diff --git a/doc/src/Howto_bioFF.rst b/doc/src/Howto_bioFF.rst
index c5cd8f1f0b..35aa609e5a 100644
--- a/doc/src/Howto_bioFF.rst
+++ b/doc/src/Howto_bioFF.rst
@@ -20,18 +20,18 @@ The `CHARMM force field <https://mackerell.umaryland.edu/charmm_ff.shtml>`_ :ref
 
 .. math::
 
-  V = & \sum_{bonds} E_b + \sum_{angles} \!E_a + \!\overbrace{\sum_{dihedral} \!\!E_d}^{\substack{
+  V & = \sum_{bonds} E_b + \sum_{angles} \!E_a + \!\overbrace{\sum_{dihedral} \!\!E_d}^{\substack{
          \text{charmm} \\
         \text{charmmfsw}
-      }} +\!\! \sum_{impropers} \!\!\!E_i +  \!\!\!\!\!\!\overbrace{\sum_{pairs} \left(E_{LJ}+E_{coul}\right)}^{\substack{
+      }} +\!\!\! \sum_{impropers} \!\!\!E_i \\[.6em]
+      & \quad + \!\!\!\!\!\!\!\!\!\!\underbrace{~\sum_{pairs} \left(E_{LJ}+E_{coul}\right)}_{\substack{
          \text{lj/charmm/coul/charmm} \\
         \text{lj/charmm/coul/charmm/implicit} \\
         \text{lj/charmm/coul/long} \\
         \text{lj/charmm/coul/msm} \\
          \text{lj/charmmfsw/coul/charmmfsh} \\
         \text{lj/charmmfsw/coul/long}
-      }} \\[0.6em]
-      & \qquad  \!\!\!\!+ \sum_{special}\! E_s + \sum_{residues} \!\text{CMAP}(\phi,\psi)
+      }} \!\!\!\!\!\!\!\!+ \!\!\sum_{special}\! E_s + \!\!\!\!\sum_{residues} \!\!\!{\scriptstyle\mathrm{CMAP}(\phi,\psi)}
 
 
 The terms are computed by bond styles (relationship between 2 atoms), angle styles (between 3 atoms) , dihedral/improper styles (between 4 atoms), pair styles (non-covalently bonded pair interactions) and special bonds. The CMAP term (see :doc:`fix cmap <fix_cmap>` command for details) corrects for pairs of dihedral angles ("Correction MAP") to significantly improve the structural and dynamic properties of proteins in crystalline and solution environments :ref:`(Brooks) <howto-Brooks>`. The AMBER force field does not include the CMAP term.
@@ -112,6 +112,10 @@ These styles are used by LAMMPS input scripts generated by `charmm-gui.org <http
    <pair_charmm>` and :doc:`dihedral charmm <dihedral_charmm>` doc
    pages.
 
+.. note::
+
+  TIP3P water model MUST be used with CHARMM force field not TIP4P, TIP5P or SPC. In fact, `"using the SPC model with CHARMM parameters is a bad idea" <https://matsci.org/t/using-spc-water-with-charmm-ff/24715>`_ and `"to enable TIP4P style water in CHARMM, you would have to write a new pair style" <https://matsci.org/t/hybrid-pair-styles-for-charmm-and-tip4p-ew/32609>`_ . LAMMPS input scripts generated by Solution Builder on charmm-gui.org use TIP3P molecules for solvation. Any other water model can and probably will lead to false conclusions.
+
 COMPASS
 -------