0
PhreeZone

PIA Report on low reserve canopy openings

Recommended Posts

http://www.pia.com/images/rocketlauncher/PDF/TECHNICAL_REPORTS/PIA-TR-401LowReserveOpeningInvestigationReport91316.pdf

Report is out, there is some data in it but with names redacted its hard to make many new decisions moving forward if one wanted to change their gear to get something with better drag from the reserve PC.
Yesterday is history
And tomorrow is a mystery

Parachutemanuals.com

Share this post


Link to post
Share on other sites
Id' be really curious to know which one is (A) and only has a pulling force of 18 lbs for AAD-fire situations, compared to all others having 46 to 78. And if that could be a cause for problems (maybe it's a container with a completely different reserve tray design, like racers, with the two bottom pins, so the results have to be very different and can't be compared?).

All other numbers across 7 manufacturers seem close enough to not make a difference under *most* situations.
I'm standing on the edge
With a vision in my head
My body screams release me
My dreams they must be fed... You're in flight.

Share this post


Link to post
Share on other sites
Interesting. I assume that this was created in part to address John Sherman's criticism of Bill Booth's PC design?

Observations/Questions:
1. If I called up a manufacturer, could they tell me what letter they are, or is there a NDA on the data? (I assume the manf were told what letter they were).

2. The sample sizes were small, and there was a lot of variation, but there are some clear differences. The highest sustained drag was about 1.6x the lowest, not as much as the 3x John mentioned in a post a few years ago http://www.dropzone.com/cgi-bin/forum/gforum.cgi?post=4286259#4286259

3. Also interesting is the fact the terminal AAD fired deployments drag measurements were lower than the hand-deployed terminal deployments. This was true for all manufacturers. I assume the hand deploy is tipping the body and getting the pc in the airstream faster (no burble).

4. Would be really good to have documented the time for deployment too, and alt loss, at least to bag extraction or line extension, but with just a load cell for testing that would be challenging.

Seth
It's flare not flair, brakes not breaks, bridle not bridal, "could NOT care less" not "could care less".

Share this post


Link to post
Share on other sites
The ripcord deployments are pulling a ripcord, launching the PC without allowing the bag to be extracted. The bridle was attached to a three ring at the yolk a few inches from the bag. The peak force is the spike on the load cell as the PC hit the end of the bridle attached to the three ring that towed the PC until release. The sustained force is the PC being towed at terminal.

The AAD force is essential the extraction force of the free bag. These are lower because the PC is not towed. A manually activated cutter cut the loop and initiated normal reserve deployment. The PC never had to exert it's maximum force to extract the bags.

Note that only 7 manufacturers participated. Not all of the most popular rigs are represented. All were invited. Some declined and some never responded. The manufacturers got the data for their own gear and agreed not to use it for marketing. No, I don't know which letter is which rig.

In watching the video opening times were not significantly different and no bag was delayed in extraction. The load cell was in line in the bridle and transmitting wirelessly.

But YMMV depending on your combination of components, rigging, attitude, speed and a multitude of factors.
I'm old for my age.
Terry Urban
D-8631
FAA DPRE

Share this post


Link to post
Share on other sites
Terry. thank you for those explanations. Did the author of this report not have time to write up some explanations of the data? According to your explanation, the "AAD Fire" line in the chart does not seem to be what many people would think it is.

I am puzzled by the findings described in the report, that is, "... the Committee Members have yet to locate evidence that supports or indicates a systemic or specific equipment design issue.", considering this video: https://www.youtube.com/watch?v=vaYQ6iP8zlg

Share this post


Link to post
Share on other sites
SethInMI

Interesting. I assume that this was created in part to address John Sherman's criticism of Bill Booth's PC design?



3. Also interesting is the fact the terminal AAD fired deployments drag measurements were lower than the hand-deployed terminal deployments. This was true for all manufacturers. I assume the hand deploy is tipping the body and getting the pc in the airstream faster (no burble).


Seth




There is no 'hand deploy'. The non aad data, either at terminal or sub terminal started with the pull of the reserve ripcord. The pilot chute launched and reached the end of the bridle. The bridle was attached to the yolk of the rig with a three ring release a few inches (maybe a foot) from the bag. The forces are the peak when the PC hit the end of the bridle and the constant force as it was towed in stable free fall. Then the bridle was released and the PC allowed to extract the bag. All deployments were video taped. The load cell transmitted wirelessly and was small so didn't really effect the function much at all. The AAD deployments were made flat and stable with a manually activated cutter. The peak is the peak measured by the load cell as the PC deployed the reserve as it would normally for a total malfunction/AAD fire. No delays in bag launch were observed. All jumps were made by the same jumper using the same techniques and body position.

And no this was about PC design, this was about why are people bouncing after AAD fires? Are bags being held up by tight rigs and are PC's drag strong enough to deploy bag. For the seven rigs submitted (all were invited) no issues were seen although differences in towed PC force were observed. Don't try to guess which was which. You won't get it right.;)
I'm old for my age.
Terry Urban
D-8631
FAA DPRE

Share this post


Link to post
Share on other sites
Thanks for the reply. When I said "hand deployed" I meant using your hand to start deployment by pulling the ripcord, not the clearest choice of words.

I agree I totally misread the chart and mentally doubled the number of jumps to split them into standard deployment (for peaks) and 10+ sec drougefall deployments.

So to clarify, all the non-AAD initiated launches had two steps:
1. pull the ripcord to launch PC
2. pull the "drogue release" to initiate deployment after waiting the 10+ seconds.
It's flare not flair, brakes not breaks, bridle not bridal, "could NOT care less" not "could care less".

Share this post


Link to post
Share on other sites
peek

Terry. thank you for those explanations. Did the author of this report not have time to write up some explanations of the data? According to your explanation, the "AAD Fire" line in the chart does not seem to be what many people would think it is.

I am puzzled by the findings described in the report, that is, "... the Committee Members have yet to locate evidence that supports or indicates a systemic or specific equipment design issue.", considering this video: https://www.youtube.com/watch?v=vaYQ6iP8zlg



"A systemic or specific equipment design issue" The goal here was with a properly sized container for the reserve used (container chosen by manuf. for the specific reserve make/size used for all tests) do we see bag extraction force requirements higher than PC drag forces. In reality the prime data is just how much drag do these PC's create in real life freefall behind a jumpers burble. The towed PC data gave us that. The single AAD launch for each gave us one extraction force measurement. Data is limited due to cost of program.

That doesn't mean that any particular rig (which may not have been tested) with an over stuffed reserve container and an over stuffed main container and a loop 3 inches too long may not have issues launching as seen in the referenced video.
I'm old for my age.
Terry Urban
D-8631
FAA DPRE

Share this post


Link to post
Share on other sites
councilman24

The AAD deployments were made flat and stable with a manually activated cutter... And no this was about PC design, this was about why are people bouncing after AAD fires? Are bags being held up by tight rigs and are PC's drag strong enough to deploy bag. For the seven rigs submitted (all were invited) no issues were seen although differences in towed PC force were observed. Don't try to guess which was which. You won't get it right.;)



These tests don't really make sense to me. The manufacturers provided containers with optimally sized reserves, and all deployments were in the optimal body position with the reserve pulled straight up and out of the container. This same test would have been performed by each manufacturer thousands of times already, so it's not exactly surprising that no problems were found. Am I missing some crucial detail here?

Primary Root Cause D test wasn't done, so you have no idea what the maximum bag extraction force is for each container. That's the number that needed to be compared with the minimum sustained subterminal PC force. As it stands, the tests show that all of these containers are able to deploy their reserves under very specific (pretty much ideal) conditions. [:/]

Share this post


Link to post
Share on other sites
Only one manufacturer (that I know of) had the pull force data for their PC in real free fall behind a jumper. They were the ones who had put together the wireless load cell and data system. And I'm not sure they had used it in this way before. I know some others had wind tunnel data. What's new is all of the numbers. And this is the first time we have bag extraction forces. Especially with an AAD deployment. And if rigs are tested thousands of times its by customers trying to save their life.

These tests took two years and several 10s of thousands of dollars. Remeber PIA is a volunteer organization made up of competitors. Thats why the data is blinded and participants had to agree not to use the data in advertising.

There has been doubt by some that even in optimum circumstances rigs would perform perform well. This is a beginning. This is data that most manufacturers didn't have for their own rig. This is public data on container/PC performance that didn't exist before. This data took hundreds of volunteer hours and lots of PIA dollars. What's sad is the manufacturers that chose not to participate.

Okay, say thank you now.:P
I'm old for my age.
Terry Urban
D-8631
FAA DPRE

Share this post


Link to post
Share on other sites
The thanks go to Dave Singer of Perigrine and PIA. As a member of the technical committee I was around for some of the planning and saw the video and data in the spring.

Lots of folks think that PIA had lots of resources and ability to take on tasks. In reality we meet twice a year, have one part time employee for administrative functions and limited resources. All of the activities PIA does is done by industry volunteers. In this case PIA funded modification of the rigs for PC towing and the test jumps with video and data recording. Dave put in lots and lots of hours in analyzing and compiling the data. This was partly in response to in adequate information learned when he did wind tunnel testing for his then employer. The committee realized that the only useful data was from a PC in freefall behind a jumper with all of the real life conditions that imposed.

This seems like an easy task but with people administering it in their free time and the requirement for all of the testing to be done by one jumper it takes time, effort and money.
I'm old for my age.
Terry Urban
D-8631
FAA DPRE

Share this post


Link to post
Share on other sites
bhop

So does the AAD load in the appendix equate to the force needed to extract the bag? If that was the load measured on the load cell during an AAD cut I'd almost want the 18lb instead of a higher one if that's all the pilot chute had to pull.



I need to review things again myself but essentially yes. The jumper accelerated to teriminal and then fired the AAD cutter with a manual button. The force reported is the peak reading the load cell in the bridle recorded. This is a dynamic system with momentum of the still bag (relative to the rig) the PC slowing down relative to the rig, and the PC having to over come the momentum of the bag when it reaches the end of the bridle.

Also this is one reseve/ rig size combination with one rigger. YMMV. This was done to at least have one data point other than goround extraction measurements and to compare to the towed force measurements of the PC. The answer is they all worked.
I'm old for my age.
Terry Urban
D-8631
FAA DPRE

Share this post


Link to post
Share on other sites
I thought it interesting how much lower the Sustained loads were compared to the Peak loads at terminal velocity. And even the Sustained loads are low for terminal velocity compared to figures published for pilot chutes only.

The key difference is that these tests are with the pilot chute stretched out on a reserve bridle behind a jumper, with all of the burble effect.

It takes time to dig up numbers but here are a few:

-- A Jumpshack publication on calculated drag of some pilot chutes , based on tests: At "terminal velocity" forces were 145-180 lbs for Racer PC's (big & small), the MA-1, and Vector I PC. (And Vector II more like 75). The numbers are of course just from one company with a particular viewpoint but better than nothing.
[Source: I think the jpg I had of these values is from one of John Sherman's videos from about 2011 ?? when he was quite active in the pilot chute debate]

-- An Australian Parachute Federation study from the early 1980s ?? tested pilot chutes with a scale outside of a testing vehicle at up to 120mph. Loads of 98 to 146 lbs were seen for some spring loaded pilot chutes of the era (except with the Vector II at 84 lbs).

In contrast, the new PIA Sustained values 85 to 124 lbs for Average Peak force at an unknown terminal velocity. That's a good chunk less than those tests mentioned above. And that's after the pilot chute hits the end of the bridle with a bunch of momentum.

The Average Sustained at terminal was just 34 to 57. On a rig by rig basis, the sustained values were thus often just 45% to 55% (very roughly) of the Peak forces.

Sometimes the Average Sustained forces were not much higher than in the subterminal cases. Although since they were averaging over 10 seconds... the "subterminal" case isn't all that subterminal any more after that long. I'd want to see some speed info to judge whether the small gap between terminal and subterminal forces were really a surprise or not.

The 34 to 57 is far lower than the typically 100 to 150 or more values in the other tests.

That's a huge loss. Since the data is from completely different tests, the reasons for the difference aren't clear.

But it does suggest that the burble effect is larger than many might expect. I suspect the average dynamic pressure far back in the burble isn't hugely less, but if a pilot chute is being batted around in turbulence it could be spilling air and not working efficiently. Who knows. And that's not just 5 or 6 ft away, but at the end of a perhaps 12 or 15 ft long reserve bridle.

But the forces should still be enough. Lets say there's a tight pack and it does take 18 lbs to extract the freebag from the rig (a number at the top end of what John Sherman suggested as being acceptable). Forces of 85+ peak forces will extract that.

And even a sustained force of 34+ lbs would then accelerate a say 7 lb medium sized reserve away at 34/7 = 4.8g acceleration or more.

Anyway, all this is just some playing with rough numbers and quick speculation. Any additional data on spring loaded pilot chute drag values would be welcome.

Share this post


Link to post
Share on other sites
Yes good point! As with the subterminal case, that long 10 seconds thing, and no speed data in the report we see, makes things messy. So both "subterminal" and "terminal" aren't necessarily what we might expect them to be throughout each test.

And who knows how much body position changed as well. Just like in the case of someone with a bag lock, body position can affect the result a lot.

Edit: I expect the people involved in doing the tests weren't exactly dummies, but the info that got passed down to us is a little limited in telling us what really happened...

Share this post


Link to post
Share on other sites
The body position was essential flat and stable, slightly head high when towing. Jumper did not look back or roll shoulders (as observed in the videos several months ago).

Without reviewing the data in the report in detail the PC that was 'dancing' in the burble produced higher sustained force but more noise in the data. Those that were stable and moving very little seemed to be lower in force but smoother data.
I'm old for my age.
Terry Urban
D-8631
FAA DPRE

Share this post


Link to post
Share on other sites
We are concerned with people bouncing after AAD's fire. How about duplicating the situations with test dummies or real people jumping the actual blood stained equipment, rather than looking at test specs from manufacturers? At this rate, USPA will want us to be opening at 10k.

Share this post


Link to post
Share on other sites
gb1

We are concerned with people bouncing after AAD's fire. How about duplicating the situations with test dummies or real people jumping the actual blood stained equipment, rather than looking at test specs from manufacturers? At this rate, USPA will want us to be opening at 10k.



.................................................................................

Big thank you to Dave Singer and his assistants within PIA for providing base-line data. This base-line data will eventually lead to more drop-tests to answer more complex questions posed on this thread.

As for replicating rests with "bounced" rigs ..... the most blatant malfunctions can be replicated on the ground at slow deployment speeds. For example, the problems created by ridiculously long reserve closing loops was well understood a decade ago. Eric Fradet even sent me a diagram of a "worst case" long loop.
As for replicating "stupid tight" container/canopy combinations ...... that is a well-understood problem.

Share this post


Link to post
Share on other sites
Been discussed here a couple times, including this thread:

http://www.dropzone.com/cgi-bin/forum/gforum.cgi?post=2101957

tl;dr is: in AAD cutters installed below the PC, a too-long, unlubricated loop can fully lock the reserve tray in play in the event of AAD fire. The mechanism of it being that the overlong loop snakes through the flaps rather than going straight, and the force of the PC spring coil pins it between flaps. Since it's AAD fire, the reserve closing pin remains in place, completing the lockup.
"Skydivers are highly emotional people. They get all excited about their magical black box full of mysterious life saving forces."

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account. It's free!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
0