Reported StoryI was using a rebreather while on a wreck dive. While swimming along the deck at about 100 feet, I noticed that the solenoid was firing more often than normal for a constant-depth dive. This was a dive I had done many times, and I was quite sure that the solenoid firing was not normal.
I bailed out to open circuit and terminated the dive with my buddy following me. The bailout was uneventful, with one gas switch to oxygen to complete the small decompression obligation. After checking the oxygen sensors I was convinced that the old cells were very likely "current limited" and could not show a partial pressure more than 1.0 of oxygen. Because the set point for oxygen was 1.3 bar, my solenoid was firing too often trying to raise oxygen partial pressure.
At the time I owned five rebreathers; thinking back, I realized that I hadn't used this one in more than a year. I hadn't checked the date of the cells, and they were beyond their useful life. Since this was the first dive in a sequence, I hadn't yet spiked my cells with O2 at the end of the dive to confirm that they weren't current limited.
CommentsOxygen cells in contact with oxygen produce current measured in millivolt (mV). The output increases with the increasing partial pressure of oxygen. Normally, in contact with air which contains 21 percent oxygen, the cell's output is in the range of 8-13 mV and in oxygen it is 38-62 mV. This range varies a bit by manufacturer.
The calibration process at surface before the dive includes flushing the cell with 100 percent oxygen (partial pressure of oxygen (PPO2) = 1 bar), reading and recording the output of the cell in mV, and mapping that value to the value of 1.00. The assumption is that output would linearly increase with increasing oxygen pressure at depth through at least a PPO2 of 1.6. In this case, all the cells were already partially exhausted and could not respond with output greater than 1 mV regardless of how high the PPO2. Cells with such failure are called "current limited cells."
The unit's computer was set to maintain PPO2 of 1.6 bar in the breathing loop; thus, when output from the cells indicated that it was just 1 bar, the computer activated solenoid (electrical valve) to add more oxygen in the loop. Since the cells' output did not change, solenoids were activated again and again, and this alerted the diver that something was wrong. Switching to the bailout gas source was the right move because the diver was at risk of oxygen toxicity since the PPO2 in the loop was much higher than indicated on the display.
The primary mistake here was starting the dive without checking the sensor's expiration date. There have been several known fatalities on rebreathers with old oxygen cells, and this diver was clearly in an area of very high risk. Even cells that are within the rebreather manufacturer's recommended service life may become "current limited."
To mitigate the risk of oxygen toxicity, when approaching the manufacturer's declared lifetime for the cells, the diver could run a lower setpoint (1.0) on the first dive in a sequence and check the linearity of the cells at the end of the dive with a higher PPO2.
For more information, see http://www.advanceddivermagazine.com/articles/sensors/sensors.html.
— Bruce Partridge