No matter how you cut it, you're fitting an arbitrary curve to two untested, assumed, pseudo data points.Originally Posted by Gary
Welcome to the Cave Diver's Forum.
No matter how you cut it, you're fitting an arbitrary curve to two untested, assumed, pseudo data points.
Andrew Ainslie
Almost extinct cave diver
What empirical data are you using to come up with the 50% turn gives you a 50% chance of survival? If I were to test this, say at little river, I'm willing to bet in 10 tries I would make it out 10 times without going to a stage, breaking your chart.
My problem with those numbers is that they don't account for ANY factors. In a low flow/no flow/siphon system they would skew high on the fatality per event side, and in a high flow cave they would skew low. If I swam in without a light or guideline and followed those numbers would my chances change? yes. The options are endless and without any baseline runs or testing it's really just a pretty graph. You have a great idea, just need to get all the factors and facets down. The fact that you're thinking about the correlation between deaths and turn pressure shows you are looking for an answer to "How Much Is Enough". We all want to know the answer, but I'm sure the answer changes each time we enter the water.
You might want to pick up this quarter's NACD Journal, They did specific gas management tests from back in 02? and 2 out of 3 tests they "died". Testing is possible, prudent and might save some lives. Guessing won't unless you can prove the outcome.
I was taught to be a thinking diver, and I do like to consider what my options are before I develop a proper gas management plan for a given dive. I don't base my survival on an arbitrary psi, but cubes required to exit the cave with 2 major malfunctions. Oddly enough that's always been more conservative than thirds for most dives. If it's just a short swim dive with a buddy I've got plenty at thirds due to my lower RMV than all my buddies. I normally have turn pressure at the safety stop.
Really? If it takes you 100psi per 100ft, you're going to get 1650ft in, and only have enough gas to swim 1350ft. That's simply not possible without extreme flow, possibly even more than mainline @ Devil's system.
How so? The majority of cave dives I take less gas to get out than to go in. Unless you have something to slow your exit when diving halves 1 in 2 times, this stat doesn't add up. We're not planning gas in best case scenario, but statistics need to take it into account. I guarantee the odds of making it out of an average tourist cave like Madison, Ginnie, Peacock, etc with well maintained gear are better than 1 in 100 and maybe even 1 in 1000 when diving halves. That's not good enough odds to bet your life on, but your saying that you have a better chance playing Russian roulette with a 6 shot revolver than diving 1/2's at Ginnie, which simply isn't true at all.
How can depth not effect this answer? Someone with a .65 sac at Wakulla is going to use about 784psi per minute, meaning he can't even support a buddy for 30 seconds with that extra 100psi. The same diver at Peacock is going to use 207psi per minute, and can support a buddy for almost 4-5x as long.
Either we have a huge misunderstanding of what you're trying to convey or one of us is terrible at statistics :-/
Last edited by jj1987; 10-07-2009 at 03:14 PM.
It was specified as "no flow" (flow changes the starting and ending points to unknowns.)
If you swim in to half you swim out the same gas. Your efficiency will be more, less or equal with about equal chance of being better or worse. If better - live. If worse - die.
50%
The assumptions are simple enough to be very safe but even if the chances were significantly off (say 1 in 10 instead of 1 in 2) it would have very little effect of the calculation starting at aprox 1 in 10000 and dividing over 400-500 psi.
4th root (10000/10) = 5.62
4th root (10000/2) = 8.41
5th root (10000/2) = 5.49
And though the starting point may change based on flow, gear, etc there is little reason to assume the factor changes overmuch.
JJ: are you saying there is no chance a person can swim out more efficiently than they swam in? Is there no variability in human performance? That swimming to half they can swim out successfully every time but 1 psi past half they have no chance whatsoever?
Clearly that isn't the case. Even if they swam in until they hit 1000psi they'd have some chance of swimming out on half what they swam in on - even if the chance is immeasurably small. They may have had something slow them down going in - but not on the way out. And on the other side they may not be able to swim out on the same that they went in on. They could get a good run in and have a number of delays getting out.
No flow, I can still beat it every time. You use more gas going in because of the drysuit and/or wing inflation going in. I know if I'm marginal on gas I'm going into conserve and I guarantee I'll make it out on less than half the gas. I still say you need to do a study. Test it!
There's going to be a point where you have a zero survival rate within those 2-400 psi. These tests will only prove that x amount of times a diver will swim out to a certain footage and return with x amount of gas or none. The amount will skew with depth as well because of wing and drysuit inflation. It has to be tested at a steady depth in a controlled environment with several different divers of different competency before the number will be viable.
The idea is sound but the differences between individuals will wreck your theory's curve. From this test you can begin to overlay situational data points such as flow, vis, etc. There are several people here with a much better understanding of how to calculate the flow and consumption rates at odds with the rate of travel. This can be done if given the proper amount of data in the proper environment.
I'm saying that there's better than 50/50 odds of making it out alive if you dive halves. I'm guessing the chance is more like 999/1000 or at the very least 99/100. Sure you can pickup speed on the exit, we've all proven that by diving Ginine, Manatee, JB, Maidson, etc.....even Peacock we often exit faster than we enter with as light of flow as there is there, especially when you factor in lighter stage bottles and such.
But since you've acknowledged that, that further proves that the odds of diving halves and getting away with it are a LOT better than 50/50. The higher flow the system, the better chance you have here...but your statistics didn't factor in flow, either in how it helps speed your exit, or how it slows your entrance (and increases sac).
Also, could you comment on why 100psi is a 5x safety factor in both 20ft mexico caves, as well as 300ft deep WKPP or west FL caves?
I think you're onto a key component here, James.
Some issues slow you down by a percentage of your normal exit time. Losing a fin, or bad vis, are examples of this. For such events, reserving a MULTIPLE of your gas makes sense - for example, reserving double the amount used in to exit (i.e. thirds). Depth doesn't enter the equation.
For other issues, like looking for a buddy, briefly losing the line, running a jumpreel across a line entrapment in a silt out, or an entanglement, one needs to assume a fixed time period to sort the problem out. You might decide to allow 5 minutes for one of these events. Here, the gas reserve DOES depend on depth, and one whould have to increase the amount as one got deeper, based on SAC rate, depth, time required and tank size (we can all do the calculation from there).
For this reason, thirds may be insufficient at (say) 300 ft to handle even one entanglement, given how quickly we use our gas at depth, whereas it may be enough to handle a whole bunch of them in 20 ft caves.
Andrew Ainslie
Almost extinct cave diver
Exactly. Let's say you're in a 300ft deep cave and swim the wrong way, 100ft until you see a line arrow that corrects you. That 100psi barely helps, since barely a minute of gas, where as in peacock, you could easily swim 100ft and back on 100psi (about 5 minutes IIRC from when I calculated it earlier).
I really wish more majors required discrete math courses. They really force you to think about "hidden" variables that can really change the T/F value of your statement.
For example-
Today I was talking to a customer at work, he ships 1in x 1in items (replacement parts for compressors), so I programmed his shopping cart to add the items end-to-end, thinking this would give the smallest possible box size, so that UPS's online tools api would give the best estimate on shipping as possible when you combine boxes. However, if you order 100 items that are 1in x 1in, my original solution would result in a package that's 1in x 1in x 100in, which due to the length, is a MUCH higher shipping cost.
The correct answer is to add 1in at a time to each dimension to the box, and check to see if you can fit the items, and if not, add 1in to the next dimension, continuing this until everything fits. This results in shipping 100 items in a 10in x 10in x 10in box, which actually winds up costing the same as a 1in x 1in box, other than the weight difference, and literally less than half of what it costs to ship a 1in x 1in x 100in item.
Even this fails. You then need to check to see if the items fit in a flat rate box, and compare the shipping cost for the smallest possible box, and the flat rate box, and choose the cheaper of the two, since it varies depending on location.
But wait, even that fails. You need to estimate the cost of cardboard since flat rate boxes are free, and make sure that the shipping price for the best sized box doesn't exceed the cost of flat rate when you include the cost of packing materials.
Point being - it's nearly impossible to think of all cases in a simple linear formula. It took over a year for us to find this bug in the software I wrote, but after finding it, you feel stupid that you even thought it would work in the first case. I think Gary's idea might work with a small number of test cases if it started with accurate data and appropriate limitations (maybe 60-90ft deep caves with moderate flow, without stages, without scooters, with a buddy), but there's just not enough input to expect it to be even remotely accurate as it was presented.
To answer the original question, I dive my stages differently depending on the dive and have fun doing it....
Safe diving,
Rich
Right, depends on cave, dive plan, buddy/solo, sidemount/backmount, scootering/swimming, etc. Make a plan taking into account the variables and go from there. No one rule takes care of everything.To answer the original question, I dive my stages differently depending on the dive and have fun doing it....
Safe diving,
Rich
Posting Permissions
Reply With Quote
Bookmarks