Perhaps your issues relate to the high flow rate and density turndown. Coriolis meters could help with some of the problems. What accuracy is required at each end of the flow spectrum?
Yes the turndown of just under 600:1 is the problem. At first I wanted to go with thermal mass but the steam portion seems to stop the suppliers. At the low end I would like +/- 2% accuracy.
Consider two parallel Coriolis meters of different sizes; and a switching valve to select one or both based upon the flow rate through the smaller meter.
I'm not understanding the problem here. Is the 101.36 kPa (a) upstream or downstream pressure? In either case, the 38% steam would condense at that temperature and pressure and you'd have water and CO2 or water and N2. I have to assume that the only pressure you mentioned is downstream and you have some room to play with pressure upstream (without that assumption, this problem has no solution).
Since you posted it in Measurement, I'll assume you are trying to measure this stuff with a 600:1 volume turndown. The only way I know of to get that kind of range is with a critical flow nozzle--use something like a V-cone or a Taylor Choke in front of a critical flow nozzle and build a table for the RTU that XX press upstream of the nozzle is YY kg/hr. Your +/- 2% with 600:1 volume turndown is quite doable.
The only way you can have 100% vapour is if the pressure is 60 kpa or the temperature is 190 C and then it may not be that because of changes in composition. The data you presented is two phase flow and the liquid phase will not have a pH of 7.0.
The only device that will measure this is a coriolis meter opr better yet two meters as JLSeagull recommended.
