Question How much of an IP drop is acceptable?

[Tanks A Lot]

For what it's worth, halocline is absolutely right with what he says.

The venturi effect is noticeable, depending on what regulator and what ports selected. I have played around with it in the past by inserting small flow restrictions between the LP port and LP hose. The smaller the aperture, the less the IP was affected at the end of the hose.
Take a MK21 for example. Having a second stage and IP gauge on two adjacent ports, the drop in IP was smaller than when selecting two opposing ports.
That being said, the effect is really small. As for how applicable this is for the scenario here I very much doubt it. I was never able to see a difference of more than 0.3 bar that I could attribute to the venturi effect, but again, different ports and configurations make a difference.

I do not think that the venturi effect is of any significance to this particular scenario.

 

[tursiops]

Some questions:
1) Where is he flow restriction that is essential to generating the Venturi effect?
2) Flow past an orifice is not sufficient for there to be a Venturi effect; there must also be a flow restriction to cause a speed increase in the flow and thereby generate the pressure drop.
3) There may be confusion here with the Bernoulli effect, in which a flowing fluid contains some energy (proportional to the fluid velocity squared), and thus has some pressure drop on the wall of the flow container in order to have conservation of energy.
4) All these "laws" are derived for incompressible laminar flow fluids, and an be used only in concept for gases in a turbulent flow, as in a regulator.
5) My point is that the flow past the orifice is not fast, so the pressure on the inner walls of the reg is not reduced much due to the flow.
6) Even if all the laws applied, and the flow were (say) 10 ft/sec, the pressure on a 3.8-inch diameter section of the reg wall would only be about 0.2 psi, not enough to worry about for the IP measurement.

 

[Tanks A Lot]

These are excellent questions that I must admit I do not have the answers to. Thinking about it, venturi effect may not be the appropriate answer to what I measured to begin with, or it may be, depending on how one looks at it.

1) That gave me food for thought. I always assumed that the opening of the LP outlet was an orifice. You mentioning it specifically, doesn't leave me all so sure about this though.
2) Absolutely right! Here was my train of thought, which I'm not sure anymore is correct anymore:

• The gas expanded past the seat and is now flowing towards the second stage.
• The gas that flows towards the second stages passes one of the other LP ports (This depends heavily on configuration).
• The other LP port is acting as an orifice and the air past it gets drawn towards the air flow.

Now the problem I see after I think more about it, is that this second LP port orifice is not enclosed in the loop, as would be in a traditional venturi-effect scenario.
3) This is not applicable to the drop in IP, as the gas in a traditional at an LPI attached IP gauge shouldn't be flowing. With inline gauges, this may be of interest, but as you mention the effect is negligible.
5) I'm not entirely following here, are you speaking about the orifice of the seat/orifice combo, where pressure dropped from cylinder pressure to IP?

That being said, I'm certain that the flow of the air effects IP even measured at an LPI. What I did in the past was essentially making the outlet which goes to the LPI where my IP gauge sit smaller smaller by using some gaskets. The smaller I made the outlet, the more stable IP was. Again, the drops were tiny and I used a class 1 digital gauge to even measure it.

Mares uses their DFC to pull the diaphragm down. One could think about that little hole of the DFC towards the diaphragm as similar to a LP port.

I'm sure there is something more to it and I get this wrong somewhere, but what I measured did make intuitively sense to me as well. Thanks for making me think about this again and I look forward to learn why this is indeed happening. The obvious solution - ventui-effect - doesn't look all that cut and dry to me anymore.

 

[rsingler]

Here's some data to help us keep theorizing.
I just put a Mk25 and a Mk11 (with HFP and regular LP) on the bench.
I ran 8 SCFM through all regs, with a supply downregulated to 1000 psi, so falling supply tank pressure wouldn't change the results.

I don't know the Physics principle, as it doesn't seem to be Venturi, but my working hypothesis was that a measurement port pointing toward flow might see a higher pressure due to impact of onrushing molecules than a port pointing away from flow, as entrained gas in the measurement port is sucked into the flow.

My results were much as @Tanks A Lot suggested above.

Mk25:
Static IP: 131.2 @ 1000 psi supply
Reg attached to a lateral turret LP.
IP @ 8 SCFM:
IP gauge on adjacent port : 121.8
IP gauge on opposite port: 121.1
IP gauge on end port (facing piston shaft): 122.1

Mk11 data to follow...

 

[rsingler]

For the Mk11, the reg always pulled from a HFP.
As you can see, the HFP gets its jet of gas directly adjacent to the channel below the diaphragm.

The regular LP gets gas down the special channel adjacent to the HFP. Therefore, I theorized that HFP measurements might see a higher value due to "piling up" of molecules aimed directly at the port, while the LP might see a lower pressure as molecules are entrained in the flow path to an adjacent regulator. As for an opposing port, I had no idea.

Here's the data:
Static IP @ 1000 psi: 130.2
Reg pulling from HFP at 8 SCFM.
IP at opposite HFP: 123.3
IP at opposite LP: 124.0
IP at adjacent LP: 122.7

 

[rsingler]

As others have suggested, the measurement error is real, but small.
I'll let someone else tell me what the principle is called, but the Mk25 data seems to show that even if pressure is dropping from the static ~130 psi, dynamic IP measured at an end port is slightly higher than either an adjacent or opposed lateral port, suggesting that flow impact on the column of gas going to the gauge is raising relative pressure.
The opposite appears to be in play with opposing lateral ports, where gas is "pulled" from the measurement port due to the downrushing flow of gas to the opposite LP. An intermediate effect is seen at an adjacent port, perhaps due to local turbulence.

In the Mk11, we can imagine something similar.
The gas flow to the reg at the HFP entrains gas molecules to the extent that the immediately adjacent LP showed the lowest measured dynamic IP.
On the opposite side of the reg, flow was more turbulent, but even then, the HFP opposite the outrushing gas showed a lower dynamic IP than the far distant LP on the opposite side.

So what's that called? Seems to be a Venturi-like effect, even if I can't identify an "orifice".

And BTW, please ignore the high dynamic inhalation resistance on the magnehelic. I just grabbed that second stage from the drawer. It was one quickly reassembled after a class with a slight freeflow, so I cranked on the adjustment knob to seal it, hence the high effort @ 8 SCFM. Properly tuned, that Deep6 X usually pulls about one inch at that flow.

And incidentally, we see a dynamic IP drop that is more in keeping with what the OP expected when he created this thread: 7-11 psi dynamic fall in IP at a substantial flow. With a full tank instead of 1000 psi, it might have been a little smaller. With more molecules at depth and 5x gas density it might have been a little larger. But certainly not 20-30 psi.

 

[mycoco]

Does anyone know why most of reglator sets IP as 9.5bar/137psi?
Is ita standard or safty requirement?
If the IP set maybe 8.5bar or 7.5bar, diver can still breath easily?(of course 2nd stage need also set as same crack effort).
I am just curious, why 9.5bar/137psi

 

[Nemrod]

Does anyone know why most of reglator sets IP as 9.5bar/137psi?
Is ita standard or safty requirement?
If the IP set maybe 8.5bar or 7.5bar, diver can still breath easily?(of course 2nd stage need also set as same crack effort).
I am just curious, why 9.5bar/137psi

The IP is not always 137 psi across the board for all regulators, brands and models. Current Scubapro product for the balanced first stages calls for a static IP of 133 to 142 psi at 230 bar (3336 psi). Such a range is compatible with and within the adjustment range of their downstream and balanced second stages.

 

[mycoco]

The IP is not always 137 psi across the board for all regulators, brands and models. Current Scubapro product for the balanced first stages calls for a static IP of 133 to 142 psi at 230 bar (3336 psi). Such a range is compatible with and within the adjustment range of their downstream and balanced second stages.

Yes, IP is not always stays at 137psi for sure.
My question is how does it setup the pressure range?
Maybe 80 or 100psi can supply the breath demand, why almost every brand setup around 130-140psi as the IP standard

 

[Mobulai]

Awesome analysis everyone so far (I am still on page 2, still reading the rest -- but I think I had an answer for this one), and thank @rsingler for running the exp.s and providing data -- methodical as always

So what's that called? Seems to be a Venturi-like effect, even if I can't identify an "orifice".

I won't claim more than a pretty good guess not a definitve answer, I only know of this phenomena (or had to deal with it's complex analysis) and understand in context of diesel injection: much higher pressure range (1.5K-4.4Kbar) -- where everything can feel like an orifice at the speeds and flowrates incorporated -- and, the fluid is incompressimble (art these ranges)
it doesn't even have a "formal" name [that I know of -- ask reg designers]:
Pressure Wave Analysis (I would mostly say "running a 0d/1d/2d pressure wave analysis for blabla")

The weird thing is, and that is a bit paradoxial, the ration of variation of Re number dictates how negligibel it is in making the calculation feisable or not (computationally), if we talk tiny relative drops in comparison to great supply/working pressure your equations get simpler; as Re# variance is small, if we are talking losing 1/3 of our WP, bummer, more differential terms; luckily wan't the case for me (mostly)
So to an extent, it is a Venturi dependant phenomena, at least highly impacted by it

An intresting tool/hqck here is actually building an analogus model of the "chambers" and "passages" and "orifices" to other physical systems that are.. ehem, without hardcore CFD computing power, well analytically/numerically impossible.. similar and simpler to compute analytically/symbollically
So the "chamber" is like a capacitor or a spring, a passage is a resistor, something like a turbine/propeller/ poppet in a solenoid pilot driven injector might be considerd an indcutor; battery for tank, molocules fo gas for coloumbs and masss flowrater for current add some tears over partial diffrentials maths depending on the model detail you want; et voila 1st/2nd/3rd order equations are born in a simpler way and you avoided a costly (or inacessible) CFD run/computers.

I have mostly run 1D (2nd order P.D.E.s) versions of this for design work, 2D is a true nightmare (to just code knowing it would take a machine ages to figure it out, let alone derive the model symbollycally) as this is enough for such a case (diesel inj sys); I only had to do 2D once for an air intake with AirMass sensor housing (that's a really simple construction btw compared to valves) for a grad class seminar and I kinda swore never again (but diameters here are drastically bigger so yet again very far dynamic ranges than a reg)

I suspect for regulators... hmm, I bet with compressible gas at lower ranges of pressure.. 2D?
I assume, and suspect even more (dangerous territory): it's like waves/tide puping in/out at the cliff/cave/cavern that's at sea, or how reverbration of sound works or even how just snuffed candles can suck the flame through the plume of smoke close to them and reignite:
The closer a disturbance (press. diff.) occurs somewhere (a drop), the more it has an impact on the closest points to it (with some kind of positive feedback amplfying it even); like how a tiny change in a viva knob (I am thinking old g250) while tuning orifice/lever height has a wild impact on freeflow sheer violence when trying to "ride" the edge of the reg

oh kinda reminds me.. sidebar/ADHD.. dear Dr. Rob, how do I tune orifice and VIVA on a g250 and not chase my tail itteratively? I feel like a headless chicken chasing that 3sec purge standard (I might better ask in a different thread, tomorrow maybe)​

Bigger defferntials in pressure (like Turret head port/opposite to piston/HFP on 11/17/19) would amplify, a bigger(CS Area to Volume ration) "chamber" also would be stiffer and this might dampen the impact of IP drop measuring from there (?)....

I suspect also that's the source of magic for the "swirl" action, be that in a flowthru piston knife egde / cone seat, or in an engine aspirating fuel in a way that it burns as it moves through the piston chamber (but that'S treading on flashpoints and ignition by compression so other factors at play her as well)

"The great Venturi an inline gauge or flowmeter would exprience is the most violent, and the further away fro the path of air the more dampend it is (stifness, resistance)" fits in my head with "the pressure rate diff (wave), the voltage drop... as in an AC circuit"
IDK I feel like I am getting too esoteric now

Maybe not (to some/all these guess), I haven't fallen that deep down the rabbit hole of pressure waves and complex CFD as I moved on to other things soon after (this mainly taught me good enough coding skills to stop doing it completely lol)

probably that's something reg designers can tell us (or maybe LPG world?) -- I will keep waiting for an answer but not to holding my breath (tho I knida am, and wanna know)

Funny, I got here as I am seeing something similar to OP's issue but opposite on a H-75P (mk25); currently measuring at 150bar, longhose opposite Piston for SS (g250v @2.7cmWater), BC hose (adjacent port) IP guage drop on inhalation (to 115 PSI; It sits at 125, but drifts withinin 5 mins to 135, 30 mins to ~145PSI, 1 hour to 150PSI, currently testing overnight); for sure needs to be opened up and is gunky -- but that will have to wait a while, just a pre rebuild check

EDIT: oh It's just page3; ok I haven't missed much after