PropsSI failed ungracefully with Water::IF97

[jccberthet]

Description

PropsSI failed ungracefully, with IF97::Water

Steps to Reproduce

import CoolProp.CoolProp as cp
cp.PropsSI('Q', 'P', 6.4743994800040303e+06, 'H',1.2394090698400016e+06 , 'IF97::Water')

Expected behavior:

Expected to return zero, or something indicating that the fluid in fully liquid.

Actual behavior:
Raised an exception:
ValueError: PropsSI failed ungracefully :: inputs were:"Q","P",6.4743994800040303e+06,"H",1.2394090698400016e+06,"IF97::Water"; please file a ticket at https://github.com/CoolProp/CoolProp/issues

Versions

CoolProp Version: 6.4.1
Operating System and Version: Windows 10
Access Method: Python

Additional Information

[ibell]

Is that statepoint valid? And the only problem is the value that is returned for Q? How about density?

[henningjp]

Other props return values in the liquid phase, Q is undefined here. This point is just outside the vapor dome in the liquid region. HEOS returns Q = -1 (i.e.. undefined) and does not throw an error. Nudging the IF97 enthalpy value up to 1239.795 returns a sensible Q value (Hsat is 1239.7942 at this pressure).

I think the point here is that IF97 should return something other than an ungraceful failure if Q is undefined.

Workaround: The calling code should check to make sure H is inclusively between the saturated liquid value and the saturated vapor value before making the call.

[ibell]

@henningjp can you add an exception for Q < 0 or Q > 1 for surface tension?

[henningjp]

Yes, once I get through this current project. I’ll tag it. Should be a quickie.

…

On Sat, Feb 20, 2021 at 11:52 AM Ian Bell ***@***.***> wrote: @henningjp <https://github.com/henningjp> can you add an exception for Q < 0 or Q > 1 for surface tension? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#1999 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AECSHMGIEWJDNH3PQDCM5MDS77SFFANCNFSM4XQUXJQQ> .

[ibell]

Thanks, yes I should think so.

[henningjp]

@ibell Finally getting back to this issue. Was intending to modify IF97 to make Q return an undefined value (-1) in all single-phase regions; like HEOS behavior. However, your request above has me confused as you mentioned surface tension. Did we not address surface tension in #1755 ? Not sure how surface tension relates here. Unfortunately, it was so long ago, you may not remember either! 😕 I'll proceed with the original issue above at risk until apprehended.

[henningjp]

@ibell - Maybe you meant, "like" for surface tension?

[henningjp]

NVM. Surface Tension error thrown is just wrong for IF97 backend when Q < 0 or Q > 1. Fixed.
_Q issue is a little less trivial than I thought. Working on that.

[ibell]

I haven't the foggiest about what is meant about surface tension. I think that it should not be allowed for Q < 0 or Q > 1?