentration of the flight crew, (but I'm not saying that might not be a useful auxiliary function) so put them on and breathe normally.

Now, what are you breathing? In a Boeing 747 you're breathing out of an oxygen bottle, just like the pilots, just with a different mask. The airplane literally contains a big bank of oxygen bottles, bristling with valves and lines running all over the cabin to the passenger oxygen masks. As you breathe in wearing one of those masks you get some cabin air and some extra oxygen from the cylinders. Ironically, it was one of those oxygen cylinders, positioned in the nose of the aircraft, that caused the Qantas decompression in the first place.

On any other airliner, you are not breathing out of a bottle, but breathing oxygen custom made for you in response to your pulling the mask towards you, as directed in the passenger briefing. The compartment above your head contains a chemical oxygen generator, based around a compound that contains oxygen, in the same way that H2O contains oxygen, but less stable. It takes a lot of energy to liberate oxygen from water, but just ignite something like sodium peroxide (Na2O2) and it starts to decompose. Na is elemental sodium, O is oxygen, and it takes two oxygen atoms to make oxygen gas (O2). The sodium peroxide breaks down into sodium oxide (Na2O) and releases oxygen gas. Here's the equation.

2 Na2O2 --> 2 Na2O + O2

You don't have to have understood high school chemistry to count that up and see that there are two Na atoms and 4 O atoms on each side of the equation, with nothing left over. Now, that's pretty neat if you can get around the bit I skimmed over and that is that in order to start this reaction, you set the oxygen generating chemical on fire. When you pull the mask briskly towards you, you are pulling a pin to fire an ignitor. You may even smell it burning. But despite the scary sounding implications, they are maintenance free devices. The only incident I know of where chemical oxygen generators caused a fire ValueJet where a number of the devices were being transported in the cargo hold, mistakenly labelled empty. They should have been deactivated for the flight. A real drawback of chemical oxygen generation is that it is a short lived system capable of putting out about 12 minutes of oxygen, just enough to get down to 10,000'. For flight in areas with a lot of high terrain, supplemental gaseous oxygen is required. Boeing offers this as an option on the B737NG and as a retrofit on other aircraft.

It would be nice to have oxygen available that didn't have to be held in either a pressurized container or an unstable chemical compound. And there is. In the air. There exists technology to separate nitrogen from oxygen in the ambient air. While the unpressurized air at 30,000' doesn't contain enough partial pressure of oxygen to sustain consciousness, the same pressure of close to 100% oxygen does. And the cool part is that generation of oxygen in this manner is just a side effect of fuel tank inerting.

See, when an airplane takes off, typically the tanks are full of fuel. In the absence of fuel tank inerting, as the fuel is burned, the resulting space (cool word alert: it's called the ullage) is filled by ambient temperature air and fuel vapour. Given the right conditions, this mixture of fuel vapour and air can ignite, to disastrous consequences. The idea behind fuel inerting is to fill the fuel tank ullage with something inert. Like nitrogen. And now if you didn't already, you see where this all comes together. During the flight, take the air that is already going by anyway, and make almost pure nitrogen, used to fill the unused space in the fuel tanks, and then have close to pure oxygen as a byproduct. I don't know much about the actual system that will be on the B787 Dreamliner, but apparently the whole apparatus weighs about as much as one passenger with luggage, and draws 40 kW of power.

The Boeing 787 will use this system. The B787 will also make a liar out of me for what I said about bleed air, but this post is about the fuel inerting. It's so symmetrically beneficial, it's almost as if putting my laundry in the washing machine somehow created both clean underwear and groceries.


Anonymous said...

Cool! I didn't know such a device existed for the 787.

I do have one question though. Given the flammability of oxygen. Isn't pumping 100% O2 into the cabin pretty risky?

Anonymous said...

The masks really aren't designed to muffle your screams [...] so put them on and breathe normally.

"Why yes, I always breathe normally when I'm in a six hundred mile per hour uncontrolled vertical dive. I also s%$@ normally... right in my pants!"

RIP, Carlin. :-)

Paul Tomblin said...

I wouldn't want to eat the groceries that came out of that load.

Scott Johnson said...

40 kilowatts? That seems like a really huge amount of power at first glance. A quick check shows that on the B747-400, though, each engine-driven generator is good for 90 KVA, which is 72 kilowatts at a "normal" power factor, and three good ones are required for dispatch. The APU is good for 144 kilowatts more. I imagine the 787 probably has a similar power budget, so it probably fits in just fine.

Increasing the oxygen concentration in the cabin atmosphere, as Blake said, sounds like a recipe for another Apollo 1, but then again, that was 100% oxygen at sea level pressure. Fire hazard, like breathing potential, varies directly with oxygen partial pressure, not concentration.

Anonymous said...

Here's the equation.

2 Na2O2 --> 2 Na2O + O2

You don't have to have understood high school chemistry to count that up and see that there are two Na molecules and 4 O molecules on each side of the equation, with nothing left over.

Almost, but not quite. There's 4 atoms of Sodium and Oxygen on each side of the equation. There's two molecules of disodium dioxide (aka disodium peroxide) on the left, and two molecules of disodium oxide (aka sodium oxide) plus one molecule of (diatomic). But the point still stands, it all adds up.

Anonymous said...

scott: the budgets aren't that similar because the 787 eschews bleeds for most uses, and thus has much more generation available. Something like 500kw per donk IIRC.

Aviatrix said...

David-t said: Almost, but not quite. There's 4 atoms of Sodium and Oxygen on each side of the equation.

Oh, how embarrassing. I can't believe I said molecules instead of atoms. I know the difference very well. I might even still be able t explain sp3 hybidization. Thank you for catching the slip. I'm editing the article to correct it now.

Anonymous said...

According to this article, the 787 will have two 250 KVA generators per engine, with two 125 KVA generators on the APU.
I have heard more recently that the APU now will have two 250 KVA generators, but I can't find a source for that.

Dave Starr said...

This is an excellent article to provoke thought. The advantages of in flight tank inerting have long been known ... the original USAF C-5's for example had a workable Nitrogen inerting system which filled the ullage as fuel was burned but it was quickly decommissioned. Inflight inerting of civil airlners has been propsed at various times for years but always rejected by airline executives as 'unnecessary' (wonder if former TWA executives (Flt 800) still think so).

Boeing's idea of putting the oxygen to good use is indeed a breakthrough idea. More inventions are stillborn by the 'it can't be done' thought process than by technical considerations. Negative mindset is science's single greatest obstacle.

An important reason to carry much more than 10 or 12 minutes supplemental oxygen that I don't see much though applied to is range. Assume an airliner in mid-Pacific. A failure of cabin pressurization occurs, necessitating descent to 10,000 feet. Want to bet there is enough fuel on board to reach dry land given the vastly increased fuel flow?

Final observation. Oxygen is _not_ flammable.

Anonymous said...

dave: current flight planning takes long range scenarios with emergency descents into account. quoting the FAA on the recent final rule on ETOPS:

"many aircraft have passenger oxygen systems that allow extended operations at 14,000 feet."

"The final rule prohibits the dispatch or release of a flight by an airplane with more than two engines for more than 90 minutes at full cruise speed unless it has adequate fuel, considering wind and weather conditions, assuming a rapid decompression, followed by descent to a safe altitude to fly to an adequate airport, including enough fuel to hold for 15 minutes at 1,500 feet."

etc. etc.

feel free to go down the regulatory rabbit hole and read this stuff, quite fascinating really.

Anonymous said...

Very cool stuff and educational too. I didn't know some of this, and I used to maintain passemger O2 systems - of course that was a long time ago, on DH Comets and B727-100s. Things have changed.

Pumping 100% O2 into the cabin does not increase fire risk appreciably because 1) there is very little of it and it passes through the passenger's lungs first, and 2) the pressure is so low that an internal fire would probably not stay lit after de-pressurization. That's why fuel tank explosions have been rare, (not enough O2 partial pressure), but unfortunately, not rare enough.

Dave Starr said...

Garrett: thanks for the correction/explanation. I'm not a long-range pilot nor a qualified dispatcher ... the original perceived problem was pointed out to me by a guy transitioning to the 747-400 ... mainly surrounding issues he was raising with several single points of failure in the pressurization system. The fuel situation was brought up as a possible "bad outcome".

Could be he hadn't yet worked all the tables, or could be someone re-wrote the reg smarter since then ... either way it's one of those situation where I am happy to be mistaken/corrected. Thanks.