Decanting a Twin Tank Reserve

[surflung]

Decanting a Twin Tank Reserve
- I was told that Cousteau's first triple tanks used the 3rd tank as a "Decanting" reserve. They'd be diving off the other two tanks until empty, then open the valve from the the reserve tank to "decant" more air into the other two... until all three were the same pressure. Then close the valve on the third tank. Again, they'd keep diving the other two tanks until empty, then open the reserve until the tanks balanced again and close the valve on the third tank. This system allowed them to know how much they had left toward the end of a dive without having a pressure gauge.

You can do the same thing with TWIN Tanks and a J-Valve!
- After a really good re-build of the reserve mechanism on my Twin 38s, I was able to "Decant" from one tank to the other. I did it while diving at Sea Hunt Forever but only for 2 pulls and I didn't track how much time I got from each pull.
- Last week I re-built the reserve on ANOTHER set of twins (my "Bill's 38s"). Did it the same way and yesterday we tested it at the pool. The first pull gave me 8 minutes. Second pull gave me 3-4 minutes. 3rd pull 1-2 minutes. And 4th pull about 30 seconds. But even that last pull went from sucked dry to 30 seconds more of normal good breathing pressure.


J-Valve Re-Build Tips
1. The J-valve orifice and seat must be clean and smooth. I used 0000 Fine steel wool on the end of a pencil eraser to polish the orifice. Then, I blew out any dust or debris with HP air. THEN, I used a Q-tip and paper towel moistened with food grade silicon spray to wipe any remaining particles from the orifice.
2. I replaced the seat with a NOS seat I was fortunate to acquire from Captain Tom.
3. The spring may or may not need replacing but take note of any little washer/spacers. These are used to tighten/adjust the spring tension if you need it. You can test this now: Blow with just your lung pressure into the tank and flip the lever closed. Listen to see if it leaks. Flip the lever open to see that the reserve releases properly. If it leaks when the lever is closed, you may need a better seat, orifice, spring, or another spacer to tighten the spring.

[SeaRat]

Surflung,

You can do the same thing by using the twin reserve with a regulator which has a J-reserve on it. The regulator J-reserve is usually set at 300 psig. If it is mounted on a tank with a twin tank J-reserve, that reserve is set at 500 psig.



With both J-valves up, the pressure will go down to 500 psig and then be fed only by the non-reserve cylinder. When that cylinder gets to 300 psig, the center post J-valve on the regulator will engage, and warn the diver that he is getting low.

When the non-reserve tank gets to zero, the second, tank-mounted J-valve will have 500 psig, and tripping that will give you 250 psig in both cylinders. If it is then put back up, you get 250 psig in one tank to breath down to zero, and then you trip it again and get 1/2 of the 250 psig, which is what? 125 psig in both cylinders. Putting the reserve back up will then "save" that 125 psig in the reserve tank, and when the non-reserve tank is again at zero, tripping the reserve again will give you 75 psig in both cylinders. This is probably as low as it is practical to go with the reserve. If the diver hasn't surfaced by this time, something's really wrong.

Here's a diagram of the Healthways J-reserve twin tank manifold.



John

[surflung]

Surprised That Its Possible...
- My reason for posting this comes from my surprise that its possible to have a twin tank J-Valve that seals well enough to decant.
- I've heard, read, and experienced many accepted opinions on how twin tank reserves are supposed to work. Since all of the twin tank reserves we see nowadays are old, they don't often reveal how well they can and should work.
- I've purchased old twin tanks where the reserve was set and the tanks were full. And when I emptied the tanks, the reserve tank still held a good deal of pressure with no leaks. And, I had to pull the lever to completely empty the tanks.
- I've also had reserves that would slowly leak from the reserve tank to the other tank... Sometimes so that when you needed to pull the lever, there was nothing left in the reserve tank. Other times there was still SOME air in the reserve tank. And others where if you breathed gently enough, you'd actually keep up with the leak from the reserve tank and end up with both tanks empty without pulling the reserve lever.
- I've been told that the reserve seat is SUPPOSED to leak. That you're just supposed to detect it getting harder to breathe.
- My favorite "Twin 38s" had a reserve that (I thought) worked pretty well. But, I knew the seat leaked somewhat. Imagine my surprise when I took it apart and discovered that the spring was broken and it had 3-4 extra spring tightening spacers. The previous J-valve technician must have used the spacers to adjust the reserve too tight, causing it to break. I don't know why it worked for me, but now I had to repair it.
- Fortunately I had a spring from another manifold and a new seat from Captain Tom. I put it all together without the extra spacers. And it works perfectly. With the lever up, the reserve tank does not leak into the second tank... Not even if the second tank is empty.

What Made Me Think of Decanting?
- Near the end of my second 30 minute dive at Sea Hunt Forever, I ran out of air, pulled the reserve lever, and ended the dive a little early. When I checked the tank pressure after the dive, I still had 300 psi total in both tanks. These old tanks hold a lot of air at low pressure. 300 psi in two 38s is around 10- 15 minutes of dive time at these depths. I could have finished that dive with 10 minutes to spare.
- So, I figured the next time I ran out I would try pulling the lever to decant the reserve, then pushing the lever back up to see if it would give me a second reserve. And sure enough it did exactly that.

Depth Compensation
- My understanding is that a twin tank J-valve is really just a spring powered over pressure valve that you can manually open. When its not manually opened, it causes up to a 500 psi difference between the reserve tank and the other tank throughout the dive, regardless of depth. You can pull the lever at any time or depth and hear the air equalizing between the tanks.
- I read the info in Basic Scuba and Here's How I Understand it: If you pull the lever when you're deep, the reserve air will expand and give you more as you come up. If you pull the lever shallow, the reserve air will compress and give you less if you go deeper. Is this correct?

[SeaRat]

Mar 21, 2019 9:39:22 GMT -8 surflung said:

Surprised That Its Possible...

Depth Compensation
- My understanding is that a twin tank J-valve is really just a spring powered over pressure valve that you can manually open. When its not manually opened, it causes up to a 500 psi difference between the reserve tank and the other tank throughout the dive, regardless of depth. You can pull the lever at any time or depth and hear the air equalizing between the tanks.
- I read the info in Basic Scuba and Here's How I Understand it: If you pull the lever when you're deep, the reserve air will expand and give you more as you come up. If you pull the lever shallow, the reserve air will compress and give you less if you go deeper. Is this correct?

Lets walk through what actually happens with the tanks of air in a twin tank, 38 cubic foot system which is full at 1800 psig. With a 500 psig spring in the J-reserve valve on one side of the system, we have:

500 / 1800 = 0.2278 (This is the ratio of reserve air in one tank verses the whole single tank)

0.2270 x 38 ft3 = 10.5556 ft3

So we have in reserve 10.6 cubic feet of air in the single tank. If we let this tank stay at 500 psig, and the other tank goes to zero psig, then we trip the reserve, we will equalize the two at 250 psig. But we will still have in reserve 10.6 cubic feet of air. Now, if we are at nearly the surface, and breathing at a rate of 1 cubic foot per minute, we will have about 10 minutes of air. But if we go to 33 feet (34 feet is Silver Springs), which is two atmospheres, and breath at the same rate as on the surface, we will consume 2 cubic feet of air per minute, which is about 5 minutes of air supply. This is because the air is at twice the pressure, so each breath is twice as dense.

10.6 ft3 / 1 ft3/min = 10.6 minutes (surface)
10.6 ft3 / 2 ft3/min = 5.3 minutes (33 feet of sea water, 34 feet of fresh water)

Now if we were at 66 feet, the air is three times as dense. Therefore,

10.6 ft3/3 ft3/min = 3.5 minutes of air at 66 feet (68 feet of freshwater)

And so on down the depth. Now, if you pull the reserve at depth, and start ascending, what happens? We have the same breathing rate, at a surface air consumption rate of 1 cubic foot per minute. But the pressure at, say 68 feet in freshwater is three times the surface, so that rate is three times what it is on the surface. But as we ascend, the pressure lessens, and air density lessens, and therefore we consume less of that 10.6 cubic feet of reserve air per breath. So yes, as we ascend, we have more available air in reserve.

Now, Fred Roberts, in Basic Scuba, states that the spring release is slightly greater than above. We need to add the ambient pressure to the amount of pressure from the J-valve spring. Here's what he states:

All constant reserve valves are, in effect, depth compensated because of the inherent characteristics of the demand regulator. Because the regulator supplies air at ambient pressure, the spring setting will always be the reserve spring plus ambient pressure. The resulting pressure will be the cut off point. For example, at 100 feet the gauge ambient pressure is over 44 psig greater than it is at the surface; thus the regulator cut off is 44 psig higher and the reserve will therefore occur at 344 psig rather than 300....

He goes on to break down the use of twin manifolds, and the reserve on each. In each case, he adds 44 psig to the reserve (300 psig spring = 344 psig; 500 psig spring = 544 psig). But here's the thing, at the depth of 100 feet, where the engagement of the spring occurs for a twin tank (500 psig spring) at 544 psig, the total pressure of the tank would be 1844 psig too. If we go through the same division as above, where we use the ratio, we get:

544 psig/1844 psig = 0.2950
If we multiply this by the 38 cubic feet, we get 11.21 cubic feet in reserve.

38 ft3 x 0.2950 = 11.21 ft3
So it appears that there is slightly more air held in reserve at 100 feet. But at 100 feet, we have:

100 ft / 33 ft/atm = 3.03 atm
The pressure is 3.03 times the surface pressure, and so our 1 cubic foot per minute air consumption at the surface would be 3.03 cubic feet at 100 feet.

11.21 ft3 / 3.03 ft3/min = 3.6997 minutes, or 3.7 minutes of reserve air.

If we do the same for the pressure at 68 feet of freshwater, we get:
529.4 psig/1829.4 psig = 0.2894
If we multiply this by the 38 cubic feet, we get 10.9966 ft3 = 11 cubic feet in reserve.

38 ft3 x 0.2950 = 10.9966 ft3 = 11 ft3

So it appears that there is slightly more air held in reserve at 68 feet. But at 68 feet, we have:

68 ft / 34 ft/atm = 2 atm difference between the surface (at 1 atm) and 68 feet.

The pressure is 2 times the surface pressure, and so our 1 cubic foot per minute air consumption at the surface would be 3.03 cubic feet at 100 feet.

11 ft3 / 2 ft3/min = 5.5 minutes of reserve air.

Then, as we surface, the air consumption rate goes down as the pressure goes down. Therefore the amount of air available is greater as we surface. The air doesn't actually compress at depth, but the pressure is increased. Here is a table that Fred Roberts developed in Basic Scuba (page 166):



John

[SeaRat]

This might help people working on their USD J-valve manifolds.


For years I did not know that the parts that the arrow is showing were there. You have to go from the other side of the valve to remove them, and clean them out. Once that is done, a positive, air-tight reserve can be established with this valve.

John

[nikeajax]

Mar 21, 2019 16:14:11 GMT -8 SeaRat said:

Lets walk through what actually happens with the tanks ...

John

See this is why we love John so much I'm not even gunna try ta read that an' expect to un'erstan' it: I work best with hands on-empirical data... But I'll b'lieve ya buddy!

JB

[dead_dog]

This is not new. Back in the early '70s, SCUBAPRO had a twin valve with reserve on each end. Only made a few years, now hard to find.

[surflung]

Mar 27, 2019 12:42:21 GMT -8 dead_dog said:

This is not new. Back in the early '70s, SCUBAPRO had a twin valve with reserve on each end. Only made a few years, now hard to find.

That’s funny SoCal... nothing new about vintage I mean 😜. Learning about vintage gear and techniques actually is new to many of us and refreshing. Good explanation SeaRat.

- SeaRat you had a typo using 10 ft instead of Roberts 100 ft depth that confused me at first but I went back to the actual book and saw it was all for 100 ft depths.
- I don't think it means the same as "depth compensated" but Roberts assertion is that the reserve at 100 feet will be the spring strength PLUS the additional ambient pressure at depth. So, a 300 psi single tank reserve would give 344 psi at 100 feet.
- Additionally, a twin tank reserve with 500 psi spring on the surface would split to 250 psi per tank on the surface and give 294 per tank at 100 feet.
- So yes, the reserve will give a little more air at depth but not nearly enough to offset the increased air usage (SAC) at depth.
- SeaRat's calculations say you'd be using air 3X faster at that depth. So, decanting and continuing the dive at depth won't give any significant additional dive time... I got 8 minutes on the first pull at 10 ft deep in the swimming pool. I'd probably only get a minute or two at 100 feet. And the 2nd pull (if I was still 100 feet deep) would only give 30 secs to a minute. Better start coming up on the first pull. :oops:

[SeaRat]

Surflung,

It is an interesting thought experiment that keeps our minds agile. But going back to the original thought about Cousteau's tanks, I think they may only have done it once, at least at times. Also, the decanting produces quite a sound, with one full tank emptying into two empty tanks. That sound was discussed in Cousteau's The Silent World, where he talked about everyone hearing the decanting, and knowing that at least one of their group was on reserve air. From what I recall (I cannot right now find it) the whole group on a deep dive would then head for their decompression stops.

_MG_5130 by John Ratliff, on Flickr
Here is Cousteau's triple tank system.

_MG_5131 by John Ratliff, on Flickr
Here is the valving of the Cousteau triple tank system (original).

These photos came from Ryan Spence's collection exhibited at the Foss Maritime Museum in Tacoma, Washington.

John

[retpo]

I haven't done any research on this, but just wondering out loud. On a set of twin 38's with a single reserve lever, wouldn't the air transfer between the tanks to equalize out without going through the regulator. If you attached a stand alone tank pressure tester on the "O" ring, would the air still equalize when the reserve lever is pulled?

I always thought that when you go deeper, your reserve supply does not actually change but you are breathing thicker air so the restriction the J valve causes occurs faster. Therefore the diver feels the restriction and pulls the lever/rod sooner than if in shallow water. Or am I just old tired and confused :-)

In other words the air travels between the tanks while in the valve manifold rather than having to go through a regulator.

[SeaRat]

Apr 5, 2019 20:17:32 GMT -8 retpo said:

I haven't done any research on this, but just wondering out loud. On a set of twin 38's with a single reserve lever, wouldn't the air transfer between the tanks to equalize out without going through the regulator. If you attached a stand alone tank pressure tester on the "O" ring, would the air still equalize when the reserve lever is pulled?

I always thought that when you go deeper, your reserve supply does not actually change but you are breathing thicker air so the restriction the J valve causes occurs faster. Therefore the diver feels the restriction and pulls the lever/rod sooner than if in shallow water. Or am I just old tired and confused :-)44 psig

In other words the air travels between the tanks while in the valve manifold rather than having to go through a regulator.

Retpro,

Your thinking is correct, and is why in my first calculation I simply used the 500 psig behind the spring to look at the amount of air available. But Fred Roberts, in the quote from Basic Scuba, made the case that we have to add the absolute pressure to that calculation. At first I did not understand why, but then I realized that the diver would Experience out-of-air ("OOA") when there was still 44 psig in the "empty" tank. So in a sense, the system is depth-compensated.

John