Onboard the space shuttle Columbia (STS-107), experimental flame balls have been doing some strange and wonderful things--e.g., flying in corkscrew patterns and beating like human hearts.

"[I'm calling this one] Howard," deadpanned astronaut Dave Brown onboard the space shuttle Columbia (STS-107) this week. He had been filming the tiny flames for some time, watching them roam around their test chamber in a lifelike search of food (fuel), when the idea popped into his head. These flame balls needed names.

Above: One of the nine flame balls in this video snippet is named Kelly. Read on to find out what's special about her. Image credit: Paul Ronney and the crew of the space shuttle Columbia (STS-107). [more video at http://carambola.usc.edu/research/SOFBALL2quickie.html]

"After that everyone started naming them," says USC engineering professor Paul Ronney who designed the experiment. "It was fun. It also helped us keep track of some of the strange things we saw." For example, two flame balls flew around in a spiral pattern like DNA. "We called them Crick and Watson."

It's more than just fun, though. These flame ball experiments--called SOFBALL, short for Structure of Flame Balls at Low Lewis number--are serious investigations into the physics of fire.

Unlike flames on Earth, which have a tear drop shape caused by air rising in a gravitational field, flames in space break apart into spheres a few millimeters in diameter. A typical floating flame ball produces 1 to 2 watts of thermal power--much less than, say, a 50 watt birthday candle. "We created some flame balls on STS-107 that emitted only 0.5 watts--a record low," he says.

Flame balls are "lean" burners; they don't need much fuel to keep going. Engineers would love to duplicate their efficiency in the engines of automobiles, "but first we have to understand how flame balls work," says Ronney.

That's the goal of SOFBALL.

SOFBALL is a chamber about the size of an office trash can filled with combustible gases: "a little bit of hydrogen or methane (the fuel), some oxygen (the oxidant), and a lot of inert gas (e.g., helium or nitrogen) to dilute the mixture until it is barely flammable," says Ronney. The experiment rides in the shuttle's cargo bay inside a rack called the Combustion Module. Astronauts simply press a button to spark the mixture and voila ... flame balls. Their temperature, brightness, heat loss, and the composition of their gaseous byproducts are all monitored by built-in instruments. SOFBALL was built and tested at NASA's Glenn Research Center under the guidance of project scientist Karen Weiland and project manager Ann Over.

Above: The SOFBALL Experiment Mounting Structure (EMS) has a built-in spark ignitor, temperature sensors, a radiometer and more at http://microgravity.grc.nasa.gov/combustion/sofball/sofball2_hardware.htm.

This is SOFBALL's second flight; the first was in 1997, also onboard Columbia (STS-83). In those days the experiment ran for only 8 minutes. "We didn't think flame balls could last more than a few minutes," explains Ronney, "but we were wrong. Many of them were still burning when SOFBALL's control computer automatically ended the test. We needed more time."

So, during STS-107, SOFBALL has operated for periods as long as three hours. And "we've seen some extraordinary things," says Ronney.

Crick and Watson are examples. Ronney says he has no idea what would make a flame ball fly around in a spiral. "Flame balls move for two reasons," he explains. "First, when they exhaust the fuel in their vicinity, they drift toward regions with more. They follow the fuel like a little organism. Second, they can drift due to slight accelerations of the shuttle." Neither of these effects would produce a corkscrew flight path.

Left: A spiraling flame ball. Image credit: Paul Ronney and the crew of the space shuttle Columbia (STS-107). [more video at http://carambola.usc.edu/research/SOFBALL2quickie.html]

Howard is another example of something Ronney had never seen before. "Howard was suicidal," he says. Instead of following the fuel like a flame ball should, Howard headed straight for the walls of the chamber--a fuel-poor region. "He promptly went out. We saw several more flame balls like this, too." It's another mystery.

The SOFBALL experiment also produced the biggest and the smallest flame balls ever recorded--ranging from 2 mm to 15 mm across. "We named one of the biggest ones 'Zeldovich' after the Russian physicist who predicted flame balls in 1944." A well-meaning astronaut named one of the flame balls 'Paul Ronney,' "but it turned out to be small and short-lived--a wimp," Ronney laughed.

Oscillating flame balls were another first. "About 15 years ago John Buckmaster at the University of Illinois and Guy Joulin of CNRS in Poitiers, France, predicted that flame balls about to run out of fuel should oscillate. You've probably seen something like this in low burning candle flames, which jump up and down in a rhythmic pattern just before they go out. We had never observed these oscillations in flame balls before, but now we have--in two flame balls named Buckmaster and Joulin." The period and duration of the pulsations reveal a great deal about the inner workings of flame balls, adds Ronney. It's a very important result.

Right: A SOFBALL radiometer signal showing flame ball oscillations. [more at http://carambola.usc.edu/research/SOFBALL2quickie.html]

Ronney's favorite flame ball, though, is Kelly. "Before the mission began I said I wanted to send a flame ball around the world. Kelly almost made it." The shuttle circles Earth once every 90 minutes; Kelly burned for 81 minutes--the longest-lasting flame ball ever recorded.

"Kelly's experience is a fascinating example of group dynamics among flame balls," says Ronney. "She was created, one of nine flame balls, in a gaseous mixture of hydrogen, oxygen and sulfur hexaflouride (the inert filler). All the others began drifting around the chamber, looking for food, competing with one other, while Kelly remained motionless at the center. Before long, the others were exhausted; they had drifted too close to the walls and winked out. Kelly was left all alone with a chamber full of fuel."

"It pays to be patient," notes Ronney. And he should know. Ronney discovered flame balls in 1984 in a drop tower at NASA-Glenn in Ohio where the weightless lifetime of a flame ball is only a few seconds. He's been waiting almost 20 years for data like this....

There's a sign in Ronney's office: When the Gods want to punish you they answer your prayers. "It's going to take us years to analyze all these results!"

He's delighted.

Editor's Note: Astronauts ignited and filmed 55 flame balls during the STS-107 mission (which is still ongoing as this story is being published); 33 of them received names after trend-setter Dave Brown christened one of his "Howard."

Dr. Paul Ronney (http://carambola.usc.edu/) of the University of Southern California has prepared a summary of SOFBALL results (http://carambola.usc.edu/research/SOFBALL2quickie.html) from STS-107.

SOFBALL home page (http://microgravity.grc.nasa.gov/combustion/sofball/sofball_index.htm) (NASA/Glenn); SOFBALL experiment fact sheet (http://spaceresearch.nasa.gov/sts-107/107_sofball.pdf) (OBPR)

more NASA combustion research links: Candle Flames in Microgravity (http://microgravity.grc.nasa.gov/combustion/cfm/cfm_index.htm) (NASA/GRC); Microgravity Combustion website (http://microgravity.grc.nasa.gov/combustion/) (NASA/GRC); Combustion Module Home Page (http://microgravity.grc.nasa.gov/combustion/cm/cm_index.htm) (NASA/GRC)