I agree an IP gauge on the end of an inflator hose has possible limitations when it comes to accurately measuring the magnitude of IP swing. The geenral trend and the final stablized IP are fine, but just not the more dynamic aspescts of it.A while back I started a thread about the relationship between IP drop and flow. I got curious when my MK10 and MK15 both experienced a larger IP drop than my MK5 while using the same second stage, same IP gauge, and even same LP inflator hose for the IP gauge. It got to the point where I talked with Peter Wolfinger, the guy that wrote Reg Savvy and was a SP engineer.
Anyhow, I never did get a definitive answer why my regs acted like that, when the MK10 is supposed to have higher flow than the MK5 and the MK15 clearly is a higher flow reg. If you are using the same 2nd stage purge to create the demand, theoretically there should be an inverse relationship between IP drop and flow.
Eventually I tried to really tax the 1st stages by putting 2 balanced/adjustables on them and going full purge (they have really strong purges) on both at the same time. I found that the MK15 initially dropped 15-20 and then leveled off at about 7-10 PSI under IP. The others dropped a full 15-20 under that kind of demand.
In the end I explained the results by guessing that there must be some variance in venturi effect which could lower the pressure in the LP inflator hose in the presence of increased flow through a different LP port. IOW, I don't think it's possible to really accurately measure the IP drop in dynamic flow situations accurately with the set up I have.
BTW, you can really tell a difference between the IP drop with different MK20 pistons. I was working on a really creep-ridden MK20 last summer and eventually tried all three piston styles. The MK25 composite piston has a much smaller IP drop than the other two in the same reg.
Of course none of this makes any difference in dive performance. For example, a relatively med-high performance 1st stage with a maximum flow rate of 150 SCFM would theoretically empty an AL80 in around 30 secs, much faster than the tank valve itself. Second stage flow rates probably max out at about 50 SCFM, and I couldn't imagine any diver, even the most hyperventilating panic-stricken one, drawing more than 10 or 15 SCFM.
An IP gauge on an inline adjuster makes a lot more sense and it pretty much ensures that you are measuring at the second stage. But then again, if the adjuster itself reduces the flow, you can't really see the full impact of maximum flow rates. Still the one Scubatools sells has worked very well for me and does not seem to impede the flow rate much more than the orifice itself.
There is also I think an effect on IP caused by response time. I think the composite Mk 25 piston demonstrates less IP swing as it is much lighter and allows the valve to repond faster to IP drops. In that regard the lubricant used can have an effect as can dirt, salt, etc - or for that matter water rather than air in the ambient chamber as the water is denser with more mass and takes more energy to move and displace as the piston moves back up and down of the seat.
Response time is I think also impacted by piston head diameter in a balanced piston reg. The larger piston head requires less pressure drop to develop the same amount of force to open the valve, and in general a larger piston can maintain a slightly larger distance off the seat at a given IP drop than a smaller piston if the IP drop is identical. So a larger piston diameter equates to greater working range and greater flow, all other things being equal, at an identical IP drop. It is not coincidence that SP went back to the larger piston head size with the Mk 15, and with the Mk 200 and Mk 2 Plus, leaving the Mk 10 and MK 3 as fairly short small head piston experiments.
In an unbalanced piston design a smaller piston head means a larger change in IP as the tank pressure drops, so larger is better on two levels in that situaion.
I think you are probably most accurately displaying the differences at high flow rates where you observed the IP at high sustain flow rates was smaller for the Mk 15 compared to the Mk 5 and Mk 10. Doing the same thing at low rates won't show the differences as the capabilities of the lesser performing regs are not really challenged.