[TRNSYS-users] Calculating properties for ethylene glycol water

[RUBIO RUBIO, JOSÉ JAVIER]

Jeff,

You are absolutely right. In fact, I did some time ago a test in a similar way as you said and the result was practically the same because the temperature range in my simulation is small and the properties barely change.

Thanks,

Regards,

Jose Javier.

De: Jeff Thornton <thornton at tess-inc.com>
Enviado el: lunes, 21 de junio de 2021 17:00
Para: TRNSYS users mailing list at OneBuilding.org <trnsys-users at lists.onebuilding.org>
CC: RUBIO RUBIO, JOSÉ JAVIER <josejavier.rubio at edu.upct.es>
Asunto: Re: [TRNSYS-users] Calculating properties for ethylene glycol water

José,

Before you decide to code your own model to calculate the properties for Ethylene Glycol I want to remind you of a few things that many users forget.  If you use variable fluid properties then every model that that fluid flows into will need to be re-coded; pipes, tanks, valves etc.  Otherwise you will run into major energy balance issues in your system.  And it’s not simply just moving the properties in those models from parameters to inputs - it’s the fundamental governing differential equations for the models that also change. With density now a function of temperature, the mass of the fluid changes in a pipe or tank forca given volume - which means you have to re-derive the entire solution.  And beyond that you have to now code special controllers to move the fluid.  Imagine a case where a tank of water is just sitting there and slowly losing energy to the environment.  Usually that’s a pretty simple case to solve.  But now as the fluid cools, the density increases and the mass in the tank now takes up less volume - so water will enter the tank to take up the newly freed volume.  Is this water coming from the pipe leading to the tank, the pipe leaving the tank?  In most systems it comes from an expansion tank and flows into the tank from either, or possibly even both, pipes...

At TESS, we have a series of models that incorporate the change in fluid properties with temperature as part of our high temperature solar library.  But long before you decide to go down the variable property route you should ask yourself if the small change in fluid property accuracy is worth the major changes you’ll need to make.

Run a simulation with constant “cold” fluid properties and then one with constant “hot” properties.  How different are the results - likely not very different at all.

Jeff

Sent from my iPhone


On Jun 21, 2021, at 6:19 AM, RUBIO RUBIO, JOSÉ JAVIER via TRNSYS-users <trnsys-users at lists.onebuilding.org<mailto:trnsys-users at lists.onebuilding.org>> wrote:

Dear TRNSYS users,

Anyone knows some way for calculating the properties of the ethylene glycol water (density, specific heat and thermal conductivity) by using the temperature as an input? I have thought in creating a new type in Simulation Studio to code it in FORTRAN, which would call REFPROP to calculate the properties (although REFPROP doesn’t have ethylene glycol water in its database), but I don’t really know how to do it properly.

Thanks in advance.

Regards.

Jose Javier.


_______________________________________________
TRNSYS-users mailing list
TRNSYS-users at lists.onebuilding.org<mailto:TRNSYS-users at lists.onebuilding.org>
http://lists.onebuilding.org/listinfo.cgi/trnsys-users-onebuilding.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.onebuilding.org/pipermail/trnsys-users-onebuilding.org/attachments/20210621/596bdcc8/attachment.htm>