Jul 23, 2008

Lil' Joe, T-7 days...

While I'm still working refining the payloads with the folks who will be launching the HALE mission, I thought I'd at least put up a few more facts and pictures about them here in the blog. Of my two payloads, Lil' Joe is perhaps the most ambitious... or riskiest. Looking somewhat like a blue loaf of bread with a tailfin, the goal is to replicate, in a very small (& cheap!) way, the famous leap of Col. Joseph Kittinger from over 100,000'. The NXT will try a much more modest goal: being released from roughly 80,000', it will free-fall for a short period of time (something like 20 seconds to a minute or more - the exact length of time is under discussion) before "reeling out" the tailfin, which then functions (I hope!) as a drogue 'chute to pull the main parachute out of the body. There will be no control from the ground, or even communication, with the payload during this process - either the NXT "does it on it's own", or the Nevada desert gets a new crater lined with blue foam and very small pieces of ABS.

The command and control for Lil' Joe is fairly simple: when it is turned on, the NXT-G program starts writing timestamped 3-axis accelerometer data to a datalog as fast as it can. Every five seconds it checks a "free-fall flag", and if that flag has not been set it erases the data & starts again. This way, the final datalog should contain a few seconds of data from just before the payload is cut free, extending through the free-fall and the parachute deployment beyond... even though the NXT doesn't know in advance when it is being released. A second sequence running in parallel with the first is responsible for detecting free-fall and deploying the parachute after that. It simply hangs in a loop, waiting for at least 0.5 seconds of near free-fall (less than 0.25 G's) acceleration, and once it detects that sets the "free-fall flag", as well as starting a count down for the parachute deployment. At the moment of deployment this sequence pauses datalogging briefly, runs the motor to unwind the tether & deploy the parachute, and then resumes datalogging (datalogging is paused because a poorly-timed write to the flash memory can mess up the internal clock and motor control, disasterous if the motor in question is releasing your only parachute).

Just how fast and how far Lil' Joe will fall is poorly known: a 20 second free-fall with no friction at all can produce a velocity of 440 mph after a fall of more than a vertical mile. But even with as thin as the air is up there, there will be some friction, & as the payload plummets deeper in the atmosphere and the atmosphere gets thicker aerodynamic effects will limit it more and more. So it's really an unknown right now. The good news is that with the size of the payload bottom and the size of the parachute (24"), it looks like parachute deceleration might be reasonable at under 10 G's for a couple of seconds... but that means the heavy portions of the payload (batteries) need to be very firmly cradled by the parachute lines directly. The team at the University of Nevada at Reno are working with me on perhaps adding some things to the payload (such as a convection baffle for after the tailfin comes off, and perhaps stronger webbing).

There are more pictures of Lil' Joe up in my Brickshelf gallery:

HALE Photos on Brickshelf

As well as a short YouTube video of Lil' Joe. This is really aimed at explaining to the launch team how it is unpacked and packed, but it also shows (near the end of the video) what the internal mechanism does during free-fall. Hopefully this wasn't the last time I ever get to video it :).

Lil' Joe preparation on YouTube



--
Brian Davis

4 comments:

Jim Kelly said...

Brian,

As someone who used to jump out of airplanes (my wife won't let me anymore - some silly thing about insurance not paying up) I can tell you this - when that chute opens, you better really trust in the nylon connecting the chute to your pack.

What were your thoughts on the quick deceleration and how to keep the chute connected to the payload? At the speeds your possibly looking at, my concern isn't that the drag chute pulls the main, but that the main doesn't pop open so fast that the force it exerts rips away the connection...

Anyway - looking forward to hearing about the success - I'm very positive about this and I wish you and the other teams much luck and good will.

Jim

Brian Davis said...

Jim wrote:

> What were your thoughts on the quick
> deceleration and how to keep the chute
> connected to the payload?

The idea is that payload is *not* tied to the parachute from above it, but that the main tether from the parachute (in the pictures, a multi-stranded cord) are screw-gated onto a "sling" that the SPOT/NXT sit within. If you try to catch the skydiver by having his wrists tied to the shroud, you'll yank his arms off... so you use straps under the thighs. Lil' Joe uses the same system. Wil it be enough? We'll see... but yes, I've thought about it and I'm concerned as well.

The other thing to do is try to calculate what the actual accelerations are as the parachute opens. I'm not sure what they are for humans (anybody have hard facts?), but from simple modeling of the Lil' Joe payload, they don't appear to exceed 10 G's, which isn't bad (OK, it's be really bad and painful for you or me, but is no problem at all for an NXT or SPOT supported from below... I hope).

--
Brian Davis

Anonymous said...

This is so exciting Brian!!!!

Goodluck Lil' Joe
LEGOmom

Parax said...

Brian, What is triggering Lil Joes Plumment? is it Altitiude or Time? and is the relase mechanism Lego too?

Good luck!

Related Posts Plugin for WordPress, Blogger...