search

usnistgov / REFPROP-issues

A repository solely used for reporting issues with NIST REFPROP
26 stars 13 forks source link

Units are not correct for kinematic viscosity and thermal diffusivity #646

Open Qattussen opened 1 year ago

Qattussen commented 1 year ago

In the Fortran code for Python wrapper, the kinematic viscosity is divided by a factor of 100 to convert it from cm^2/s to m^2/s. This is not correct, because the correct factor would be 10^4. This also affects the thermal diffusivity. I took the refprop interface as a reference. Please check! Attached is a screenshot of the deviation of the Python wrapper to Refprop interface

To reproduce results:

Temperature: 25°C Pressure: 7.9 bara Gas Composition: {'Methane','Ethane','Propane','Butane'} Gas Fraction: {0.0965 0.1667 0.3717 0.3651} ref_state = {'OTH',0,0,0,1.01325} model = {'NIST',1}

ScShot (2)

ianhbell commented 1 year ago

What version of REFPROP? There is a known bug with kinematic viscosity in REFPROP.

Qattussen commented 1 year ago

Version 10.0

heikone commented 1 year ago

Hi, also using this version:

image

Is the known bug already fixed in a later version and how can I get it?

Greetings, Heiko

ianhbell commented 1 year ago

For now my recommendation is to calculate kinematic viscosity yourself as viscosity divided by density. The development version fixes this bug but has regressions that make it not generally suitable.

PieterCornelisse commented 1 month ago

Many thanks for the quick reply. I see now that this issue was reported before. Will wait for the new version to come out then. When will that be?

ianhbell commented 1 month ago

Many thanks for the quick reply. I see now that this issue was reported before. Will wait for the new version to come out then. When will that be?

Sorry I cannot give an estimation. But you can get a development version from me: ian.bell@nist.gov that fixes this issue at least.