Regulator maintenance intervals

[tbone1004]

Well, you did know that going in....
Getting back on topic.. Most parts kits are available with some manufacturers even offering direct purchase. A good few of the OEMs have pushed their recommended interval from yearly to 2 years.

There are good shops that will test the set and tell you, "no, it's working fine. I'll tune it for you, but you don't need a rebuild just yet." Other shops, "yeah, good thing you brought it in. It needs a rebuild for sure!" and not even put a gauge on it.

Most is not accurate. Very few are freely available, and only one major regulator manufacturer allows it *Poseidon*, all other major brands explicitly forbid it in this country *Scubapro, Apeks/Aqualung, all of Huish, etc.*. Poseidon has a 3 year interval fwiw.
I will agree that the standard should be to take it in annually for a $20 checkup where the cracking pressure and IP are checked and adjusted if necessary, and the regulator is inspected for cleanliness *i.e. salt water intrusion*. If they fail, then service, if they don't, then let it go.

 

[Bigbella]

Even old stalwarts like Poseidon, now recommend service at 24 month intervals, to maintain warranty, rather than annually; and a bit less of one for other products, such as their FFM . . .

 

[Fyffer]

Zeagle is every two years on newer regs. Flathead 7 and newer.

 

[loosenit2]

@tbone1004 Appreciate the insight on DEMA and their availability.

The responses here have been interesting, and honestly, about what I expected. What appears to be lacking is any data that would drive decisions. A key component to any 3rd or 4th generation maintenance plan is the analysis of reliability of parts and systems, either from testing, or observed field data, without that it is just guessing, perhaps educated guessing but still guessing.

It is possible that manufacturers have that kind of data and use it to drive decisions, but like so many things in the SCUBA industry it appears to be all proprietary and closely held. I will continue to monitor the thread for any new insights. I am still hoping that D6, with their direct to consumer and self service friendly model, might be able to shed some additional insight when DEMA is over.

 

[jgttrey]

I don't think this the right analysis - or at least a complete paradigm for analyzing the issue. A couple of reasons for this:

1. The operating environment is too variable. There's no reliable set of time-between-failure data and there won't ever be because so many of the "failures" are the result of the huge variation in the way equipment is used and maintained post-dive rather than the natural lifespan of some part.

Consider the differences between (1) weekend quarry diver who dives in fresh water and cares for his gear, (2) a salt water vacation diver who lets the valet dive op rinse (or not) his gear and only uses it twice a year and (3) a regulator used on a bailout cylinder that gets a lot of water time, but not a lot of use. In a vacuum, or cared for perfectly, regs can and should function for a very long time. It's how they are used that shortens that. There's just no way to boil it down to one number. If you flood your first stage with saltwater, you can throw the statistics about theoretical part lifespans right out the window.

2. For 90+% of divers, "service" is more about inspection than replacing parts. You can check the health of your regs with an IP gauge, leak tests, and some simple tests for cracking pressure, etc., as described above. But, if you want to look inside the first stage to see whether it is bright and shiny or starting to corrode from being flooded, you have to open it up. That's the expense of service, not the parts. And just opening it up requires that some parts be replaced anyway, like first stage diaphragms.

To me the current regime actually makes some sense. A relatively short service interval that gets beat into new divers so their gear at least gets checked. More sophisticated divers that bother to learn more about their gear will quickly realize that their regs won't explode on the 366th day. And, folks that service their own gear probably own enough of it that they can just dive it until it fails a function check and then service it.

This is one of those areas where perhaps we are not looking for the "optimum" schedule, but the one that that is calibrated for the least common denominator.

 

[loosenit2]

1. The operating environment is too variable. There's no reliable set of time-between-failure data and there won't ever be because so many of the "failures" are the result of the huge variation in the way equipment is used and maintained post-dive rather than the natural lifespan of some part.

An insightful point but not at all unique to SCUBA. RCM is applied readily to the manufacture of passenger vehicles. Personally owned vehicles are subject to very similar variations in use and maintenance that you describe. Some owners are meticulous about on-road use and adhering to a service schedule, others use there vehicles in a very rough and tumble way and are much more cavalier about maintenance. That variance invariably contributes to a lack of reliable data on MTBF once vehicles are in common use by owners.

What is still possible is generating service schedule based on the probability of failure of individual parts in the design and testing phase. As a piece of equipment is designed and manufactured it should be tested to determine failure probabilities in isolation and then in combination with other design elements. That testing results in an initial service schedule for inspection and replacement of parts that prevent failure while in use. This is of course subject to extreme examples of failure at the extremes of probability modeling.

The service intervals and methods I am familiar with in SCUBA equipment now is either inspect or overhaul. If an inspect item indicates a problem it generally does not result in selective replacement of parts, but total overall with a full parts kit. If it is time for maintenance it is done as a full overhaul. This methodology clearly falls into Generation 2 maintenance schemes

 

[jgttrey]

But diving is not driving a car and the consequences of a breakdown are more severe. For Joe Average Diver, planning around mean time between failures is not of much comfort. We're not trying to optimize fleet maintenance. Plan around the dumbest driver. It's life support equipment.

Part of the reason you don't usually selectively replace service kit parts is that they are cheap relative to the labor. It's the same reason you replace the oil filter when you get an oil change even though, in theory, they might have different service lives. It isn't worth the labor to do one without the other. The other reason is that once disassembled, reusing old parts can lead to whole new problems as they have "taken a set."

 

[Edward3c]

One thing I’ve not seen mentioned is the deteriorating of O rings when not used. A reg only used once a year is more substantial that one used throughout the year.

 

[Bigbella]

One odd thing that potentially discourages consumers, from more regular maintenance is the rather recent trend, among some shops, to only service a very few common brands; and to farm out everything else, at no small expense, A local shop, in Northern California, does almost no in-house repair any longer; sees itself more as a ersatz travel agency; and ships most everything to another branch.

Compounding that, is a newer lack of confidence in the woefully poor or minimal training of many scuba techs nowadays -- whose certification, for working on life support equipment, typically involves a couple of hours, sipping Nescafe, at a Sheraton conference room; and whose piss-poor salary, usually at the lower end of the dive shop totem, almost guarantees transience . . .

 

[loosenit2]

But diving is not driving a car and the consequences of a breakdown are more severe. For Joe Average Diver, planning around mean time between failures is not of much comfort. We're not trying to optimize fleet maintenance. Plan around the dumbest driver. It's life support equipment.

Part of the reason you don't usually selectively replace service kit parts is that they are cheap relative to the labor. It's the same reason you replace the oil filter when you get an oil change even though, in theory, they might have different service lives. It isn't worth the labor to do one without the other. The other reason is that once disassembled, reusing old parts can lead to whole new problems as they have "taken a set."

Again a fair point about risk but I think that RCM accounts for this as well. The aviation industry uses RCM as a basis for maintenance as well. Airlines and airplane manufacturers are must less accepting of risk then are automobile manufacturers and thus there maintenance schedules are much more aggressive; however, they use the same methodology. Formulating the probability of failure of parts and then inspecting and replacing them before they fail, the difference is in what probability of failure they accept. Where an automobile manufacturer may accept only a .05 (5%) probability of failure at x hours/miles and aircraft manufacturer may develop a more aggressive maintenance schedule that accepts only a .005 (.5%) probability of failure.

Your point on labor and design prohibiting selective replacement of parts is not lost on me either and is a valuable insight to the conversation.