Official Statement Regarding Colorado Vigil Firings

[billvon 2,428]

>Another point is that if the plane reaches 1500ft, the Cypres also
>becomes active and if the plane descends to the activation zone,
>the Cypres will also activate if the pressure variation measured is fast
>enough.

Definitely true, which is why it's a good idea to turn off a cypres before landing with the plane. However, it seems to be quite resistant to firing in response to transient pressure events, as the World Team incident demonstrated - and as opening doors at between 1000 and 1500 has never, to my knowledge, caused a cypres misfire, despite the popularity of doing that.

[brettski74 0]

Quote

This talk of "decision time" is not an accurate representation of what's going on. When the Cypres's firing parameters are met, it fires within milliseconds. When the Vigil's firing parameters are met, it fires within milliseconds. There is no delay when either unit is thinking (in effect) "I know I have to fire now, but I'm going to wait a while to close the cutter switch."

I'm not suggesting that this is what is happening. I'm not talking about the decision to activate. I think that might be where the confusion comes from.

To reference the incident that prompted all of this, the device was in an aircraft that was not descending - it registered a zero descent rate just prior to the pressure spike. In Vigil's responses they mention that the device detected the freefall in 0.625 (5/8) seconds. The firing parameters of the device in Pro mode require a descent rate of 35 m/s and an altitude between 840' and 150' (body position corrections notwithstanding). From a a descent rate of zero, it will take over 3.5 seconds under gravitational acceleration alone to reach 35 m/s descent rate - and that's assuming no drag. Once you add in drag, it will take even longer. Even if you consider the lower activation descent rate for Student mode of 20 m/s, that's still going to take over 2 seconds with no drag and even more with drag. I simple don't see the necessity to detect freefall in only 0.625 seconds.

The bigger picture from all of this is that the device should have some idea of what is realistically possible in freefall and be able to rule out impossible situations to prevent misfires at inopportune times - particularly when inside an aircraft. I will grant that determining this based on barometric pressure readings alone may not be as simple as I may have initially made it out to be, but a device that can think at one moment that it's at or near zero descent rate and less than a second later decide that it's in freefall at a descent rate in excess of 35 m/s doesn't make a whole lot of sense either.

[brettski74 0]

Quote

If I am reading his argument correctly, he is saying that the unit should take into consideration the current speed of the skydiver towards earth, and the maximum possible acceleration that skydiver could achieve due to gravity, and IF the skydiver somehow accelerated with ludicrous speed towards earth, not fire because impossible acceleration has been achieved, and thus the speed readings are inaccurate...

Mostly correct. No acceleration is impossible, but the device is designed to save skydivers in freefall, which means that they are accelerating under the influence of Earth's gravity alone, so it should give some consideration to what is possible under such circumstances.

Quote

I think he is disputing that a skydiver can accelerator from under canopy to firing parameter speed in less than 5/8 of a second.

Almost correct, but again missing some of the details. It is impossible for a skydiver can accelerate from under canopy with a specific range of descent rates to a descent rate exceeding 35 m/s in less than 5/8 of a second. This is a fairly well accepted law according to Newtonian physics.

Quote

I wonder however, after having video of a spinning mal where I looked at my altimeter twice, thus know my decent rate to be well over 50 MPH

For the record, 50mph is about 22.2 m/s. Under the influence of Earth's gravity only and ignoring drag, it would take about 1.3 seconds to reach a descent rate of 35 m/s. In the real world, it would take longer than that. For you to accelerate to a descent rate of 35 m/s with no drag in a time of 0.625 seconds, your initial descent rate would need to be about 28.9 m/s, which is around 65mph. Again, these calculations ignore the effects of drag, so in the real world, your initial descent rate would have to be higher than this.

There's no reason why the device can't have the benefit of the pressure and other derived data while you're under your high speed malfunction and plummeting to the earth at around 50mph. If you look at the graphs downloaded from the devices that misfired, they actually show consistently negative descent rates from about -7 seconds to -3.5 seconds as well as descent rates at or close to zero for another 2 seconds after that and prior to the beginning of the pressure spike. This suggests to me that the device is still making pressure measurements and altitude/descent rate calculations prior to detecting freefall and that in the event of an activation, it is logged somewhere therefore it must have been available to the device at the time of activation also.

Now, you could say that due to pressure variations induced by body position, we can't be sure that the descent rate of 50mph is accurate or calculated based on pressure changes from other causes. We can say, however, that virtually all skydives start from aircraft that are in relatively stable flight and therefore at descent rates at or close to zero. This being the case, there is no need for the device to detect that I'm in freefall in such a short space of time as 0.625 seconds as there is no way I can reach the activation speed in anything close to this time under the influence of Earth's gravity alone. The only plausible exceptions to this stable flight assumption that I can think of are stunts or aircraft emergencies, however, I'm not sure that I'd want an AAD firing while I'm in or attached to an aircraft in a nosedive, anyway.

[dorbie 0]

Quote

Quote

Conclusion: In those 2 cases the Vigils activated because the firing parameters were met.

And, IMO, it is a bad design if it can fire due to only the door opening.

And if it saves you from a low chop or low bailout then it's a brilliant design that saved your ass. As the document says there are tradeoffs, no set of choices can produce the perfect AAD.

You yourself can sit down with a piece of paper and specify what YOUR ideal AAD activation altitude would be, the descent rate and altitude required and any other heuristics. So come up with them and share with us all.

Simply saying it's a bad design because one aircraft ride had a pressure variations that met it's firing criteria is a bit of a stretch. Vigil could change the microcode & 'detune' it in a heartbeat if they thought it made sense, it's not a decision they make lightly.

[Ron 7]

Quote

And if it saves you from a low chop or low bailout then it's a brilliant design that saved your ass.

Firing because the door opened is not the same as a low chop or a bailout.

Low chop... You were in freefall
Bailout... You are in freefall.
Door opening could result in losing a whole plane full of people.

Quote

You yourself can sit down with a piece of paper and specify what YOUR ideal AAD activation altitude would be, the descent rate and altitude required and any other heuristics. So come up with them and share with us all.

Why? I am not in the market to make an AAD and there are units that already work fine with parameters that not only sound good, but have been proven over the years.

I don't have to create a new wheel design to make the statement that a square wheel is not a good idea.

Quote

Simply saying it's a bad design because one aircraft ride had a pressure variations that met it's firing criteria is a bit of a stretch.

1. It has happened more than once.
2. If you can't see how a reserve being fired just due to the door opening is a danger... You didn't pay attention in your FJC.

"No free man shall ever be debarred the use of arms." -- Thomas Jefferson, Thomas Jefferson Papers, 334

[sundevil777 94]

Quote

Quote

Quote

Conclusion: In those 2 cases the Vigils activated because the firing parameters were met.

And, IMO, it is a bad design if it can fire due to only the door opening.

And if it saves you from a low chop or low bailout then it's a brilliant design that saved your ass. As the document says there are tradeoffs,

The vigil claims to have some advantage in acting (detecting freefall) faster. I think it is marketing hype or worse - a characteristic that has proven to be dangerous with no advantage. As others have said in this thread, AADs don't pause once they've detected their firing criteria. Please explain to us how the Vigil acts in a way that is an advantage, and I have yet to have it explained how it would behave any differently than a cypres in a low cutaway scenario. The claim is there, but it doesn't seem to stand up to scrutiny.

Quote

no set of choices can produce the perfect AAD.

You yourself can sit down with a piece of paper and specify what YOUR ideal AAD activation altitude would be, the descent rate and altitude required and any other heuristics. So come up with them and share with us all.

The folks that designed the Vigil had a clear target to beat. They didn't even match their competitor in such an important characteristic as being able to recognize when the data is inconsistent with skydiver physics.

Quote

You yourself can sit down with a piece of paper and specify what YOUR ideal AAD activation altitude would be, the descent rate and altitude required and any other heuristics. So come up with them and share with us all.

Simply saying it's a bad design because one aircraft ride had a pressure variations that met it's firing criteria is a bit of a stretch. Vigil could change the microcode & 'detune' it in a heartbeat if they thought it made sense, it's not a decision they make lightly.

It isn't a matter of "detuning" it. It is a matter of making it smarter.

I really do wish there was better competition out there to drive down prices and push the quality up.

People are sick and tired of being told that ordinary and decent people are fed up in this country with being sick and tired. I’m certainly not, and I’m sick and tired of being told that I am

[dorbie 0]

Quote

Quote

Quote

Quote

Conclusion: In those 2 cases the Vigils activated because the firing parameters were met.

And, IMO, it is a bad design if it can fire due to only the door opening.

And if it saves you from a low chop or low bailout then it's a brilliant design that saved your ass. As the document says there are tradeoffs,

The vigil claims to have some advantage in acting (detecting freefall) faster. I think it is marketing hype or worse - a characteristic that has proven to be dangerous with no advantage. As others have said in this thread, AADs don't pause once they've detected their firing criteria. Please explain to us how the Vigil acts in a way that is an advantage, and I have yet to have it explained how it would behave any differently than a cypres in a low cutaway scenario. The claim is there, but it doesn't seem to stand up to scrutiny.

Quote

no set of choices can produce the perfect AAD.

You yourself can sit down with a piece of paper and specify what YOUR ideal AAD activation altitude would be, the descent rate and altitude required and any other heuristics. So come up with them and share with us all.

The folks that designed the Vigil had a clear target to beat. They didn't even match their competitor in such an important characteristic as being able to recognize when the data is inconsistent with skydiver physics.

Quote

You yourself can sit down with a piece of paper and specify what YOUR ideal AAD activation altitude would be, the descent rate and altitude required and any other heuristics. So come up with them and share with us all.

Simply saying it's a bad design because one aircraft ride had a pressure variations that met it's firing criteria is a bit of a stretch. Vigil could change the microcode & 'detune' it in a heartbeat if they thought it made sense, it's not a decision they make lightly.

It isn't a matter of "detuning" it. It is a matter of making it smarter.

I really do wish there was better competition out there to drive down prices and push the quality up.

Making it smarter is an easy thing to say, and it is not difficult to make it more complex, but there is a philosophy and I think probably a wise one to keep the heuristics simple in these devices.

As for the claimed advantages, they are not the same as their competition, parameters and philosophies vary in nuanced ways that produce a different outcome. I'm not here to justify or argue for Vigil algorithm, I jump a cypres2 but if you cannot discern a scenario where arming altitude differences can be of benefit I don't think anyone can move you past your anti-Vigil rhetoric.

[dorbie 0]

Quote

Quote

And if it saves you from a low chop or low bailout then it's a brilliant design that saved your ass.

Firing because the door opened is not the same as a low chop or a bailout.

But it can look the same to an AAD.

Descending quickly in an aircraft is not the same as freefall but a student cypres can fire under those circumstances.

Diving a canopy is not a chop, but someone died because their AAD thought it was and this had been anticipated for years. A decision was made that it wasn't enough of a risk factor to compromise safety on a low chop (my interpretation of the decision to do nothing about it) (edit to add , or any borderline mid-speed mal).

AAD heuristics are tradeoffs between firing when you want it to fire and not firing when you don't. There is no device that can give you perfect decision making, it is not possible.

[sundevil777 94]

Quote

if you cannot discern a scenario where arming altitude differences can be of benefit I don't think anyone can move you past your anti-Vigil rhetoric.

OK, so to the point of arming altitude differences. A vigil arms itself considerably lower than a cypres. For the scenario of someone getting out below 1400 and not pulling, the Vigil will save you. That scenario clearly goes to Vigil. Not a very significant scenario though.

I don't want to bash the Vigil. It is just that their product deserves it. I hope they improve, I want there to be good competition.

People are sick and tired of being told that ordinary and decent people are fed up in this country with being sick and tired. I’m certainly not, and I’m sick and tired of being told that I am

[Ron 7]

Quote

But it can look the same to an AAD.

It did not look the same to a CYPRES. So, it can also NOT look the same.

To ME when two devices are put into the same *false* situation... The one that reacts incorrectly has a problem.

Quote

Diving a canopy is not a chop, but someone died because their AAD thought it was and this had been anticipated for years.

Yep and that ALSO was a bad thing. It has now been addressed. Unless you think that it was a bad thing they invented the swoop CYPRES?

Quote

AAD heuristics are tradeoffs between firing when you want it to fire and not firing when you don't. There is no device that can give you perfect decision making, it is not possible.

If you think an AAD firing due to a door opening is fine.... Please don't get on the same plane as me.

"No free man shall ever be debarred the use of arms." -- Thomas Jefferson, Thomas Jefferson Papers, 334

[billvon 2,428]

>As the document says there are tradeoffs, no set of choices can
>produce the perfect AAD.

Right - but there are certainly choices that produce better AAD's and worse AAD's. If all AAD's were poor compromises, then we'd all be jumping FXC's and Sentinels. (They're a lot cheaper.)

>Simply saying it's a bad design because one aircraft ride had a pressure
>variations that met it's firing criteria is a bit of a stretch.

There have been a great many. Four due to open doors, over a dozen due to transient pressurization incidents.

[dorbie 0]

Quote

>As the document says there are tradeoffs, no set of choices can
>produce the perfect AAD.

Right - but there are certainly choices that produce better AAD's and worse AAD's. If all AAD's were poor compromises, then we'd all be jumping FXC's and Sentinels. (They're a lot cheaper.)

>Simply saying it's a bad design because one aircraft ride had a pressure
>variations that met it's firing criteria is a bit of a stretch.

There have been a great many. Four due to open doors, over a dozen due to transient pressurization incidents.

The transient pressurizations are irrelevant to most jumpers and outside the intended design envelope IMHO, it's seriously misleading to bring them up in this context. The door oppenings are the primary concern.

As I said the thing could be 'detuned' or reparameterized (and it has nothing to do with a smarter device) based on new door opening pressure data, but the decision should be data driven and Vigil is conducting those experiments. They're building a data logger that doesn't require firing, says so right there in their document.

I'd have thought just sticking a sensor and logger in a box on board a several AC rather than one rig would be the way to collect a lot of data fast and augment with in rig data on the same runs.

Hmmmm... an engineering company conducts tests and data gathering after a statistically unlikely outcome is uncovered in an attempt to improve their product. This is an OUTRAGE! Puh-lease.

[billvon 2,428]

>The transient pressurizations are irrelevant to most jumpers

Many jumpers, yes. However, I'd be willing to bet that most active jumpers will find themselves inside a pressurizable aircraft at some point in their skydiving careers. (C-130 for example is a popular boogie aircraft)

>and outside the intended design envelope.

Agreed. Some jumpers may decide to use an AAD that has a wider design envelope.

>Hmmmm... an engineering company conducts tests and data
>gathering after a statistically unlikely outcome is uncovered in an attempt
>to improve their product. This is an OUTRAGE! Puh-lease.

??? I think it's a great idea, and hopefully it will lead to them being able to fix the current problems with the Vigil.

[dorbie 0]

Quote

??? I think it's a great idea, and hopefully it will lead to them being able to fix the current problems with the Vigil.

Every AAD has problems. There is no such thing as a perfect AAD.

[skydiverek 56]

Quote

Quote

??? I think it's a great idea, and hopefully it will lead to them being able to fix the current problems with the Vigil.

Every AAD has problems. There is no such thing as a perfect AAD.

Brand "A" can have one two problems, brand "B" can have ten "problems". Quantify. Saying that both brands have "problems" removes the basis for comparison (quantification).

[sundevil777 94]

Quote

Quote

??? I think it's a great idea, and hopefully it will lead to them being able to fix the current problems with the Vigil.

Every AAD has problems. There is no such thing as a perfect AAD.

What a gem of insight that is.

No perfect AADs, yes we can agree on that.

What we apparently don't agree on is the extent of how unintelligent (not smart) the vigil is. The folks designing it had a clear target (the cypres) to do better than. They have clearly missed the mark. The vigil doesn't need to be "detuned" to not have fired when a door opens, it needs to be smarter.

People are sick and tired of being told that ordinary and decent people are fed up in this country with being sick and tired. I’m certainly not, and I’m sick and tired of being told that I am

[billvon 2,428]

>There is no such thing as a perfect AAD.

Agreed. But some come closer than others.

[Remster 24]

Quote

The folks designing it had a clear target (the cypres) to do better than

That's where the comparison fails. I dont know if (nor do I think that) they wanted to so better than Airtec. You don't need to make a product technically better to compete with another. Pricing, total cost of ownership, ease of use, distribution channels, etc are other factors that can come in.

Looks like in the end, you do get what you pay for.

Remster

[dorbie 0]

Quote

Quote

Quote

??? I think it's a great idea, and hopefully it will lead to them being able to fix the current problems with the Vigil.

Every AAD has problems. There is no such thing as a perfect AAD.

What a gem of insight that is.

No perfect AADs, yes we can agree on that.

What we apparently don't agree on is the extent of how unintelligent (not smart) the vigil is. The folks designing it had a clear target (the cypres) to do better than. They have clearly missed the mark. The vigil doesn't need to be "detuned" to not have fired when a door opens, it needs to be smarter.

Instead of talking vague nonsense about "smartness" propose an algorithm. Simply adjusting a few factors in the Vigil and keeping the algorithm generally the same could have prevented these firings, but it IS a tradeoff, it is not "smarter", no matter how much ill informed nonsense you post about it needing to be smarter. There are also scenarios where it performs well that some are bending over backwards to ignore.

You could try to make it smarter, but that's potentially fraught with more pitfalls than you realize or anyone can easily even quantify. It's just a bad strategy for a device like this.

Vigil is understandably cautious about making adjustments, as I have said you want this kind of alteration to be data/evidence driven as any competent engineer will tell you.

There is a lack of data and this is a rare event, if these were popping off all the time you might take a different approach but under the circumstances the right thing to do is appreciate the saves Vigil has under it's belt vs. risks associated with jumping with it (no AAD is without such risks), as they collect the data to inform any required adjustment.

If Cypres was out after the Vigil it would have been banned SOMEWHERE because of it's activation altitude and maybe its disarming deck, and all the loons would have been bashing Cypres for years. Because these are AADs everyone assumes that performance envelopes on newer products should be the same in every respect but better (in unspecified ways) than the incumbent. But it is rarely how real innovation happens. The Vigil will be better in some scenarios and the Cypres in others. In this case the Vigil after years of use & many saves is showing signs taht it may be a bit twitchy under extremely rare circumstances, and this is rare despite FUD that has tried to lump in decompression related firings.

And FYI I jump a cypres 2 and wouldn't change it, I don't have a dog in this hunt.

[dorbie 0]

Quote

>There is no such thing as a perfect AAD.

Agreed. But some come closer than others.

And that determination depends on scenarios and priorities.

Your abject reply linked below told me all I need to know about your ability to offer a reasoned and credible response on non Cypres AAD related matters:

http://www.dropzone.com/cgi-bin/forum/gforum.cgi?post=3825061;page=unread#unread

AADs are useful and save lives, even imperfect ones.

[piisfish 135]

Quote

AADs are useful and save lives, even imperfect ones.

true, my life is imperfect and has been saved by an AAD

scissors beat paper, paper beat rock, rock beat wingsuit - KarlM

[brettski74 0]

Quote

Instead of talking vague nonsense about "smartness" propose an algorithm. Simply adjusting a few factors in the Vigil and keeping the algorithm generally the same could have prevented these firings, but it IS a tradeoff, it is not "smarter", no matter how much ill informed nonsense you post about it needing to be smarter. There are also scenarios where it performs well that some are bending over backwards to ignore.

Several people, myself included, have already posted in this thread that the device should have some concept of what is possible in freefall due to the Earth's gravity. Without doing experimental research, nobody here is going to be able to give you the complete and final algorithm, but it starts with looking at second differences in the altitudes deduced from barometric pressure readings. The device's design is flawed if it is programmed to fire less than 1 second after it believed it was in level flight when even on the most paranoid setting (which it wasn't) it would take at least twice that long to reach firing speed, and on the Pro setting (which it was) more than 3 times that long.

Quote

You could try to make it smarter, but that's potentially fraught with more pitfalls than you realize or anyone can easily even quantify. It's just a bad strategy for a device like this.

It's more important for the device to avoid firing when it shouldn't than it is for the device to fire when it should. Whether you like that reasoning or not, statistically, the vast majority of skydivers will never need their AAD to save their life. That being the case, it's difficult to accept a device that might endanger my life by firing when it shouldn't. It's a bad business strategy to do nothing about this.

Quote

Vigil is understandably cautious about making adjustments, as I have said you want this kind of alteration to be data/evidence driven as any competent engineer will tell you.

Agreed. I am an engineer.

Quote

If Cypres was out after the Vigil it would have been banned SOMEWHERE because of it's activation altitude and maybe its disarming deck

That's a bit of a reach. How do these parameters endanger anyone that is not already in danger?

Quote

all the loons would have been bashing Cypres for years.

I was not a skydiver when CYPRES was first released, but I was under the impression that many people did bash the CYPRES for some time before it started to prove itself in the field.

Quote

Because these are AADs everyone assumes that performance envelopes on newer products should be the same in every respect but better

I for one, don't, so not everybody. I do however, expect that misfires are treated seriously. I may be mistaken, but all of the CYPRES misfires that I can recall were down to either a faulty device and led to a recall, or the device operated as designed and changes were made, such as releasing speed CYPRES. Vigil's attitude until this most recent incident seems to have been "The device is the best. It did exactly what it was designed to do. The end." I'm encouraged by their response to this event so far, but that doesn't mean I don't expect improvements to come from it. Their competition has been shown in the past to be able to determine that non-freefall events did not require activation where Vigil did. This proves that they could have done better. As a current Vigil customer, I'd like to see that they do.

Quote

And FYI I jump a cypres 2 and wouldn't change it, I don't have a dog in this hunt.

I jump a Vigil 2. For someone who has no dog in this hunt, you seem very passionate about it.

[billvon 2,428]

>And that determination depends on scenarios and priorities.

Of course. In transient-pressure-change scenarios, Vigils tend to fire more often than other AAD's. If that's OK, the Vigil could be a good choice for someone.

My priority is not having my reserve deployed unless I absolutely, definitely need it. Which is why I use a different AAD. You may have different priorities.

[pchapman 262]

It was brought to my attention that the Vigil II manual now states:

Do not open the door of the plane during the flight in the activation zone to avoid a
possible pressure variation, which could result in an unexpected activation.

There was no fanfare about this change that I'm aware of.

This is in the current manual that was published some time in June 2010, v II.0.3. Earlier versions didn't have that. (At least not II.0.1 which had been around a while -- I guess the "II" refers to the Vigil II, not revision 2, so this new manual is in essence v 0.3)

The "activation zone" would I suppose be at least the range from 150 ft (arming) to 840 ft + 260 ft = 1100 ft to include the belly vs. non-belly pressure adjustment.

In Student mode, 1040+260= 1300 ft max.
If Tandems are on board, no door open until 2040 + 260 = 2300 ft max.

Technically this new rule could prohibit, say, sliding open an Otter door after seat belts are off when students or tandems are on board.

In practice though, I figure it is more a precaution for C-182's and the like, with a smaller cabin volume and doors that can go from well-sealed to popped open very quickly. Perhaps 'cracking' the door open for a moment while holding the handle might be better than just letting the door fly right up.

For intentional door openings, the altitude range is of most concern if there's a tandem on board, and one were thinking of throwing a wind drift indicator or dropping a hop & pop.

For unintentional door openings, there are obviously still concerns about rare but not impossible AAD firings.

[sundevil777 94]

Totally lame.

They might as well just admit their product is stupid.

For those willing to put up with a stupid AAD, the price of used Vigils at least should go down, if not even the new units. Demand goes down...

People are sick and tired of being told that ordinary and decent people are fed up in this country with being sick and tired. I’m certainly not, and I’m sick and tired of being told that I am