According to my calculations the Trilateral Flash method not work at all.
Now, I expect to see practical measurements that show that I am completely wrong. I will not change my opinion until I have seen convincing practical measurements.
This might be something also. In this data set the steam mass flow is constant. As you can see the Power decreases when the x-value decreases. These calculations are made with built-in volume ratio =1, which means that flashing/condensation takes place after the opening of the control volume.
I think the paper "Experimental and modeling study on adiabatic two-phase expansion in a Cylinder" by Hiroshi Kanno and Naoki Shikazono" also is worth to study when we talk about two-phase expansion. The time factor is interesting.
Of course you get a torque because of the pressure difference over the expander.
What I mean with "does not work at all" is that the control of the operation is very difficult. Hitherto what I have seen there is a spread of measuring point when the media quality is close to zero.
May I also say that what I have understand as "Trilateral Flash Circle" is expansion from quality x=0. I have seen calculations with x=1 and they have also been called "Trilateral Flash Cycle"!!??
I think the paper "Experimental and modeling study on adiabatic two-phase expansion in a Cylinder" by Hiroshi Kanno and Naoki Shikazono" also is worth to study when we talk about two-phase expansion.
May I also point out the data sheet "Easy method to predict effect of R245fa inlet quality. on th..." is when flashing is 100 %. I.e. immediate. In practice the flashing is not so immediate. It is time dependent, which will reduce the efficiency.