Double valve on a single tank

[Akimbo]

…Really curious about why someone would want a double valve on a single tank.

It is all a risk calculation. Although hard statics are almost nil, it is fairly well accepted that the most likely gas supply failure is a leaking second stage. On a no-decompression no-overhead dive you can just surface and fix it. When you can’t just surface you must weigh the probability of failure against your willingness to invest in and carry more gear.

A Y-valve with double dip tubes provides full redundancy of the first and second stage regulators and some protection from the dip tube getting blocked, but does nothing to protect against a blown cylinder-to valve O-ring, blowing a burst disk, or an empty cylinder due to a leak or diver error.

The next step in redundancy is a pony bottle or independent doubles of sufficient size that you can safely surface from a soft or hard-overhead dive. Some feel that manifolded doubles with an isolation valve is the ultimate in redundancy, but that does not provide the gas redundancy of a pony or independent doubles unless you dive with the isolation valve normally shut.

You could carry the analysis to extremes by putting an octopus second stage on both first stage regulators. But then you have to balance the probability of needing to share air at the same time as a failure that required shutting down one of your two regulators with the fact that the second stage is the leading regulator failure. You could also add a pony bottle to your isolation manifolded doubles. At that point, do you require your buddy to gear-up the same?

There comes a point where the backup systems become so complex and physically burdensome that they conspire to make you less safe — or so much work it isn’t worth doing. Each individual must make their own compromise based on abilities, finances, dive profile, and risk they are willing to accept.

 

[halocline]

Some feel that manifolded doubles with an isolation valve is the ultimate in redundancy, but that does not provide the gas redundancy of a pony or independent doubles unless you dive with the isolation valve normally shut.

I think your overall explanation was quite good except this point; as long as you can shut the iso valve when needed, there's full redundancy. If the iso valve or manifold was to fail (extremely rare....) there's not full redundancy. What's the difference if you dive with it open or shut?

 

[Tortuga68]

The next step in redundancy is a pony bottle or independent doubles of sufficient size that you can safely surface from a soft or hard-overhead dive. Some feel that manifolded doubles with an isolation valve is the ultimate in redundancy, but that does not provide the gas redundancy of a pony or independent doubles unless you dive with the isolation valve normally shut

Once you have a failure on independant tanks - pony or otherwise - that requires a valve shut-down, you've effectively lost the gas in that tank; with manifolded doubles, you still have access to 100% of your remaining gas - that's the purpose of a manifold, in addition to not having to switch back & forth between second stages during normal diving. No need to dive with the isolator shut, you may save more gas that way in the event of a failure, but even with it open you'll still have more usable gas than an equivalent independant setup

 

[Akimbo]

Once you have a failure on independant tanks - pony or otherwise - that requires a valve shut-down, you've effectively lost the gas in that tank; with manifolded doubles, you still have access to 100% of your remaining gas - ...

There's the conflict. That is true providing the gas loss failure does not occur at the end of the dive and/or your planned gas reserve is still there after you correctly diagnose and shut the right valves down. For example, an extruded cylinder O-ring or blown burst disk will cause very rapid gas loss in both cylinders of manifolded doubles and that gas is lost anyway. Not that big a problem at the beginning of the dive but at the end is another story.

Providing you follow the rule that a pony bottle is only used in an emergency, then you stand very little risk of that supply being compromised. That is an argument favoring either mounting your pony bottle valve-down or side-mount, assuming your primary is valve up. The reasoning is that physical damage to both regulators far less probable.

...that's the purpose of a manifold, in addition to not having to switch back & forth between second stages during normal diving. No need to dive with the isolator shut, you may save more gas that way in the event of a failure, but even with it open you'll still have more usable gas than an equivalent independant setup

Usable gas on independent doubles is the same providing you alternate during the dive. Granted, the probability that there will be remaining gas for your isolation valve to save is higher since cylinder valve O-ring extrusion and failed burst disks are rare relative to failed second stage and inflator valves. But this brings you back to that risk calculus.

This is a secondary reason I run with the isolation valve shut. The primary reason is progressive equalization.

Following up on Sat Diver and Bob/Grateful Diver's comments: Aside from protection, the second reason for valve-down is the way I use the isolation valve on doubles, which seems to be different than just about everyone in the US.

For me, isolation valve access must be quick, comfortable, and covenant because I open and close it 3-6 times per dive. An ex-British Navy diver first turned me on to progressive equalization. To me it makes perfect sense. You start the dive with the isolation valve shut. Once you breathe down the primary cylinder you reach back, equalize the two cylinders, and close the isolation valve again. The reserve aspect is a hallmark of most triple and quad cylinder rigs used by Cousteau and is far more effective than the early J/Reserve valves.

Each equalization can correspond with a pre-determined event based on the dive plan. For example:

1. Start meandering back toward the boat
2. Look more aggressively for the anchor
3. Stop fooling around and find the anchor
4. Forget the anchor and leave bottom right now

IMHO, the entire failure mode analysis behind the isolation manifold would be better served by leaving the isolation valve shut. If an O-ring, hose, or burst disk fails, gas is only lost from the one cylinder. You can still shut the regulator stop valve if the failure is upstream and potentially safe more gas — especially when you can't react immediately or misdiagnose the problem on the first try.

 

[Tortuga68]

Usable gas on independent doubles is the same providing you alternate during the dive

Not really... yeah they're going to have equal amounts of gas in them. Then when you have to shut one down for whatever reason, bye bye to that gas

Anyway, you're obviously happy with the way you do it, good for you

 

[Akimbo]

I think your overall explanation was quite good except this point; as long as you can shut the iso valve when needed, there's full redundancy. If the iso valve or manifold was to fail (extremely rare....) there's not full redundancy. What's the difference if you dive with it open or shut?

Sorry, didn’t see your post before replying to Tortuga68 where I may have partially addressed your question. Diving with the isolation valve normally closed has a greater probability of saving usable gas when the failure is on the cylinder-side of the valve seat — granted that is a relatively rare failure compared to upstream, but much higher flow rates = less time to react.

You are also likely to lose more "vital" gas when a failure occurs at the end of your dive when already at your planned reserve. The problem can is exacerbated when you are slow to shut the isolation valve; perhaps through misdiagnoses of the problem, high currents, entanglement, narcosis, injury, trying to find your buddy, etc. Most serious accidents are compound problems, often three or more. Does this make sense (I know, I need a copy editor)?

Keep in mind that this is all a probability versus convenience compromise. You have to ask yourself what you are willing to give up for each higher level of safety and the complexity that goes with it. Especially when each added level protects against a progressively lower probability of failure. If you can comfortably cycle the isolation several times per dive you can take advantage of the safety advantage and progressive equalization. It isn’t worth it if you are miserable doing it.

Let’s face it; the vast majority of recreational divers will never need to share air, let alone a Y-valve or pony. Just as very few technical divers will ever need to use the isolation valve due to a failure… thankfully. Naturally the equation changes the farther we push the limits and personal experiences cause us to question our sense of invincibility.

 

[halocline]

At one point I considered diving with the iso valve slightly cracked; enough to allow equalization in the tanks, but quick to close with a single twist. The standard procedure is to dive with it completely open so that there cannot be confusion as to which way it turns to close. This does mean that closing it takes more time. The ideal solution would be a re-designed iso valve that would close with one turn, I suppose. Anyhow, I can close it pretty quickly; that's what shut down drills are for.

There's theoretical redundancy and practical redundancy; with manifolded doubles you have access to all the gas in the event of a regulator/hose failure. This type of failure is probably hundreds of times more common than valve/tank o-ring/manifold failures that force isolating the tanks in order to avoid losing all gas. In order to lose all the gas in both tanks no matter what, there would need to be two separate failure points in the manifold; one side of the barrel o-rings (all of them) and the valve itself, or both sides of the barrel. This is exceedingly unlikely; in fact, I would venture a guess that failure on both tank o-rings or burst discs of independent doubles is just as likely. I've heard the nonsense that someone might get smacked with a rock hard enough to blow out the manifold; well, considering its about 3" from your head, I suspect in the event of an impact of that magnitude, you've got much bigger problems.

 

[stairman]

I dont see where manifolded doubles are more redundant than independants. If you happen to bump the isolator hard enough, your gonna lose em both. There is more 0 rings to the system as well. I dove doubles for years and even an h valve thru cavern training, never used a slingshot and have heard that the closeness of the orfices can cause reg clearence problems. IMO, sidemounted independants are the safest scuba unit there is. You can feather a leaky reg easily without reaching behind your head. You have an spg on each tank, and I have a wing inflator on both regs as well with a dual Nomad. Switching regs every 300 psi or so is not a concern and I never bump my head anymore on a valve. I realize entering a boat with them on would be a problem since most ladders arent wide enough so I would opt for a single on my next boat dive since redundants in openwater isnt so important.

 

[Akimbo]

At one point I considered diving with the iso valve slightly cracked; enough to allow equalization in the tanks, but quick to close with a single twist. ...

It is an interesting problem. There are devices called flow fuses that will automatically shut after a certain flow rate, but then you have to consider that they introduce their own failure rates aside from being expensive. I think most people would agree that the money would be better spend on greater physical protection.

 

[halocline]

I dont see where manifolded doubles are more redundant than independants. If you happen to bump the isolator hard enough, your gonna lose em both. There is more 0 rings to the system as well.

I wouldn't say "more" redundant, I'd say that the type of redundancy is likely to be more useful due to the availability of all gas in the most common type of equipment failures. Please provide a single, actual example of some diver, sometime in history, smacking the isolator valve hard enough to cause loss of gas in both cylinders while diving. The presence of the extra o-rings in the manifold improves reliability, because they are in series. IOW (on my type of manifold) three consecutive o-rings have to fail; one sealing and two back ups, before gas is lost. That's on each side, meaning if by some bizarre fate all three failed on a side, you could still shut the valve and have gas in one cylinder. But, people are somehow programmed to think "more o-rings=less reliability" no matter what.

The 'manifold vs independent doubles' is an often-discussed topic with, like so many scuba topics, lots of assumptions and opinions but little actual data to support some of the claims made. I guess it comes down to two main issues off the top of my head:
1. Which is more likely, catastrophic manifold failure or failure in both tanks on the same dive
2. Which is the more likely human error; not balancing tank depletion or inability to shut the iso valve

Depending on how you view those issues, you'll probably tend to prefer one or the other gear configuration.