In the book "Against the Tide: Rickover's Leadership Principles and the Rise of the Nuclear Navy", Rear Admiral Dave Oliver (Ret.) narrates some of his experiences in the USS George Washington Carver, SSBN-656, launched in 1966, and in which he was hired as reactor controls officer.
In particular, he mentions the problem of hydrogen generation and build-up which occurs during the battery charging process.
According to the Wikipedia entry for oxyhydrogen, also called "Knallgas" for obvious onomatopoeic reasons,
At standard temperature and pressure, oxyhydrogen can burn when it is between about 4% and 95% hydrogen by volume". So one would really like to keep hydrogen concentration in an underwater closed space below the 4%-by-volume mark. Incidentally, the most bang for the mix would be reached at the 2:1 hydrogen:oxygen ratio (I think this means that we would have a 66%-by-volume hydrogen concentration, not sure).
In a diesel submarine, hydrogen generation is not a problem. Batteries are charged by running the diesel engine while on surface or schnorkeling. The diesel engine is air-breathing and is configured to suck air through the length of the submarine hull. Air circulation is high enough that free hydrogen never reaches dangerous values.
In nuclear submarines however, batteries can be charged while fully submerged. Thus free hydrogen can become a problem quickly. To mitigate the risk, air is made to circulate throughout the hull and the hydrogen is captured (likely, the air is made to circulate through a catalyzer).
The author narrates an event "in early 1968" whereby George Washington Carver was put into "Condition Baker" while battery charging was ongoing. In Condition Baker, the boat is compartmentalized and airflow thus evidently stops. Hydrogen concentrations in the compartments skyrocket immediately to dangerous levels. It turns out that "Condition Baker" was meant to make a collision between a diesel submarine operating at periscope depth near commercial vessel lanes and a surface ship survivable. Nuclear submarines would operate far from commercial lanes and setting them into "Condition Baker" was just a procedure fossil, now dangerous in its out-of-context application.
Now, on page 42, the author says the following which is not entirely in accordance with known oxyhdrogen behaviour:
...the process of charging batteries, even on board a nuclear submarine, is one of the most dangerous routine evolutions the submarine performs. The procedure produces hydrogen, especially in the last hour of the charge, and hydrogen is not welcome on board a submarine. As most high school chemists have learned, if the hydrogen in a flask reaches the value of 8 percent, it spontaneously combines with the oxygen in the flask. The chemical reaction produces a lost of heat - lots and lots - which on board a submarine is enough to peel the hull back as if it were an overripe banana.
Were could the "8% hydrogen will kill your boat" rule come from? Is it a rule of thumb? Would it not be better to keep the concentration below 4% instead?
The 8% are serious business and have Hollywood precision. The author is monitoring hydrogen concentration using two measuring devices (how does one measure hydrogen concentration btw?). During the Condition Baker event:
...the hydrogen needles had leaped to the red area of the gauge. The were now indicating 3 1/2 and 4 percent. One gauge, two needles, one indicative of the conditions in the forward part of the battery well, the second reading hydrogen concentrations in another sector. No-one in the manoeuvering said anything for the next minute except "The charge is off, Sir." Every eye was focused on the climbing hydrogen needles. Everyone was silently willing them to stop. Both quivering indicators had passed 6 percent, and their thin points were ever creeping upward toward the next number.
When we built Carver, I had helped install the sensors that fed those needles. An electrician and I had tried to optimally position them in the well, but our efforts had really just relied on a guess. What if I had been been wrong by just a few inches -- a few tenths of a percent? The announcement of Baker had set in process an evolution that completely isolated the battery well from any cleansing airflow. We were not going to restore the diluting flow until we reached the surface. Were the gauges stabilizing around 7 1/2 percent, or was I engaging in wishful thinking?
I do not want to impugn the author's recollection, but I feel he is going a bit overboard with the description here. Would anyone in the know be able to elucidate?