I just graduated from NCSU's aero department (www.mae.ncsu.edu)
with these guys- I did the aeronautical design project, but the Tumbleweed project
turned out alot better than I expected. From what I heard, the guys at NASA were
much more enthusiastic and receptive to their design than previous years' projects
(a mars balloon, among others).
As far as steering, several options I remember hearing their team members discuss
throughout the year were actuating the planar sails on the inside
of the carbon fiber ribs, some sort of anchor for times when prevailing winds
were in unfavorable directions, and changing the mass distribution and inertial
characteristics. However, I think the point is that they wanted a simple, lightweight,
easily reproducible design; the whole idea of a tumbleweed is based
on utilizing what you don't have to bring along (interestingly enough, it was
first thought of when a test rovers wheel blew away and escaped the crew
out in the desert, I believe).
I remember their presentations mentioning research they did on the atmosphere
and typical wind climate, average surface qualities (rock size, etc), and how
it would affect their design. The biggest concern I heard judges from industry
at the southeast AIAA student conference (http://www.AIAAstudentconference.org)
echo was over oddly shaped debris gathering inside the design and weighing it
down.
The pictures you see on their website are from the senior design picnic less
than a month ago. The actual design is even larger. The wind was calm, but it
would only take about a 10 knot gust to start rolling. And yes, the girls in
the pictures are real engineering students, and I know them all- don't worry,
they're a lot smarter than to let you guys explore any surfaces. But I'll let
them know you brought it up. Also, Dustin isn't copping a feel. I think his
family was a few yards outside that picture, but it does look sort of funny.
If anyones curious, look up our design project too (http://www.mae.ncsu.edu/courses/Mae478-479/team3/ webpage/frameset.htm)
on a hopefully soon-to-be autonomous 200 knot jet aircraft (pretty decent for
~12 lb. thrust). The section pages are still empty, but there are lots of pictures
and a couple videos.
i just got my aero degree yesterday (literally)- it's already reassuring to see headlines like this everyday.
on a more serious note, i'm continuing to a graduate degree in CFD studies, but it is a huge mistake to get rid of too many unique resources like this. our AIAA chapter just had a guy down from Langley speaking about research in such tunnels, and while I know they are antiquated, so is most prevalent consumer aerospace technology. regardless, the experimental side of aerodynamics is important; many boundary layer methods are based on curve-fits or redundantly-proven data obtained from these experiments.
computational aerodynamics gives a (relatively) cheap and widely available way to conduct "testing" which otherwise might not feasible or achievable (i.e. chemically reacting, high temp, hypersonic, unusual reynold's number, or varying composition flows), but errors do occur (approximation, method, roundoff, etc). theoretical aerodynamics gives good background, and provides understanding, a way to interpret results, and intuitive explanations for correcting problems with design. however, experimental aerodynamics are still extremely important. computer simulation and theoretical approaches can only take you so far.
as an example, on our senior design project, SLA models were dontated for wind tunnel testing, in addition to validation using CMARC (computer code) and traditional analysis on paper. while the computer simulation provided the most usable information (stability derivatives, lift values), and traditional analysis came next (drag buildups on paper-the computer code was inviscid), the wind tunnel gave alot of information that could not be obtained before flight and with reliability any other way (stall patterns on the wing, neutral point determination, etc). however, the SLA model was much more expensive than the entire rest of our project.
wind tunnel testing on a large scale can alleviate errors (nondimensionalization, boundary layer buildup on tunnel walls, measurement error, etc) and provide otherwise unattainable results in some cases. as a pilot and aerospace grad, i wouldn't want to test-fly a vehicle only proven in computer code and on a two-foot scale in a wind tunnel.
Slashdot Mirror
As far as steering, several options I remember hearing their team members discuss throughout the year were actuating the planar sails on the inside of the carbon fiber ribs, some sort of anchor for times when prevailing winds were in unfavorable directions, and changing the mass distribution and inertial characteristics. However, I think the point is that they wanted a simple, lightweight, easily reproducible design; the whole idea of a tumbleweed is based on utilizing what you don't have to bring along (interestingly enough, it was first thought of when a test rovers wheel blew away and escaped the crew out in the desert, I believe).
I remember their presentations mentioning research they did on the atmosphere and typical wind climate, average surface qualities (rock size, etc), and how it would affect their design. The biggest concern I heard judges from industry at the southeast AIAA student conference (http://www.AIAAstudentconference.org) echo was over oddly shaped debris gathering inside the design and weighing it down.
The pictures you see on their website are from the senior design picnic less than a month ago. The actual design is even larger. The wind was calm, but it would only take about a 10 knot gust to start rolling. And yes, the girls in the pictures are real engineering students, and I know them all- don't worry, they're a lot smarter than to let you guys explore any surfaces. But I'll let them know you brought it up. Also, Dustin isn't copping a feel. I think his family was a few yards outside that picture, but it does look sort of funny.
If anyones curious, look up our design project too (http://www.mae.ncsu.edu/courses/Mae478-479/team3/ webpage/frameset.htm)
on a hopefully soon-to-be autonomous 200 knot jet aircraft (pretty decent for
~12 lb. thrust). The section pages are still empty, but there are lots of pictures
and a couple videos.
i just got my aero degree yesterday (literally)- it's already reassuring to see headlines like this everyday. on a more serious note, i'm continuing to a graduate degree in CFD studies, but it is a huge mistake to get rid of too many unique resources like this. our AIAA chapter just had a guy down from Langley speaking about research in such tunnels, and while I know they are antiquated, so is most prevalent consumer aerospace technology. regardless, the experimental side of aerodynamics is important; many boundary layer methods are based on curve-fits or redundantly-proven data obtained from these experiments. computational aerodynamics gives a (relatively) cheap and widely available way to conduct "testing" which otherwise might not feasible or achievable (i.e. chemically reacting, high temp, hypersonic, unusual reynold's number, or varying composition flows), but errors do occur (approximation, method, roundoff, etc). theoretical aerodynamics gives good background, and provides understanding, a way to interpret results, and intuitive explanations for correcting problems with design. however, experimental aerodynamics are still extremely important. computer simulation and theoretical approaches can only take you so far. as an example, on our senior design project, SLA models were dontated for wind tunnel testing, in addition to validation using CMARC (computer code) and traditional analysis on paper. while the computer simulation provided the most usable information (stability derivatives, lift values), and traditional analysis came next (drag buildups on paper-the computer code was inviscid), the wind tunnel gave alot of information that could not be obtained before flight and with reliability any other way (stall patterns on the wing, neutral point determination, etc). however, the SLA model was much more expensive than the entire rest of our project. wind tunnel testing on a large scale can alleviate errors (nondimensionalization, boundary layer buildup on tunnel walls, measurement error, etc) and provide otherwise unattainable results in some cases. as a pilot and aerospace grad, i wouldn't want to test-fly a vehicle only proven in computer code and on a two-foot scale in a wind tunnel.