'Lil Joe's Parachute almost-failure
1 view
Skip to first unread message
Peter Wallhead
Aug 16, 2008, 4:56:32 AM8/16/08
to HALE TEAMS
This message is mainly directed at Brian, but feel free to jump in
also.
Brian do you know the course of the near parachute failure on 'Lil Joe
yet? I read that the parachute became tangled on the webbing (?), any
updates on this?
Would using a separate drogue 'chute, not just your fin/tail/cap thing
have helped with the parachute deployment?
Also, what material did you use for the parachute construction?
Cheers,
Peter Wallhead.
also.
Brian do you know the course of the near parachute failure on 'Lil Joe
yet? I read that the parachute became tangled on the webbing (?), any
updates on this?
Would using a separate drogue 'chute, not just your fin/tail/cap thing
have helped with the parachute deployment?
Also, what material did you use for the parachute construction?
Cheers,
Peter Wallhead.
Brian Davis
Aug 16, 2008, 4:22:08 PM8/16/08
to HALE TEAMS
On Aug 16, 4:56 am, Peter Wallhead <Peter.Wallh...@gmail.com> wrote:
> Brian do you know the [cause?] of the near parachute failure on 'Lil Joe
what evidence we have. The ground team found the parachute tangled in
the shroud lines and the tether to the tailfin cap:
http://www.unr.edu/nevadasat/balloonsats/missions/HALE/recovery.html#6
(there are a whole series of pictures here - the ground crew team did
a *fantastic* job of documenting this!)
If you look at that picture carefully, the thin yellow twisted
(actually "unwound") string leading to the left is attached to the
tailfin cap, and then leads through the twisted section to attached to
a screw-gate oval attached to the webbing loop that is at the middle
top of the main parachute. I'd originally had this *not* attached
there, but attached to the bottom of the 'chute, to prevent tangling.
But the ground team thought this was unlikely to work, and reattached
it to the top of the 'chute. That certainly deployed it, but the then
heavier-than-the-'chute tailfin probably hung down beneath it,
partially or completely collapsing one side of the 'chute, and the off-
center lift on the 'chute caused it to spin, twisting up the tailfin
tether and the main tether like kids on a playground swingset, winding
it up tightly.
Doesn't that sound reasonable? Logical, coherent, and certain to be
true? The problem is it's all guesswork, based on what I *think* might
have happened and a few pictures. My son jokingly maintains it hit a
bird in mid-flight, and honestly there's no way to disprove even that
outrageous hypothesis - there's just too little data.
> Would using a separate drogue 'chute, not just your fin/tail/cap thing
> have helped with the parachute deployment?
Perhaps - but then you have the problem of how to deploy the drogue
(and I wanted a "true freefall" - nothing hanging out during the free-
fall portion, so a pre-deployed drogue was considered "unfair" for
Lil' Joe). You could still get some issues with a drogue - that could
tangle as well, and it looks like initially there's so little air that
these sort of issues can be a problem.
> Also, what material did you use for the parachute construction?
Eric's team supplied me with the parachute, from a commercial
manufacturer (for very large model rockets). It's a rip-stop nylon,
and handled everything just fine. There is no sign of any issue with
the parachute itself.
There are some very curious clues that I'm not sure what to make of
yet, and a few observations...
1) Some damage to the bottom of the tailfin cap *might* indicate it
bounced against something hard several times... actually, it's a type
a damage that post-flight testing has shown takes something hard (a
hammer) repeatedly striking the foam - in other words it is NOT
landing damage (the blue foam is remarkably good at handling impacts).
This could indicate a partial "predeployment" or separation of the
tailfin, followed by it being smashed repeatedly against the LEGO
positioning fingers internal to the shell... but again, it's hard to
say. Another possibility is the metal screw-gate oval being slammed
against it during the descent (likely, given some of the photos). As a
comparision you can see where the LEGO "cord-lock" was twisted and
snapped into the foam on the way down, and it shows significantly less
damage.
2) Actual damage to the main payload shell is remarkably absent. At
first this made explaining the "ejection" of the internals through the
bottom (or at least to break out the bottom) really hard to
understand. If the shell hit the ground and was stopped while the
internal mass was still traveling forward (to punch out the bottom as
seen in the photos), then I expected damage to a bottom edge of the
shell - and couldn't find it. This was answered when I tried some
"impact studies" - taking pieces of foam and whipping them at the
ground as hard as possible to simulate a 30+ mph impact. Even on
concrete or rocks, there is almost *no* damage to the foam - after a
dozen impacts, the edges start to look a little rounded, that's it. So
it seems the "bottom blowout" needs nothing more to explain it than
the final impact with the payload shell.
3) Every sticker on the tailfin is missing (there were four, two self-
adhesive, two held on with nothing more the Scotch tape). There were
two on the payload shell, both self-adhesive, that remained on,
although they only remained attached to the shiny metalized tape on
the payload shell itself (one straddled the boundary of tape/foam, and
the portion on the foam is completely loose). Note I put exactly the
same stickers on Gypsy as on Lil' Joe, attached to the same surfaces
with the same mechanisms, so I have a way to compare - all the
stickers on Gypsy are still there. Clearly Lil' Joe suffered a more
traumatic descent than Gypsy and the main payload stack.
4) While there is very little damage to the main payload shell, there
is one very curious thing I've been unable to reproduce. On the side
under the "black instructions" (a two step activation checklist drawn
on the side of Lil' Joe by the ground team), there is a very odd
"hyper-smooth" area. The entire outside of Lil' Joe was shaped with a
hot-wire cutter, so it's fairly smooth to the touch (there are now
plenty of small nicks and dings in it, at least one with small pieces
of plant embedded in it, but nothing big). But this one patch is
smoother than fresh-cut foam - a very glassy or glossy texture, very
obvious to the fingers. Nor is this part flat: it is very roughly
rectangular, with both the top and bottom segments slightly depressed
relative to the (still glassy smooth) section between them. I can't
rule out that this was something that happened during shipping, but I
can not replicate the texture due to foam-on-foam friction or
compression. Could this be due to rubbing from one of the tethers or
webbing? Well, perhaps... but it's on the lower half of the payload
shell, while most of the tether system should have been streaming out
of the upper half, nowhere near the payload shell (and instead, the
impact damage noted on the tailfin is not seen on the main shell).
I've no idea what this could be - it's a true mystery to me.
5) We've verified experimentally, by building a mock-up of the payload
internals and a smashing it into a mock-up of the bottom of the
payload shell, that the three-leaved blow-out seen on the bottom of
Lil' Joe is exactly the pattern that would be expected. The neighbors
wondered why I was making a lot of little pieces of foam out of
carefully constructed big pieces of foam, but hey, that's science :).
6) There is some unexplained damage to the *inside* of the payload
shell, way down deep adjacent to the SPOT unit. But I'm not sure
exactly what the shape of that surface was before launch either (while
the outside was smoothed, the inside was rougher surfaced, as I just
needed to get everything in). It looks like something sharp (LEGO axle
sized) repeatedly hit it over a large area. There are indeed two two
LEGO axles sticking towards that surface... but only at one point. For
the damage seen to have been caused by these, the internals would have
had to shift deeper in the shell (through the bottom) by at least four
inches... and from the materials recovered on-site (like the black
cushioning foam that was in no way tied into the other materials), the
bottom blow-out must have occurred at ground level.
Note: I've been trying to get every bit of information out of the Lil'
Joe payload first, as I need to mail that one out to LEGO HQ in
Billund, so I've not done this detailed a post-mortem on Gypsy. I
*have* tried to do some measurements on Lil' Joe just by throwing it
into the air in violent spins to see what the accelerometer records...
and it doesn't look like I can solve the mystery that way either. But
it does indicate it's a really violent ride down (in other words, by
violently throwing Lil' Joe into the air in chaotic spins, I can't get
accelerometer readings as violent as it experienced after the first 10
seconds of free-fall). Interesting...
--
Brian "next time I'll write more" Davis
> Brian do you know the [cause?] of the near parachute failure on 'Lil Joe
> yet? I read that the parachute became tangled on the webbing (?), any
> updates on this?
I'd love to know what happened - but it's hard to reconstruct from
> updates on this?
what evidence we have. The ground team found the parachute tangled in
the shroud lines and the tether to the tailfin cap:
http://www.unr.edu/nevadasat/balloonsats/missions/HALE/recovery.html#6
(there are a whole series of pictures here - the ground crew team did
a *fantastic* job of documenting this!)
If you look at that picture carefully, the thin yellow twisted
(actually "unwound") string leading to the left is attached to the
tailfin cap, and then leads through the twisted section to attached to
a screw-gate oval attached to the webbing loop that is at the middle
top of the main parachute. I'd originally had this *not* attached
there, but attached to the bottom of the 'chute, to prevent tangling.
But the ground team thought this was unlikely to work, and reattached
it to the top of the 'chute. That certainly deployed it, but the then
heavier-than-the-'chute tailfin probably hung down beneath it,
partially or completely collapsing one side of the 'chute, and the off-
center lift on the 'chute caused it to spin, twisting up the tailfin
tether and the main tether like kids on a playground swingset, winding
it up tightly.
Doesn't that sound reasonable? Logical, coherent, and certain to be
true? The problem is it's all guesswork, based on what I *think* might
have happened and a few pictures. My son jokingly maintains it hit a
bird in mid-flight, and honestly there's no way to disprove even that
outrageous hypothesis - there's just too little data.
> Would using a separate drogue 'chute, not just your fin/tail/cap thing
> have helped with the parachute deployment?
(and I wanted a "true freefall" - nothing hanging out during the free-
fall portion, so a pre-deployed drogue was considered "unfair" for
Lil' Joe). You could still get some issues with a drogue - that could
tangle as well, and it looks like initially there's so little air that
these sort of issues can be a problem.
> Also, what material did you use for the parachute construction?
manufacturer (for very large model rockets). It's a rip-stop nylon,
and handled everything just fine. There is no sign of any issue with
the parachute itself.
There are some very curious clues that I'm not sure what to make of
yet, and a few observations...
1) Some damage to the bottom of the tailfin cap *might* indicate it
bounced against something hard several times... actually, it's a type
a damage that post-flight testing has shown takes something hard (a
hammer) repeatedly striking the foam - in other words it is NOT
landing damage (the blue foam is remarkably good at handling impacts).
This could indicate a partial "predeployment" or separation of the
tailfin, followed by it being smashed repeatedly against the LEGO
positioning fingers internal to the shell... but again, it's hard to
say. Another possibility is the metal screw-gate oval being slammed
against it during the descent (likely, given some of the photos). As a
comparision you can see where the LEGO "cord-lock" was twisted and
snapped into the foam on the way down, and it shows significantly less
damage.
2) Actual damage to the main payload shell is remarkably absent. At
first this made explaining the "ejection" of the internals through the
bottom (or at least to break out the bottom) really hard to
understand. If the shell hit the ground and was stopped while the
internal mass was still traveling forward (to punch out the bottom as
seen in the photos), then I expected damage to a bottom edge of the
shell - and couldn't find it. This was answered when I tried some
"impact studies" - taking pieces of foam and whipping them at the
ground as hard as possible to simulate a 30+ mph impact. Even on
concrete or rocks, there is almost *no* damage to the foam - after a
dozen impacts, the edges start to look a little rounded, that's it. So
it seems the "bottom blowout" needs nothing more to explain it than
the final impact with the payload shell.
3) Every sticker on the tailfin is missing (there were four, two self-
adhesive, two held on with nothing more the Scotch tape). There were
two on the payload shell, both self-adhesive, that remained on,
although they only remained attached to the shiny metalized tape on
the payload shell itself (one straddled the boundary of tape/foam, and
the portion on the foam is completely loose). Note I put exactly the
same stickers on Gypsy as on Lil' Joe, attached to the same surfaces
with the same mechanisms, so I have a way to compare - all the
stickers on Gypsy are still there. Clearly Lil' Joe suffered a more
traumatic descent than Gypsy and the main payload stack.
4) While there is very little damage to the main payload shell, there
is one very curious thing I've been unable to reproduce. On the side
under the "black instructions" (a two step activation checklist drawn
on the side of Lil' Joe by the ground team), there is a very odd
"hyper-smooth" area. The entire outside of Lil' Joe was shaped with a
hot-wire cutter, so it's fairly smooth to the touch (there are now
plenty of small nicks and dings in it, at least one with small pieces
of plant embedded in it, but nothing big). But this one patch is
smoother than fresh-cut foam - a very glassy or glossy texture, very
obvious to the fingers. Nor is this part flat: it is very roughly
rectangular, with both the top and bottom segments slightly depressed
relative to the (still glassy smooth) section between them. I can't
rule out that this was something that happened during shipping, but I
can not replicate the texture due to foam-on-foam friction or
compression. Could this be due to rubbing from one of the tethers or
webbing? Well, perhaps... but it's on the lower half of the payload
shell, while most of the tether system should have been streaming out
of the upper half, nowhere near the payload shell (and instead, the
impact damage noted on the tailfin is not seen on the main shell).
I've no idea what this could be - it's a true mystery to me.
5) We've verified experimentally, by building a mock-up of the payload
internals and a smashing it into a mock-up of the bottom of the
payload shell, that the three-leaved blow-out seen on the bottom of
Lil' Joe is exactly the pattern that would be expected. The neighbors
wondered why I was making a lot of little pieces of foam out of
carefully constructed big pieces of foam, but hey, that's science :).
6) There is some unexplained damage to the *inside* of the payload
shell, way down deep adjacent to the SPOT unit. But I'm not sure
exactly what the shape of that surface was before launch either (while
the outside was smoothed, the inside was rougher surfaced, as I just
needed to get everything in). It looks like something sharp (LEGO axle
sized) repeatedly hit it over a large area. There are indeed two two
LEGO axles sticking towards that surface... but only at one point. For
the damage seen to have been caused by these, the internals would have
had to shift deeper in the shell (through the bottom) by at least four
inches... and from the materials recovered on-site (like the black
cushioning foam that was in no way tied into the other materials), the
bottom blow-out must have occurred at ground level.
Note: I've been trying to get every bit of information out of the Lil'
Joe payload first, as I need to mail that one out to LEGO HQ in
Billund, so I've not done this detailed a post-mortem on Gypsy. I
*have* tried to do some measurements on Lil' Joe just by throwing it
into the air in violent spins to see what the accelerometer records...
and it doesn't look like I can solve the mystery that way either. But
it does indicate it's a really violent ride down (in other words, by
violently throwing Lil' Joe into the air in chaotic spins, I can't get
accelerometer readings as violent as it experienced after the first 10
seconds of free-fall). Interesting...
--
Brian "next time I'll write more" Davis
Peter Wallhead
Aug 16, 2008, 6:50:21 PM8/16/08
to HALE TEAMS
Brian,
Thanks for your (very thorougher) reply to my query's regarding Lil'
Joe's parachute. It will take a while to process it all, but I'm sure
I can come up with some more questions for you and maybe some
answers :)
Please excuse the spelling mistakes in the original message, I was
having one of those days.
Cheers,
Peter Wallhead.
Thanks for your (very thorougher) reply to my query's regarding Lil'
Joe's parachute. It will take a while to process it all, but I'm sure
I can come up with some more questions for you and maybe some
answers :)
Please excuse the spelling mistakes in the original message, I was
having one of those days.
Cheers,
Peter Wallhead.
Reply all
Reply to author
Forward
0 new messages