0
460

The science of canopy openings: Limited Subterminal Configuration Comparison

Recommended Posts

Here is a result of a discussion Walt Appel and I had with Tim Harris. Tim's work is groundbreaking IMHO. Tim did a frame-by-frame analysis of jumps at Bridge Day and analyzed the data. However, as always, these results should be taken with a grain-of-salt. I shattered my femur several years ago while speculating that an experimental gear design would work based on a simple (and imcomplete) analysis.

On another note, I am aware of technology that would allow position sensors to be installed on a canopy in a way that would not affect the parachute opening. No one has data on how a parachute really opens.


Name: Tim Harris
E-Mail: **********************
Subject: Mr. ******* (BASE 460) & Mr. Appel
Body of Message:

Mr. ******* (BASE 460) & Mr. Appel

I'm not sure that my opening paper still exists in electronic form. I can tell you essentially what it said. It was titled "Limited Subterminal Configuration Comparison".

The paper contained about 200 data points. Each data point was from a single jump. I measured duration of freefall, duration of deployment, heading of opening, number of cells and aspect ratio, and some other information. I also determined the basic rig configuration, such as bagged vs. free-packed & slider up vs. slider down/off, from the video of the jumps. The idea was to look for correlation of opening performance to rig configuration. Reducing this data from video tape was about as much fun as having teeth pulled.

The data presented in the paper showed that sliders slow down openings, and greater airspeed (longer freefall delay) speeds up openings, as one would expect. In the 1 to 4 second freefall range, lack of slider made the opening 5 to 10 times faster. There was plenty of variablity, or "scatter", in the data which is why the results were inexact. Lots of variables involved.... The data also showed that bagged canopies opened slightly slower than free-packed for the same delay. The bagged vs. free-packed difference, however, was less then the scatter within any given configuration, essentially "down in the noise."

What I didn't put into the original paper was the heading information. The data showed that heading of opening correlated more strongly to duration of deployment than did aspect ratio. In fact, off-heading openings correlated more strongly to short freefalls and their resulting slow openings than to any other variable that I measured. The "lore" back then and still to some degree today is that higher aspect ratio (A/R) canopies open more off heading. Off-heading openings seem to be less a result of high A/R than they are of short delays, after a first scientific glance. Body position is another factor to consider, and the exit point video that I used doesn't lend itself to determining if shoulders are level. Does body position effect heading? Experienced jumpers say yes overwhelmingly, so it probably does.

With so many variables involved, it becomes difficult or impossible to separate out individual effects or to determine their relative order of effect on opening performance. This is why it is important to listen to the opinions of experienced jumpers. This is also why it is important to question and scientifically test those opinions whenever possible. For a long time blood letting was an accepted and commonplace medical practice. It also doesn't work, and was responsible for fatalities in some extreme cases. Get it?

I was jumping high aspect ratio canopies at Bridge Day, and many experienced jumpers told me that it was bad BASE gear and it would open off-heading. I used that gear because it opened extremely fast, and I was more concerned with getting a canopy open quickly than I was over the heading of opening. My 13 jumps on that gear in all four basic rig configurations opened within about 10 degrees of my body axis on 11 jumps, and about 90 degrees off on 2 jumps. The experienced jumpers were only partially correct; the heading was not bad. It seems other things can have more of an effect on heading than just aspect ratio. 13 jumps is not enough to bet a life on, and I would never jump that stuff off of a building.

It WAS bad BASE gear, however, because it had a flat glide slope and a fairly high (read "fast") wing loading at or above 1.0 lbs/sq.ft. This would make it dangerous when opening near an object and dangerous to land in a tight and/or rough LZ. If you don't believe this, see Will Forshay's Lemmings Bloopers video. What good is hitting the sand bar if you have to run it out into the water? I wasn't concerned about landing. I didn't care about getting wet if necessary, and Bridge Day wasn't any tougher than the power off landings I had to make to get my helicopter instructor rating. I do realize that stuff like that will tend to catch up with a person. I don't think I'll jump that gear any more from fixed objects. Maybe at bridge day with calm low water. Then again, there's always those rappelling ropes....

I had good packing help/experience and very good body position on those jumps, which may have overcome ill effects of high A/R, if they really exist. Those same experienced jumpers, however, also recommend a solid 3 second delay for a "better" opening. Coincidence? I think not. Remember, slower openings tend more to go off heading. It pays to listen to experience. If you have an urge not to heed the voice of experience, however, I would recommend many skydives on a new piece of gear or pack job before BASE jumping it. Those jumps should be recorded on video and/or measured in every way possible to gather maximum data if one is trying something new. Detailed log records and video of pack jobs are also wise.

Another item of interest was the effect of bagging vs. free packing on heading of opening. It looked like bagged canopies on average didn't open off-heading more often than free-packed. The occasional bagged canopy off-heading opening would tend to be farther off-heading, however, than the average free-packed off-heading opening. To trend this more solidly, thousands to data points should be used, not the few hundred that I used. If this trend does bear out with a larger sample size, it may be an indication that the mechanism causing off-heading openings may be different between free-packed and bagged deployments. This is no surprise, since methods by which the canopy is exposed to the air and brought to line stretch are totally different in bagged vs. free-packed deployments.

Sleeves and pods are an entirely different issue here, and were not widely used when I first collected data.

Parachute opening performance is a complex phenomenon. There are thousands of variables involved. So far the human brain has been the best computer for determining what helps and what hurts deployment characteristics. The risk there is that emotions, health & other factors often cloud memory and judgement. Although the "feeling" an experienced jumper has regarding what is good or bad is very important, it is not always correct or absolute. That is why non-bias measurement and formal scientific analysis are also important.

The disketts w/ my original subterminal comparison paper were heat damaged when I left them in the car while travelling for work. I've been looking for another electronic copy in my stored files, which is not a simple task. I do not have a computer of my own and my files are in all different formats from all of my friends/relatives' computers that I have used over the years. I'm on full time travel for work, which complicates the issue. I am fairly sure the original database that I used still exists. It was compiled from frame-by-frame analysis of about 200 jumps on a Hi-8 format Bridge Day video shot in 1992.

I have recently served notice to leave my current job, and I need to revisit this subterminal deployment research if I get some time. Unfortunately BASE jumping research won't pay the bills very well. Since the initial effort in '92 I have reviewed other video to increase the size of the database. My friend and fellow scientist James W. O'Reilly has also started reducing jump video into useable format for the same effort. Validity increases with sample size. The more data the better it represents reality.

The data entry process is very time consuming. Each video frame is one thirtieth of a second, and each jump last for several seconds, often a few hundred frames. Although many frames may be glossed over during the freefall portion of the jump, other frames during the exit and especially the deployment must be meticulously scrutinized, often many times repeatedly, for proper determinations to be made. I have posted some of these review techniques before. Ideally the work could be spread out over many people, however this increases the chance of inconsistent data review/entry techniques corrupting the data. At the very least, all parties participating in video analysis should stay in contact with each other. This will reduce duplication of effort as well as eliminating the chance of repeat data points finding their way into the same data base. The video analysis techniques must be standardized, and that is the main reason I should find that paper. It outlined the technique used.

The most subjective part of the analysis was the definition and subsequent determination of "open" for slider up deployments. Perhaps the best definition to use is "slider half way down". Determining this using video from directly above the opening, however, is challenging to say the least. It must often be guessed by rewinding the video from slider all the way down (full size canopy) to the partially constricted state of inflation that "looks like" the slider is half way down.

For anyone interested in hard core formal parachute performance modeling and analysis, I suggest you check out the Aerodynamic Decelerator Systems Technology Conference at the following web address: http://www.engr.uconn.edu/~adstc/
These are the folks that land parachutes on other planets. They have aerospace money to do research, which is loads more than the sport parachuting industry can afford to fund, let alone the limited BASE community. Some of the systems, however, are very similar. There is a lot of ram air deployment research presented at the ADSTC. I'm planning on going to the June conference and seminar, for "fun". I'd love to show a BASE video there, although many of the science purests there would frown on such activity. There aren't many skydivers there, let alone fixed object types.

Maybe some day we will actually see BASE gear and techniques used to evacuate tall burning buildings, as it was originally conceived to do back in the days of Da Vinci. The more commonplace and wide spread our activity becomes, the better the chance of funding research and further improving the technology.

Tim Harris
March 99

Link: Aerodynamic Decelerators Systems Technology
Added on Date: 07:43:56 3/11/99
http://www.engr.uconn.edu/~adstc/
Looks like a death sandwich without the bread - Steve Deadman Morrell, BASE 174

Share this post


Link to post
Share on other sites
0