What Happens when an Isolator Manifold Fails?

[Akimbo]

Your first move should be to isolate unless you are 100% certain which post failed (bet your life on it). Anything else is nonsense...

That is one of the many reasons I keep my isolation valve shut unless I am equalizing cylinders. The technique is called Progressive Equalization.

You treat any losing-gas failure like independent doubles (back or side mount), it doesn’t really matter which side of the first stage the leak is on. You switch to the backup cylinder and regulator if the failure occurs on the primary/online cylinder. Then you can close the post valve on the failed side and abort the dive. That can preserve some breathable gas IF the leak is upstream of the first stage HP seat.

Of course, opening and closing the isolation valve 2 to 6 times per dive can get old if it is difficult or awkward to reach, which was one of the many reasons I built a protector for valve-down doubles. For me, reaching the isolator on valve-up doubles is not that bad several times/dive in a 3mm wetsuit. Unfortunately it is much slower and uncomfortable in a drysuit with cold-water underwear.

BTW: This article is worth looking at when considering you strategies in a losing-gas failure.

Life Ending Seconds, 3000 to Zero in 72 Seconds

 

[TSandM]

The rate at which a manifold failure could empty your tanks depends, of course, on the magnitude of the failure. A wide-open manifold will empty 2 Al80s in no more than 3 minutes, assuming they were full. A leaking o-ring could give you hours to get out.

The "which to do first" argument has raged, and will rage; I remember a long PM conversation with cerich that I had at least six or seven years ago about this. The "go for the post" camp believes that you can often tell which side the bubbles are coming from, either from sound or from feel, if the leak is big, and that it makes more sense to close the post because you are actually directly influencing the gas loss by doing so. The isolator camp believes that, if you close the isolator, the most you can lose is half your gas -- the problem being that, while you are closing the isolator, you ARE losing it!

In fact, from everything I've been able to read on line in 9 years, massive gas loss situations in doubles are quite rare, so nobody has a series to show that one approach or the other is clearly superior. It has been my experience that, when I have had leaks (which have always been small ones) I have often been able to tell which side they were one, which I think is likely to be true if you don't have a heavy hood on and the leak is somewhat quiet. Wearing a heavy hood and having a massive leak might be a different situation -- but what is going to cause a doubles setup, which passed a bubble check at the outset, to suddenly develop a MASSIVE leak? Yes, a poorly installed hose could come loose, and I suppose there are some first stage failures that might create a huge leak (but obviously, such failures are like hen's teeth, because we don't read about them at all). But the most likely cause of a massive, new leak is probably scootering into the roof of something, and it would be my guess that that's more likely to be a dislodged first stage than anything else.

This is an angels on the head of a pin argument, I think.

 

[boulderjohn]

It has been my experience that, when I have had leaks (which have always been small ones) I have often been able to tell which side they were one, which I think is likely to be true if you don't have a heavy hood on and the leak is somewhat quiet. Wearing a heavy hood and having a massive leak might be a different situation --

My first tech instructor used to create simulated bad leaks by using a low pressure hose with one of those air blowers on the end. He would come up behind you, hold it next to the valve he wanted to simulate leaking, and start blasting air. I always wore a hood, and I could always tell right from left. If it is a lesser leak that you cannot hear, then you should have a buddy who is alert enough to spot it.

 

[TSandM]

I've had the same experience. Being air gunned, I have about an 80% success rate in identifying which side is involved. Having had a massive freeflow once, I can imagine a situation where the noise from behind is SO loud that you can't sort it out, and with dry gloves, you can't feel it. Still, what is going to give you that massive an air leak from a manifold that passed a bubble check on descent, unless you run into something HARD?

 

[KenGordon]

IStill, what is going to give you that massive an air leak from a manifold that passed a bubble check on descent, unless you run into something HARD?

A low pressure blanking port o-ring going pop. I have had this happen in a pool doing shutdown practice when turning a valve back on. I was surprised at how loud it was.

 

[TSandM]

Blanking port? You mean the port plugs that close unused first stage ports? Wow, I've never heard of one of those static o-rings blowing. But your story suggests a) that it occurred when pressurizing the reg, which isn't something you would be doing DURING a normal dive, and b) it's a downstream failure, which can be handled by closing the post. The OP was asking about manifold failures -- all static o-rings, and in most designs, also trapped. Yes, there are three valves, and they can fail or leak, but when I have seen valves leak, it has never been catastrophic. I suppose it could happen, but I'm not sure how.

Edited to add: I've been talking to Peter, who does our valve maintenance stuff, and aside from a burst disc failure, he can't come up with a failure that would lead to catastrophic gas loss. Tank neck o-rings can go, but they usually have been extruding for some time, and often won't pass a bubble check. Now, if you are repressurizing the system, all bets are off; that's equivalent to setting your gear up on land, which is where many failures will become clinically apparent.

 

[KenGordon]

Blanking port? You mean the port plugs that close unused first stage ports? Wow, I've never heard of one of those static o-rings blowing. But your story suggests a) that it occurred when pressurizing the reg, which isn't something you would be doing DURING a normal dive, and b) it's a downstream failure, which can be handled by closing the post. The OP was asking about manifold failures -- all static o-rings, and in most designs, also trapped. Yes, there are three valves, and they can fail or leak, but when I have seen valves leak, it has never been catastrophic. I suppose it could happen, but I'm not sure how.

Edited to add: I've been talking to Peter, who does our valve maintenance stuff, and aside from a burst disc failure, he can't come up with a failure that would lead to catastrophic gas loss. Tank neck o-rings can go, but they usually have been extruding for some time, and often won't pass a bubble check. Now, if you are repressurizing the system, all bets are off; that's equivalent to setting your gear up on land, which is where many failures will become clinically apparent.

Yes, I think user incompetence might have contributed to the failure I had just put together a pair of regs from a single setup and another first stage and got it a bit wrong wrt the o-ring alignment.

I quite often practice shutdowns on normal dives, mostly to pass the time in deco though. I've not blown an o-ring since.

 

[rjack321]

Manifolds basically don't fail. I've seen two fall off benches and they were bent but held gas. Far more abuse than is possible even smacking the ceiling while scootering.

The burst disks do like to dribble though since people tend not to replace them often, its hard to rinse in there, dissimilar metal corrosion sets in, and you get little bubbles from around the plug. You can also get dribble bubbles from the knob stems, as the single oring in there packs down it will give a slow leak out from under the knob.

Reg failures are far, far and away more common. In my experience, piston regs tend to creep and then leak via the 2nd stage, diaphrams (when they go) tend to leak past the HP seat into the sealed chamber and blow out the outer diaphragm with a bang. I don't think I've ever seen a 2nd stage fail by itself, its generally instigated by a failing 1st.

 

[Vegan Shark]

Well, I feel better after all the discussion

Does anyone have a good resource that explains how every piece of scuba gear works, with diagrams, animations, etc? I've learned a lot about tanks and valves through all this, but it's about time I learn the ins and outs of all the gear, and what causes things to go wrong.

 

[kensuf]

Well, I feel better after all the discussion

Does anyone have a good resource that explains how every piece of scuba gear works, with diagrams, animations, etc? I've learned a lot about tanks and valves through all this, but it's about time I learn the ins and outs of all the gear, and what causes things to go wrong.

The PADI Encyclopedia of Recreational Diving is a pretty good treatise on all sorts of things. And I say this as a NAUI/SDI/TDI instructor.

Google, of course, is another good source.