Any DATA will be much appreciated. Not just anecdotes, but real DATA—graphs, charts, annotated lists, etc.
Thanks,
Joe
the data simply doesn't exist, you won't find it anywhere, you can ask everyone in the industry, it is simply not possible to provide data. Why? In order to provide reasonably accurate data you'd need the following information.
Total number of cycles of the first stage
Duration that the first stage has been pressurized
Run to failure
You can put a first stage on a bench and run it to failure and get that information, but how many times do you have to repeat that test to get any sort of meaningful data?
Some of the old manufacturers may have done that when they were first setting recommended service intervals, but I doubt it was done to something that could be peer reviewed.
In the review you cited above where you talked about leaks causing inhalation of water, that leak is on the exhaust diaphragm. Too many variables cause rubber to stick and it's impossible to get any data. Something could have gotten stuck in the body and wedged the diaphragm open, it could have seen a lot of heat/sun/chlorine and gotten sticky, it could have seen a really aggressive exhale that flipped the diaphragm through the holder, etc. It's not an annual service item either, but even if it was, you can't predict it. Anything you try to extrapolate out of that for regulator failure is useless because regulator failures that don't result in an accident aren't reported. I had one a few weeks ago, happened on the surface, it's not getting reported. I could tell you the last time that reg was serviced, but not how many cycles it had, how long it had seen pressure, etc. so I can't give you a failure rate on my own regulator, heaven forbid any information that is useful to publish.
I'm an engineer, I work with data all day, I'd love to direct you somewhere to get it, but there is no real value in having that data, and whatever the value is, it is paled in comparison in the cost of getting it.
If you service your regs annually, the likelihood of you seeing a failure is so near 0 it's simply not worth worrying about. Doesn't matter what kind of regulator it is.