Question Skipping 1st stage Maintenance?

[halocline]

Ok let me try - It’s not the IP guage that is dropping IP, It’s the SPG to measure tank pressure that appeared to drop 20 bar over 4 hours when the valve was closed. IP remained constant. Then mysteriously after reopening the valve the SPG was back to 100 bar tank pressure. So I am wondering what’s happening on the SPG for it to drop when no air was consumed…

I understand that, and I thought I addressed. It means you have a slow leak. It’s either in the o-ring between the tank and regulator, or in the regulator. It could be lots of place in the regulator, so you need to submerge your tank/reg in a tub and wait for the bubble to appear.

But I thought at some other point you were discussing IP drop under demand. If not, sorry, you can ignore that part of my reply.

 

[Pearlman]

Now I no longer recall if the tank valve was shut at the time or open - I don’t see any anomalies now and am unable to replicate it. The IP lockup seems to be good. Once I get to servicing the mk19evo will take the tank to the local swimming pool and check… Thanks everyone so far!

 

[Tanks A Lot]

[...]Why did Poseidon and Conshelf have (non-compressible) liquid containers for their environmental chamber if all you have to do is increase the (air) pressure inside the environmental chamber?
I do not disagree that pressure increases in the environmental chamber as the bubble collapses. But I don't think it causes a 1:1 increase in absolute IP.

My opinion is that due to diaphragm thickness and elastic forces in the main diaphragm, the increasing air pressure which collapses the bubble in the environmental chamber isn't transmitted to the pin, but is resisted by the diaphragm.[...]

I apologize for the tangent and long post that has not much to do with the original thread, but I feel like it is an important concept to understand.

This is a difficult topic to wrap ones head around. I do not find it all that intuitive, but find that looking at the physics and drawings certainly helps.

Looking at a environmentally sealed diaphragm, it becomes clear that once the ambient pressure pushes onto the diaphragm, that force is further transmitted to the transducer, which in turn presses onto the main diaphragm:

Now it is often asserted that the transducer is necessary for the transmission of pressure, but that is false. It is necessary in a very special sense, but the need to transfer pressure isn't really the main issue.
If we look at an unsealed diaphragm first stage, it is exactly identical to a sealed one, except for the missing transducer and outer diaphragm. Clearly there is no pressure transducer involved pressing onto the main diaphragm, yet the ambient pressure suffices.

There is often a disconnect introduced here, where in the sealed version, that very same pressure wouldn't suffice anymore. If we imagine a sealed version with the pressure transducer removed, but a suitable flexible diaphragm, it becomes clear that there isn't really any difference between the sealed and unsealed version. The pressure on the outside of the outer diaphragm will be equal to the pressure between the diaphragms.
It is obvious that by removing the outer diaphragm there wouldn't be any change in pressure whatsoever and we would be left with an unsealed design. However, we must also acknowledge that there is no difference between air pressure or water pressure. Either will act onto the main diaphragm the same way.

So why the need for the pressure transducer?
The answer is rather simple. Imagine the pressure in our above picture would increase, let's say to 2bar. Without a pressure transducer, the volume of the chamber would be half of what it was before. And there just isn't enough space for the diaphragm to flex into. It would bump into the rest of the mechanism.

The engineer must make sure that pressure is transmitted from the outer diaphragm to the inner diaphragm and he must do so while keeping the volume between the two diaphragms as a semi-rigid container. As we saw above, if we treated it as a flexible container, there wouldn't be any space for the diaphragm to flex into and at a certain point it will collide with the bias spring and its retainer. The engineer can achieve this rigid container requirement by either filling the space with an in-compressible liquid, or by throwing an in-compressible part, like the pressure transducer between the two diaphragms.

The result is the same. We end up with a volumetric space as small as possible and a outer diaphragm that flexes rather little. Once it touches the transducer, it doesn't really flex at all anymore, but rather pushes on the transducer. More importantly, the pressure inside the ambient chamber between the two diaphragms is fixed at a certain pressure (Within limits, due to the diaphragms slightly flexing at the edges, it is not a true rigid container). That means that counterintuitively, the pressure inside the ambient chamber of a diaphragm regulator, does not change with depth!

We can easily imagine a design where the need for an in-compressible liquid or pressure transducer does not apply. But these are purely imaginative. We could for example elongate the chamber between the two diaphragms to an absurd amount. This would give the diaphragm enough room to flex:

Now I said before these are purely theoretical. First no one wants to lug a huge chunk of brass like this around. Secondly, there isn't really a material that would be sturdy, yet flexible enough for an outer diaphragm like this.

 

[Tanks A Lot]

This whole debate started when it was asserted that the air bubble that sometimes forms between the two diaphragms is of no concern. In a sense that is correct if we look at the above drawings. If we removed the pressure transducer, the whole mechanism would work just as well, at least in principle and with the above mentioned caveats. And here it doesn't really matter if there is a air bubble or not.

However, this would assume that the pressure would propagate as it would in an unsealed or imaginative transducer removed scenario, where the volume in the chamber between the two diaphragms changes. Yet this is clearly not the case, as we do have a transducer in place, which essentially keeps the volume between the two diaphragms at a fixed rate (Again, within certain limits, the diaphragms are somewhat flexible at their edges, so it is not a real rigid container in the true sense).


And here comes the real reason why this air bubble between the diaphragms and an outward bulge is bad news:
First off, it is important to understand that the effective diameter of the inner and outer diaphragm differ. This difference in the diameter of the diaphragm is a complex calculation and partly due to different stiffness of the materials that either diaphragms are made out of. The outer diaphragm is almost always made of an extremely flexible material, while the inner diaphragm is made up of a stiffer "sandwich". This coupled with their different diameters will results in different forces being applied, as pressure is defined as a certain force acting on a given area (P_Pressure=F_Force / A_Area).

Going back to our properly setup sealed diaphragm without a an air bubble, we can see that the crucial calculation lies between the top diameter of the transducer and its lower diameter. Whatever force is applied to the top of that diameter will proportionally be applied to the bottom of the transducer. Remember that the pressure inside the dry chamber does not change, it is a semi-rigid container!

If we look at the below graphic and imagine the air bubble having to get compressed before the outer diaphragm touches the transducer, we can see how the pressure now acts differently. Due to the outward bulge and the flexible outer diaphragm, the pressure on either side of that environmental seal will be equal. Our state at the surface would like as follows:

The trouble arises, once we start to descend. Now the pressure inside the whole chamber will change, as we had an air bubble present that was able to get compressed. We didn't have a rigid container as a dry chamber anymore, but a flexible one until the diaphragm touched the transducer again.

And contrary to popular belief, there will be no lag in change of intermediate pressure, but rather a hastening in the increase of intermediate pressure. This part is almost universally understood poorly, as our intuition jumps to the conclusion that the sensing in pressure changes would be delayed.

In a nutshell, the key-points are:

• The pressure inside the ambient chamber does not change with depth on a sealed diaphragm regulator. Ask yourself if the volume changes or if any gas gets added? The answer to both is no, so pressure must remain constant.
• As it is a rigid container, pressure must be translated by the transducer and nothing else.
• Once we have an outside bulge with an air bubble, the rigid container requirement is violated and the now flexible ambient chamber can up to a certain point directly translate pressure onto the main diaphragm. This leads to a hastening in intermediate pressure increase, not a lag of it.

I hope that this clears up things a little, as I often see the concept of a dry chamber on a diaphragm regulator and its pressure transducer misunderstood.

 

[Umuntu]

This thread …

 

[CG43]

Very fine explanation and drawings Tanks A Lot !

Unfortunately, there is still a disadvantage to the dry chamber construction.
The pressure in the chamber is always only a little over 1 bar, regardless of the depth. and lies on one side of both membranes.
The IP on the inside of the 1st stage membrane rises with the depth. This leads to the pressure differential to which the IP membrane is exposed increases with the depth.
For example, if the IP is 10 bar and you dive 100 m deep, the pressure difference inside/outside IP membrane is 20 bar also twice as much as on the surface. In the case of overcompensated constructions, this effect is even more worse .
The pressure differntial for the environment membrane is only the ambient pressure .
With the MK 19evo, SP is trying to reduce the load on the diaphragms with large pressure plates. In detail, it looks quite good.
A construction that generates unnecessarily high loads remains questionable.

 

[elmo]

We can easily imagine a design where the need for an in-compressible liquid or pressure transducer does not apply. But these are purely imaginative. We could for example elongate the chamber between the two diaphragms to an absurd amount. This would give the diaphragm enough room to flex:
Now I said before these are purely theoretical. First no one wants to lug a huge chunk of brass like this around. Secondly, there isn't really a material that would be sturdy, yet flexible enough for an outer diaphragm like this.

Another flaw in the theoretical long chamber design is that the longer chamber has a greater volume so the outer diaphragm would need to deform twice as much to balance the pressure. You end up chasing your own tail with this “solution”.

• Once we have an outside bulge with an air bubble, the rigid container requirement is violated and the now flexible ambient chamber can up to a certain point directly translate pressure onto the main diaphragm. This leads to a hastening in intermediate pressure increase, not a lag of it.

The first part you explained and illustrated beautifully: the bulging chamber performs as a flexible ambient chamber.
But I cannot see why the intermediate pressure increase would be hastened once the bulge is compressed. Surely a portion of the outside ambient pressure is transferred through the chamber to the main diaphragm, and a portion transferred as a force through the transducer. In any case the forces on either side of the chamber must balance.

My main concern with the bulging system is that for the bulge to occur something must be leaking. Either gas is leaking from the inside of the regulator (not what we want), or water is leaking from the outside, which would mean it isn’t really environmentally sealed.

 

[rsingler]

My main concern with the bulging system is that for the bulge to occur something must be leaking. Either gas is leaking from the inside of the regulator (not what we want), or water is leaking from the outside, which would mean it isn’t really environmentally sealed.

IMO, if it's not Helium transiting the main diaphragm, it's air leaking in from the outside during storage. A properly fitted environmental seal is concave when the reg is depressurized. That means the pressure in the outer chamber is below ambient.
Without the surface tension provided by seawater, a questionably designed seal like Scubapro's will leak air. When the reg is next pressurized, the seal bulges.

But that was the easier question.
I do not accept that the IP increase will be hastened with a bulge, and I still think it lags.
But it's going to take me some time to do anything approaching the beautiful graphics that @Tanks A Lot has done so nicely, to explain my position. I think the answer revolves around the flexible portion of the main diaphragm that is not held in place by the pin hat. When increased pressure is delivered by air pressure, instead of the hard transmitter base, the untethered portion of the elastic main diaphragm will move without moving the pin hat. Hence, IP will lag.
Lemme see if I can diagram that.

 

[lowwall]

How about a third opinion

I don't think the bubble makes any difference.

Without an air bubble, the environmentally sealed section stays at a constant pressure since it is a fixed amount of gas at a constant volume. That pressure is transmitted directly to the diaphragm. As external pressure rises above the pressure in the sealed section, the additional pressure gets transmitted to the main diaphragm via the transducer.

With an air bubble, the initial rise in ambient pressure increases the pressure inside the sealed section (amount of gas stays the same, volume decreases). This gets directly transmitted to the main diaphragm, just like a diaphragm reg without the environmental seal.

Once the environmental diaphragm contacts the transducer top, the pressure inside the sealed section stops increasing because the volume is now fixed. Now any increase in pressure gets transmitted via the transducer just like the no bubble scenario. But the pressure transmitted through the transducer is resisted by the higher pressure in the sealed section. This is because pressure acts everywhere, including on the back side of the top of the transducer.

Say you're at 3atm ambient. If your sealed section is at 1atm, then 1atm will be transmitted directly to the main diaphragm and 2atm will be transmitted via the transducer. If your sealed section is at 1.5atm (because of compression of an initially convex environmental diaphragm), then 1.5atm will be transmitted directly to the main diaphragm and 1.5atm via the transducer.

 

[rsingler]

I must concede after trying to diagram it all out, that @Tanks A Lot is absolutely correct. A bubble under the environmental seal does not hurt reg performance.
And I'll give the assist to @CG43 whose comments about the massive gradient between IP and environmental chamber pressure made the light bulb go on.

Coming back to write this, I see that @lowwall wrote the elegantly simple explanation already.

A fun debate, albeit a little hard on the ego.