Robin BC 'c' vs flux BC 'f'

Bibhas Kumar

unread,

Nov 13, 2025, 4:50:51 PM (8 hours ago) Nov 13

to Nek5000

Hi everyone,
I hope you all are doing great.

I have a quick confusion regarding the implementation of the variable flux boundary condition for passive scalars (ps1), flux = value*ps(1).

The way I have done it in the past is using 'f' in usrdat2() and defining flux = value* ps(1) in userbc(), hence flux is dependent on local ps(1) value.

1. Is it all right to do so? using flux = f(ps(1)) instead of a fixed value.

2. Or is it just for fixed flux values? flux = fixed, can not be a f(ps(1)).

If not, how is it different from 'c', where the right side is defined as flux = value * ps1.

3. What is the right way to use the robin BC (c)? after defining 'c' in the usrdat(). What goes in userbc()?

Thanks,
Bibhas

Shaver, Dillon R.

unread,

Nov 13, 2025, 5:18:49 PM (7 hours ago) Nov 13

to Bibhas Kumar, Nek5000

Bibhas,

See the documentation (https://nek5000.github.io/NekDoc/problem_setup/BCs.html#newton-cooling-convection-c).

For 'c ' boundaries, the flux variable is not used in userbc, only the htc and Tinf variables. If you need the flux to be a general function of T (or ps1), use the 'f ' BC. If you need it to be the specific Newton cooling function, use the 'c ' BC.

Thanks,

Dillon

From: nek...@googlegroups.com <nek...@googlegroups.com> on behalf of Bibhas Kumar <bibha...@gmail.com>
Sent: Thursday, November 13, 2025 3:50 PM
To: Nek5000 <nek...@googlegroups.com>
Subject: [nek5000] Robin BC 'c' vs flux BC 'f'

This Message Is From an External Sender

This message came from outside your organization.

--
You received this message because you are subscribed to the Google Groups "Nek5000" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nek5000+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/nek5000/7543b78d-139c-4a0c-a9bd-4903fd6cdd0bn%40googlegroups.com.