Quote: (Originally Posted by Genesis)
However, there is anecdotal evidence that it don't work that way in all (or even many) cases. Example - there were a few people killed in a CO2 fire extinguishing discharge accident a number of years ago. They were within ten feet of the door - and safety - when it went off. All evidence suggests that they fell almost literally where they stood, which implies that the blackout occurred essentially immediately. Unfortunately there was nobody near (but outside) who saw it and could thus effect a rescue, nor to report on exactly how fast they fell. (The discharge was not in response to an actual fire.)
Now granted, this was with a CO2 level VASTLY exceeding what you'd get from a breakthrough, but still - its sobering to think of in this way. What's the threshold for that sort of "bad result"? Do we know?
I think there is more to it than that. Ever tried this? Drink half a large bottle of coke. Give it a gentle swirl to get some bubbles going then squeeze the bottle and inhale at the same time (not the coke!). That's pretty much neat CO2 and it's effects are almost instant. But!!!! We never get that, since we will always have O2 and N2 in the loop and the O2 being in fairly constant amounts. Remember my "non-scrubber" test above - a million miles from the 'coke' hit.
The problem we have is that all the reports we have from people taking CO2 hits have no idea of how long they have been breathing CO2 before they feel the effects. In fact, didn't someone post that "their mate" did an entire dive without o-ring and spreader bar!!! So here's the thing; others are reading this and extrapolating that the effects of CO2 come on too fast to make any monitor or alarm useful. I'd suggest that's wrong and that maybe, possibly, the
noticable effects in some people may come on faster than
some people react. However, there is no evidence I've read that shows CO2 levels going from 0 to lots instantly in a scrubber, in fact quite the opposite, but also there are no reports of how long it took people to notice these symptoms.
So, to my mind, anything which alerts the user before they know, or even as they know, they have a problem is a good thing.