yade-users team mailing list archive

[Robert Caulk <question699421@xxxxxxxxxxxxxxxxxxxxx>]

Question #699421 on Yade changed:
https://answers.launchpad.net/yade/+question/699421

    Status: Open => Answered

Robert Caulk proposed the following answer:
Hey Zoheir,

This "flow.decoupleForces" was not designed to be used directly in this
manner. It is meant to be controlled by ThermalEngine with
thermal.letThermalRunFlowForceupdates. The problem is that
thermal.letThermalRunFlowForce does not give you the functionality you
desire because it still computes and applies the forces. I probably
shouldn't have exposed flow.decoupleForces to the user in this way. But
it is ok, your use case makes sense. So it is just a matter of fixing it
so that decoupleForces can be used in this way.

After diving in, nothing is actually broken for advection, it is running
normally behind the scenes. The issue is that the "solver->noCache" flag
is well embedded into various FlowEngine functionalities, such as the
getter functions you use (getPoreTemperature()). When you use
flow.decoupleForces directly, the noCache flag is never set properly,
and hinders the use of the getters.  Here is a merge request which fixes
your problem and allows no-remesh sims to still use getters [3]. Please
confirm if it works for you or not.

>>Another question is where cell temperature is updated in the source
code?

https://gitlab.com/yade-
dev/trunk/-/blob/master/lib/triangulation/FlowBoundingSphereLinSolv.ipp#L826

>>I think cell->info().temp() is not updated when decoupleForces is
true.

Can you link to where in the source code you see this?

Cheers,

Robert

[1]https://gitlab.com/yade-dev/trunk/-/blob/master/pkg/pfv/FlowEngine.ipp.in#L105
[2]https://gitlab.com/yade-dev/trunk/-/blob/master/lib/triangulation/FlowBoundingSphere.hpp#L255
[3]https://gitlab.com/yade-dev/trunk/-/merge_requests/799

-- 
You received this question notification because your team yade-users is
an answer contact for Yade.