Regulator Rubbish

[Nemrod]

http://www.undercurrent.org/blog/2010/0 ... r-rubbish/

This article takes me back 1968, sitting in class, discussing the merits of those new fangled flow through piston designs because they were fail safe unlike the diaphragm type. Well, I agree, a flow through piston design is safer because it has no upstream failure mode of the HP seat and can flow more air and blah, blah. This is why I used my yard mowing money to purchase a flow through piston Calypso instead of that old dangerous, busted down diaphram design of the Conshelf . You know, because my instructor said so of course . So, aside from people installing the pin incorrectly in a Conshelf, Titan, RAM etc. or not trimming to correct length in the case of a PRAM does anyone actually recall a pin failure or otherwise incapacitating a diaphragm regulator?

Okay, a pin that is too long can buckle (columnar failure) or it can push the sintered filter out of the barrel. If you do not know that. If it is too short, well, the regulator may not flow enough air at depth.

I did have a Royal Aqua Master, bought on eBait, "overhauled" by an "old Navy Master Diver" lock up solid in a pool dive at 1800 psi. Yep, it breathed fine right down to 1800 psi and then locked up. Huh? Well, it had a mix match of DA parts, RAM parts and wrong springs etc. Bryan fixed that one for me because SDII was a few days away and I did not have the kit.

Frankly, I cannot recall a diaphragm first stage failing from the upstream HP seat pin breaking, bending or anything else.

Dang, I am freaking out now, I guess I shoulda listened to my old instructor and stick with the "new" flow through design first stages.

Nem

[Ron]

I've never had anything upstream fail, first or second stage. I rebuilt all of the USD and Voit tilt valve regs, and pretty much every USD and Voit single hose. I've never had a regulator fail shut ever. I've been rebuilding and testing them for about 4 years now. That isn't that long, but I have been through pretty much the entire catalog.

Luis could comment more, but I would be willing to bet that the material specifications for the parts used in upstream first stages would not be conducive to failures based upon parts shearing or bending.

[luis]

Calling a flow through piston design as fail-safe, but not a diaphragm first stage is major nonsense. Most regulator design (that I can think of) have been intended to not fail in a catastrophic (air shut-off) condition, but some failures don’t always work as intended.

I can’t speak for the Calypso IV, but the Scubapro Mk-5 (which is the father of all flow through piston regulators) is known to have 3 failure modes were no air will make it to the second stage.

1) Some of the early swivels have been known to break off. This is not related to the flow through piston design, but it was an issue with this regulator. The swivel bolt material was changed.

2) The piston knife edge has impaled into the soft seat and pulled the seat out with it. It totally blocked the flow of air. Scubapro redesigned the seat and I don’t think there were ever any major accidents, but as far as I know this actually occurred a few times.

3) If the main spring breaks, the piston is capable of closing and there is nothing there to open it.
A balanced piston regulator (just like a balanced diaphragm) has no biased air pressure that will open the valve (or minimal bias, depending on the precision of the design). In the closed position, the term up-stream or down-steam is actually meaningless.

Only when the valves are open the balance is offset, but not by much and both cases the air pressure on the face of the seat has a tendency to keep the valve open.

There is no pressure trying to close a balanced diaphragm first stage seat since the seat moves inside the low pressure inside the balancing chamber (As opposed to a non-balanced diaphragm first stage).

I wouldn’t doubt that there has been a Conshelf (or similar) that could have failed in a closed position, but the chances are extremely unlikely (unless there was human error in the assembly). It would also take a major error to screw it up. The assumption presented in the article about the Conshelf is total nonsense…

I honestly cannot come up with any reason (theoretical, actual reports, or even anecdotal) to suggest that a balanced piston is less likely to close shut versus a balanced diaphragm, based just on this design principle (as opposed to a particular design flaw of a particular unit. And, unless I am missing something, the track record and history of a Conshelf will tell you that there were no design flaws on its basic mechanism… there have been improvements on seat material, but that just increased reliability.



Warning, don’t do this at home!!!
If you take a Scubapro Mk-5 first stage apart. It is possible to remove the piston spring and push the piston against the seat and open the tank valve. Since the piston is balanced, the air pressure is not trying to open the valve. As I mentioned, the up-stream biased does not apply when the first stage valve is closed.

Note: if there is a small leak on the piston knife edge, the air pressure will only then push the piston out. If this happens the piston will shoot like a bullet. I repeat… do not try this at home!!!



The major issue with an up-stream second stage is that they are not balanced and any increase in IP will tend to push the valve closed. They are not designed to handle the full tank pressure (like the up-stream valve in a Mistral) and if the first stage leaks it will shut off the air flow and the LP hose could burst with the higher pressure.


It is worth noticing that almost all pressure reducing regulators used in all industrial gas applications are non-balanced diaphragm regulators. Think of all the oxygen regulators use for welding. I don’t think they are ever serviced and some of the applications take these types of regulators into very hostile environments.



The first dive shop where I worked in 1971 was a Scubapro shop. I find it funny that the owner never tried to pull this story of a “fail-safe” flow through piston design. He surely made up every other BS story you could imagine (and few you wouldn’t even think of). But, he missed this one.

It was very entertaining working at that dive shop. There were many times that it was hard to keep a straight face, and then there were scary times when he started believing his own lies…

[Bryan]

The article is well named and the "rubbish" bin is where it should have been filed instead of publishing it.....Crap like this is where rumors start....Rumors turn into myths and half truths that are then accepted as fact and then spread by people who have know idea what they are talking about...Double hose regulators all breathe hard....Double hose regulators are dangerous etc etc...

I am currently working with a new Phoenix owner who's local LDS refuses to let him use it in the pool....The yoke is weak and will explode and you others around you and my store could be destroyed....I am not kidding in the least...

[Nemrod]

Well, the story is rubbish and aptly named. Under what circumstance could the little knob on the end of the push pin break off . The whole story is preposterous and likely an entire fabrication that typical dive shops promulgate to hawk their particular wares. But my instructor said it was so .

So, the dive shop owner cannot read the 3300 psi service on the yoke? Maybe he needs to go back to third grade math .

Nem

[kiwidiver]

While i am new to the double hose thing, i have serviced and dove a lot of conshelfs and other balanced diaphragm first stage regs and i have never had, seen or heard of one failing closed. In ~2000 dives in environments from the cold south of NZ to southeast Asia and 95% of those on USD balanced diaphragms i have only ever had three underwater failures (personally). one was a brand new soft seat in a conshelf SE2 first deformed and failed with a high IP (i seem to remember it was a first batch of a new seat material) the other two failures were attributed to a stupid service tech (umm it was me) who in one instance failed to correctly tighten down a second stage diaphragm and it flipped out during the dive and the last was a failed o-ring on a second stage hose that i knew should have been replaced.
I agree with Brian, there is far too much mis-information in diving and it is all driven by sales and huge egos. which is one of the main reasons i departed the dive industry.

[Bryan]

there is far too much mis-information in diving and it is all driven by sales and huge egos.

This in spades....And it makes a mockery of the great LDS owners who are still out there.

[swimjim]

I have seen a lot of regs fail. All due to extreme cold conditions. Sub 36 - 38 degree water. They have always failed in a free flow condition. At least you can sip air off it(if you don't have back up) and make it home.

Jim

[Nemrod]

Well, apparently down under they have diaphragm failures that lock the regulator. I only report the story.

You know, the Conshelf/RAM first stage is probably about as simple and reliable as one could want. I have to say that I would trust a Conshelf over just about any regulator of any sort, anywhere, anytime. Even with a dangerous old fashioned diaphragm upstream first stage.

I thought y'all might need some entertainment.

Nem

[Ron]

I told you Luis would come explain in detail. That guy is a genius. Where else can you go to talk dive gear where you have actual regulator designers on staff to ask questions?

[kiwidiver]

Well he doesn't exactly state what he found once the regulator was serviced, he stated "If the knob sheared off" I suspect there was more going on here like wrong parts or extreme neglect. not to mention the owner of the reg was a kiwi and they can’t be trusted
Early in my servicing days I had a Sherwood magnum in for a service that would just not behave itself, I think it was locking up at a certain tank pressure (my memory is a bit sketchy) I couldn't figure it out (not much can go wrong in one of these) so I sent it off to the Sherwood distributor. turns out it had a completely out of spec spring in it. I finally got out of the customer he had taken the reg apart and lost the spring so he replaced it with a random spring he found in his shed!
and thanks for the entertainment Nemrod.

[Nemrod]

OK, as an experiment, simply remove the pin. This simulates a failed pin. See what happens. Now, I already know the answer, I suspect most of you do as well.

But, the Regulator Rubbish story is still just that, rubbish. I just do not believe a Titan/Conshelf first will fail in that manner. Even the longer pin in the Phoenix is way strong and a certain person made the required calculations. I do know a pin that is too long can bend or shove the guts out of the first stage barrel. But this is an assembly error, not a mechanical flaw, two entirely different things.

James

[luis]

In case I wasn’t clear enough.

In the design of a balanced flow through piston and most balanced diaphragm (see note 2 below), the moving parts are not moving up-stream or down-stream. They are moving across the stream, or to be more accurate, the pressure forces act perpendicular (at 90 degrees) to their moving path.

Anyone that is familiar with a flow through piston (such as any of the Scubapro, Mk-5, etc.) can immediately see that the piston stem only see high pressure air around the perimeter in a radial acting compression.

In a balanced diaphragm (as a RAM, Conshelf, or older Aqua Lung Titan) it may not be as obvious, but the pressure also acts only on the radial direction around the seat poppet. There are no axial pressure forces trying to close that seat. The spring is the only thing that closes that seat.

The low pressure balancing chamber does not allow HP air from applying pressure thrust in a direction that would tend to close the poppet seat.

Several of us have experience a weak seat spring in a RAM or a Conshelf. I service a friend’s regulator that worked just fine as long as the tank pressure was below 1800 psi, but with higher tank pressure the first stage would start leaking and it got worse with higher tank pressures. I replaced the spring and everything worked fine.

The new AL Titan (and the AL Legend, as well as many newer balanced diaphragm first stages) use a geometry that gives more of a visual impression that the seat poppet moves across the flow and not up-stream. This is because they are side mounted. But in reality, they work exactly the same as a Conshelf or a RAM.

You can’t forget that fluid (air/ gas or water/ liquid) applies equal pressure in all directions (and surfaces that it is exposed to). The poppet seat on the Conshelf, the old Titan and the new Titan (etc.) are only exposed to high pressure air around its sides (perimeter), not at the ends.

In a Conshelf (etc.) the pin pushes the valve poppet seat open and the spring closes it. The name balanced, comes from: the pressure forces are balanced. The IP indirectly closes the valve by pushing the diaphragm (and pin) away and allowing the poppet spring to close it.

In a RAM or Conshelf it takes 8 to 10 Lbs. to open the first stage valve. It doesn’t matter if the tank pressure is 0 psi, 300 psi, or 3500 psi. It is very close to the same exact force, independent of tank pressure. I have measured it on several (many) first stages.

BTW, I have also measured the actual force (in pounds) to open an Aqua Master first stage (and a Mistral single stage valve) and as expected, it varies proportionally to tank pressure. I have the data and curves at home, but from memory, I can tell that it requires a lot more than 8 Lbs. of force to open the first stage of a DA Aqua Master, when you have a full tank. Note: keep in mind that the IP works against the amount of heavy spring force that works to open the valves and that is why you have a higher IP with lower tank pressure.


Note 2: the Poseidon balanced diaphragm regulator is very different to other designs, but I am not familiar enough with it to comment on its design. The Sherwood balanced piston regulator is also a different design. It uses two pneumatic pistons to accomplish a very good balancing job, but it is not a flow through design. The volcano orifice rides in a secondary pneumatic piston and there is a pneumatic balancing chamber behind this piston. It is a very good design.

[Ron]

We should sticky this.

[Nemrod]

In case I wasn’t clear enough.

The post was a tongue in cheek reference to the linked article which contends that the design of the balanced diaphragm regulator is flawed because it is a downstream design. Downstream and upstream are not really useful concepts when analyzing a failure mode for the operating dynamics of a first stage balanced regulator.

In the design of a balanced flow through piston and most balanced diaphragm (see note 2 below), the moving parts are not moving up-stream or down-stream. They are moving across the stream, or to be more accurate, the pressure forces act perpendicular (at 90 degrees) to their moving path.

Forces acting other than axially cancel themselves out and are not of any consequence.

Anyone that is familiar with a flow through piston (such as any of the Scubapro, Mk-5, etc.) can immediately see that the piston stem only see high pressure air around the perimeter in a radial acting compression.

I would agree that is the nature of a balanced piston design.

In a balanced diaphragm (as a RAM, Conshelf, or older Aqua Lung Titan) it may not be as obvious, but the pressure also acts only on the radial direction around the seat poppet. There are no axial pressure forces trying to close that seat. The spring is the only thing that closes that seat.

Forces acting other than axially are of no consequence. Yes, the spring acts to close the seat and the pin pushes the seat open. It is this failure the linked article was concerning itself with a mistaken understanding of the use of downstream and upstream terminology.

The low pressure balancing chamber does not allow HP air from applying pressure thrust in a direction that would tend to close the poppet seat.

I would agree, that is not the nature of the forces acting within a balanced diaphragm first stage regulator.

Several of us have experience a weak seat spring in a RAM or a Conshelf. I service a friend’s regulator that worked just fine as long as the tank pressure was below 1800 psi, but with higher tank pressure the first stage would start leaking and it got worse with higher tank pressures. I replaced the spring and everything worked fine.

Spring force closes the seat, ambient plus diaphragm spring pressure opposes the other forces and provides the needed force to off seat the HP seat from the volcano orifice.

The new AL Titan (and the AL Legend, as well as many newer balanced diaphragm first stages) use a geometry that gives more of a visual impression that the seat poppet moves across the flow and not up-stream. This is because they are side mounted. But in reality, they work exactly the same as a Conshelf or a RAM.

The parts appear to be rearranged to the casual observer but the function is the same.

You can’t forget that fluid (air/ gas or water/ liquid) applies equal pressure in all directions (and surfaces that it is exposed to). The poppet seat on the Conshelf, the old Titan and the new Titan (etc.) are only exposed to high pressure air around its sides (perimeter), not at the ends.

Thus those radial forces cancel one another out. Axial forces work to open and close the seat.

In a Conshelf (etc.) the pin pushes the valve poppet seat open and the spring closes it. The name balanced, comes from: the pressure forces are balanced. The IP indirectly closes the valve by pushing the diaphragm (and pin) away and allowing the poppet spring to close it.

I agree, thus the concept of downstream and upstream is wrongly applied by the author of the Rubbish article. HP air is not used to close the seat nor does it trap the seat as would be the case with an upstream design second stage. However, failure of the pin would result in a regulator not functioning, so, to some extent, the Rubbish article is not completely unfortunate. What is clear to me, even if you are not clear is that the author of the article is somewhat confused at best as to how regulators operate.

In a RAM or Conshelf it takes 8 to 10 Lbs. to open the first stage valve. It doesn’t matter if the tank pressure is 0 psi, 300 psi, or 3500 psi. It is very close to the same exact force, independent of tank pressure. I have measured it on several (many) first stages.

That seems about right.

BTW, I have also measured the actual force (in pounds) to open an Aqua Master first stage (and a Mistral single stage valve) and as expected, it varies proportionally to tank pressure. I have the data and curves at home, but from memory, I can tell that it requires a lot more than 8 Lbs. of force to open the first stage of a DA Aqua Master, when you have a full tank. Note: keep in mind that the IP works against the amount of heavy spring force that works to open the valves and that is why you have a higher IP with lower tank pressure.

I would agree.

Note 2: the Poseidon balanced diaphragm regulator is very different to other designs, but I am not familiar enough with it to comment on its design. The Sherwood balanced piston regulator is also a different design. It uses two pneumatic pistons to accomplish a very good balancing job, but it is not a flow through design. The volcano orifice rides in a secondary pneumatic piston and there is a pneumatic balancing chamber behind this piston. It is a very good design.

But, Luis, my instructor said flow through balanced piston designs are better than an old diaphragm especially it it is the one he was using and selling.

Nem