FLL 90 UV data - anything?

5 views
Skip to first unread message

Brian Davis

unread,
Aug 3, 2008, 9:28:35 PM8/3/08
to HALE TEAMS
After looking over the ozone data from the LUXPAK team, I was curious
about David Levy's FLL 90 team. I know you guys had a UV sensor under
a filter wheel, and was wondering a couple of things, if you've had
the chance to look over the data:

First, how did the data come out? Looking at the data from my CDS
sensor, I'm wondering if you got direct light into the sensor through
the filter wheel - it was on top, correct? The CDS sensor shows the
brightest peak when the sensor was looking sideways, presumably as it
rotated in and out of view of the rising sun. If any of my CDS data
(almost completely uncalibrated at this point) can help, I'd be glad
to send it along.

Second, you might want to look at the LUXPAK teams ozone readings. I'd
expect the baseline UV reading to rise throughout the mission, as the
sun rises higher and therefore is both angling closer to the sensor
and passing through a smaller amount of atmosphere. But in addition to
that, there may be a... "modulation", for want of a better word... as
the payload at very high altitude not only has a lot less atmosphere
around it, but also less ozone between it and the sun (a major factor
obviously).

Third, you mention having the NXT control the heaters... did this
work? Do you have any temperature data from the mission?

--
Brian Davis

David Levy

unread,
Aug 3, 2008, 11:18:22 PM8/3/08
to hale-...@googlegroups.com
Hi Brian,

I have not had a chance to assemble the team again to evaluate the
data. I have eyeballed the files and have an initial impression of
the results however I'd like to see how the kids interpret it on their
own without clouding their judgement.

I can however give you some background on design decisions as well as
assumptions leading up to the launch.

Yes the kids' originally had the desire to point the probe out the
side of the box with the intent of controlling the payload from
spinning or rotating the probe toward the sun via a compass or gyro
sensor. Unfortunately those designs proved to be out of reach of
their capabilities so they opted for a more simple approach of
sticking the payload out the top. The thought was enough sunlight
would register from an upward angle that they would not need to worry
about payload spin.

The team did start to question that design when they learned that the
launch took place at sunrise ( ballon#2 went up first) and also
mistakenly assumed that their flight duration would be longer then
what was experienced/
Whether these oversights ( the by team) resulted in indeterminate data
remains to be evaluated.

It may be a few weeks before I can get the team together to analyze
the data. However it may be a good exercise to gather a few team
members to extract the data from the returned payload and send it to
you in a raw format while you are awaiting the team's final
determination.


David

Brian Davis

unread,
Aug 4, 2008, 7:56:27 AM8/4/08
to HALE TEAMS
On Aug 3, 11:18 pm, David Levy <david.l...@restonrobotics.org> wrote:

> I'd like to see how the kids interpret it on their own without clouding
> their judgement.

Absolutely. I look forward to hear from them on the data.

> The thought  was enough sunlight would register from an upward
> angle that they would not need to worry about payload spin.
>
> The team did start to question that design when they learned that the  
> launch took place at sunrise ( ballon#2 went up first) and also mistakenly
> assumed that their flight duration would  be longer then what was
> experienced.

I was surprised at the flight duration as well - Gypsy's script for
ascent was cut significantly short, as these missions were shorter
than the average. But honestly I think that's just part of the "game",
if you will - you can predict, and plan, but in the end there are
always things you can't control.

With respect to the low sun angle during the flight, one thing the
team could do is try to replicate the experiment (now that they know
more about what *did* happen) on the ground. Do some measurements at
very low sun angles, and then tilt the payload so it's facing the sun.
Do this in the early morning, noon, and afternoon, so that you have a
"calibration": a reading of X under flight conditions would have been
a reading of Y if the sensor had been facing the sun, for instance.
That won't solve the problem (the sensor may simply have gotten too
little UV, period)but it might at least help understand what happened.

> It may be a few weeks before I can get the team together to analyze  
> the data.  However it may be a good exercise to gather a few team  
> members to extract the data from the returned payload and send it to  
> you in a raw format while you are awaiting the team's final  
> determination.

Well, if they want to do it... I, of course, would love to see it :)

--
Brian Davis

Eric

unread,
Aug 4, 2008, 4:14:44 PM8/4/08
to HALE TEAMS
I didn't calculated the ascent rate - did anyone do that yet?

Brian Davis

unread,
Aug 4, 2008, 5:28:34 PM8/4/08
to HALE TEAMS
On Aug 4, 4:14 pm, Eric <lego.profes...@gmail.com> wrote:

> I didn't calculated the ascent rate - did anyone do that yet?

I've go the Balloon #2 data (the one with Gypsy) up. It reached
99,687' at MET 76.4 minutes, for an ascent rate of a shade over 1,200
feet/minute, overall.

--
Brian Davis

David Levy

unread,
Aug 18, 2008, 10:32:10 AM8/18/08
to hale-...@googlegroups.com

We'll we finally had a chance to meet as a team again. The kids
uploaded the three files form the NXT onto their PC's and began to copy
the contents into Excel spreadsheets. They split into two groups -one
to examine the UV filter data and one to look at the thermostat data.

The thermostat data was basically a file with a single row that
represented the Celsius reading taken once every 20 seconds. In order
to graph the data over time, a team member chose to write a simple NXT
program to create a file with the timestamp data ( i.e. 20, 40, 60,
80, 100) so it could be laid out in a column next the the temperature
readings.

For more on the thermostat and the uploaded chart, see Jeremy's
discussion here:
http://groups.google.com/group/hale-teams/browse_thread/thread/8663ee7ec3d6433c#

The UV data analysis was a little more complicated. The raw data
consisted of two files - a timestamp file with a single column written
in seconds and of course the filter file containing a single column
with the UV readings. Both files were copied into respective columns on
a spreadsheet. I gave the kids an excel function for column b:
"=(a1 / 60 )" so they could express their time in minutes instead of
seconds.

For more on the UV analysis and the uploaded chart, see Arianna and
Pooja's discussion here:
http://groups.google.com/group/hale-teams/browse_thread/thread/561199323fcde12e#

Please feel free to add your analysis to the data as you see fit. I'm
sure the kids will welcome it and publish your findings as well.

David Levy
FLL Team 90 Payload Coach

Brian Davis

unread,
Aug 18, 2008, 10:53:07 AM8/18/08
to HALE TEAMS
On Aug 18, 10:32 am, David Levy <david.l...@restonrobotics.org> wrote:

> Please feel free to add your analysis to the data as you see fit.  I'm
> sure the kids will welcome it and publish your findings as well.

I did - and these "kids" are doing a really great, adult-style attempt
and figuring out the data! Please let us know how the visit to the
Smithsonian goes!

--
Brian Davis
Reply all
Reply to author
Forward
0 new messages