Temperature with HmassP_INPUTS with twophase fluid and tabular

[JonWel]

This input pair works to compute temperature with HEOS backend, but not with the tabular ones:

import CoolProp

fluid = "R245fa"
input1 = 4e5 # Hmass
input2 = 8e5 # P
output = CoolProp.get_param_index("T")
input_pair = CoolProp.get_input_pair_index("HmassP_INPUTS")

handle = CoolProp.AbstractState_factory("HEOS", fluid)
CoolProp.AbstractState_update(handle,input_pair,input1,input2)
println("HEOS    gives: ", CoolProp.AbstractState_keyed_output(handle,output))
CoolProp.AbstractState_free(handle)

handle = CoolProp.AbstractState_factory("TTSE&HEOS", fluid)
CoolProp.AbstractState_update(handle,input_pair,input1,input2)
println("TTSE    gives: ", CoolProp.AbstractState_keyed_output(handle,output))
CoolProp.AbstractState_free(handle)

Output:

HEOS    gives: 353.7035584315394
ERROR: CoolProp: no correct state has been set with AbstractState_update
 in AbstractState_keyed_output

Asking for other properties with the same update works (tested with density and entropy).
I think it may be related to the fact that the searched point is in two state area, so with a constant temperature.

[Astrinus]

Which version are you using? With the latest stable (5.1.1) for me works well either with BICUBIC and TTSE.

>> tmp = CoolProp.AbstractState.factory('BICUBIC&HEOS','R245fa');
>> tmp.update(CoolProp.HmassP_INPUTS,4e5,8e5)
>> tmp.T()
    ans = 353.703558431357
>> tmp = CoolProp.AbstractState.factory('TTSE&HEOS','R245fa');
>> tmp.update(CoolProp.HmassP_INPUTS,4e5,8e5)
>> tmp.T()
    ans = 353.703558431357

[JonWel]

I used a build from latest GitHub sources.
Neither generating new tables nor a new source build did help.

[ibell]

I wonder if there is something funny in the high-level wrapper of the low-level interface. Maybe its a bug.

[ibell]

Given #777, I bet this is related to something curious on your side. Anyone else on linux that can confirm this behavior? @jowr?

[jowr]

I am out of office until September and have only limited internet access, feel free to remind me again later.

My immediate guess would be to make sure that there is no broken library file lurking anywhere in the search path.

[JonWel]

I was experiencing this firstly on Windows XP (My testing code for ExternalMedia set state with p and h and calculate most of the outputs but was stuck on temperature).
I did test this code again with the latest source code on Linux and still have the same error (even with the latest nightly build for Linux 64: libCoolProp.so.5.1.2dev).
If I remembered well, the same code searching for temperature in monophasic area worked.

[ibell]

Should be fixed now - tested on linux in C++

[JonWel]

👍