I have several Conshelf yokes that were stamped 4000 psi and many with the 3000 psi. They are exactly the same yoke. There are no dimensions or material change between the two. It is just a change in the times… yokes were considered acceptable for higher pressures at one time (Cousteau used 5000 psi yoke tanks for a while).
IMHO, no diaphragm compares to a Poseidon diaphragm for flow. Take the orifices apart, and you'll see what I mean. But I'm open to others' different experience. Get a used 2960 off EBay and see what I mean. Even the old Poseidon unbalanced diaphragm 300's (Model 2305) down-tuned to 130 psi Intermediate Pressure (at near-empty tank) do better than the Conshelf with a balanced second stage. But military units have used both Poseidon and Conshelf.
That is a very interesting observation and I agree that the Poseidon Cyklon 300 has a large first stage orifice. This does rebound the IP very quickly during the breathing cycle. The trade off is that having a large orifice in an unbalanced first stage, causes a larger IP change as a function of tank pressure change. The reason most non-balanced regulators have a small orifice is to try to limit the intermediate pressure swing as a function of tank pressure change. The larger the orifice in a non-balanced first stage, the more it is affected by tank pressure.
It has been over two decades since I service a Cyklon 300 first stage, but I do recall that the IP does change a lot with the change in tank pressure (back then I didn’t actually wrote down this type of data).
I have several Poseidon Cyklon second stage (I have retired their first stage) and I have paired them up with Conshelf first stages. It is easy to adjust the Conshelf to the higher pressure and I like the consistent performance of a constant IP (independent of tank pressure). It makes it a lot easier to fine tune this regulator combination.
I have serviced a lot of Poseidon regulators (in the early 70’s in my first job at Divers Service Center in PR) and I spent a lot of time fine tuning does regulator. We did have a work bench with a regulator to supply any tank pressure I dial in, but it still was more time consuming than other regulators.
My favorite (single hose) regulator back then was the Poseidon, due to its performance (followed by the Scubapro Mk-5).
Note: I also have a bunch of Scubapro 109 second stages and they are all paired up with Conshelf or Aqua Lung Titan first stages. I find their performance being fantastic, perhaps not as good as a flow through piston, but more than adequate for feeding a couple of second stages at the same time.
In my case one of the major advantage with the Conshelf is the commonality of parts with my Royal Aqua Master, Phoenix Royal Aqua Master, etc.
The only job of a first stage is to try and maintain as constant of an IP as possible. The IP will always dip some in all first stages during the breathing cycle. The only advantage of a “higher flow” first stage is that the IP recovery will be a bit quicker and the total drop will be less. I have tested the Conshelf (and its many derivatives) for many decades and I find them all good to excellent performance.
As I mentioned, the flow through pistons are better, but the Conshelf performance are excellent.
I can’t remember the actual flow rate numbers published for flow through pistons and diaphragm regulators. What I recall is that the flow through pistons are several times the max flow of a diaphragm, but either are like an order of magnitude (10x) the flow that any second stage can use. Those flow rate test are only comparative tests and only give a relative idea of first stage performance during actual operation.
Most manufacturers have switched to a T shaped design for their diaphragm regs. I suspect this allows for a better air flow and gives more room to provide 2 HP ports.
I am almost certain that the reason for the change is for ease of maintenance and assembly. I have assembled many Conshelf with a wooden dowel, but it is a lot easier with the special assembly tool (like the one Herman makes). My observation with the “T” geometry is that it doesn’t require any special assemble tools (but I have limited experience with most of the newer first stages).
I don’t believe that the flow rate is any better with the geometry change, at least nothing significant.