# Edit to documentation. State Neumann (natural) boundary are implicitly handled

**URL:** https://community.freefem.org/t/edit-to-documentation-state-neumann-natural-boundary-are-implicitly-handled/1125
**Category:** Feature Request
**Created:** [July 28, 2021, 8:07pm UTC](https://community.freefem.org/t/edit-to-documentation-state-neumann-natural-boundary-are-implicitly-handled/1125 "2021-07-28T20:07:53Z")
**Posts on this page:** 1
**Page:** 1

<div class="post-metadata">

### Author: ![kevinmo08](https://yyz2.discourse-cdn.com/flex030/user_avatar/community.freefem.org/kevinmo08/32/217_2.png) [@kevinmo08](https://community.freefem.org/u/kevinmo08)
#### Post date: [July 28, 2021, 8:07pm UTC](https://community.freefem.org/t/edit-to-documentation-state-neumann-natural-boundary-are-implicitly-handled/1125/1 "2021-07-28T20:07:53Z")

</div>

Hi all,

I’ve seen a few posts in the past asking about implementing Neumann boundary conditions, especially homogeneous neumann boundary conditions in FreeFEM++.

I think it would be helpful to explicitly state somewhere that, since Natural boundary conditions vanish in your variational formulation, FreeFEM++ implicitly assumes any boundaries you don’t label are neumann.

Additionally, it might also help to describe how to implement Robin boundary conditions, where one implements it like:

If Th is a 3D mesh

\partial u / \partial n + g(x,y,z) = 0 on Gamma\_1

You just put in

```auto
int2D(Th,Gamma_1)(v*g(x,y,z))

```