diff --git a/doc/src/fix_add_heat.rst b/doc/src/fix_add_heat.rst
index 5b67bd4e15..6d192fd0b6 100644
--- a/doc/src/fix_add_heat.rst
+++ b/doc/src/fix_add_heat.rst
@@ -18,11 +18,11 @@ Syntax
 
        *constant* args = *rate*
          *rate* = rate of heat flow (energy/time units)
-       *linear* args = :math:`T_{target}` *k*
-         :math:`T_{target}` = target temperature (temperature units)
+       *linear* args = *Ttarget* *k*
+         *Ttarget* = target temperature (temperature units)
          *k* = prefactor (energy/(time*temperature) units)
-       *quartic* args = :math:`T_{target}` *k*
-         :math:`T_{target}` = target temperature (temperature units)
+       *quartic* args = *Ttarget* *k*
+         *Ttarget* = target temperature (temperature units)
          *k* = prefactor (energy/(time*temperature^4) units)
 
 * zero or more keyword/value pairs may be appended to args
@@ -45,9 +45,9 @@ Examples
 Description
 """""""""""
 
-This fix adds heat to particles with the temperature attribute every timestep.
-Note that this is an internal temperature of a particle intended for use with
-non-atomistic models like the discrete element method.
+This fix adds heat to particles with the temperature attribute every timestep
+at a given rate. Note that this is an internal temperature of a particle intended
+for use with non-atomistic models like the discrete element method.
 
 For the *constant* style, heat is added at the specified rate. For the *linear* style,
 heat is added at a rate of :math:`k (T_{target} - T)` where :math:`k` is the
@@ -96,7 +96,8 @@ only enabled if LAMMPS was built with that package.
 See the :doc:`Build package <Build_package>` page for more info.
 
 This fix requires that atoms store temperature and heat flow
-as defined by the :doc:`fix property/atom <fix_property_atom>` command.
+as defined by the :doc:`fix property/atom <fix_property_atom>` command or
+included in certain atom styles, such as atom_style rheo/thermal.
 
 Related commands
 """"""""""""""""
diff --git a/src/GRANULAR/fix_add_heat.cpp b/src/GRANULAR/fix_add_heat.cpp
index 01bc22cdf4..8d6bfcdc36 100644
--- a/src/GRANULAR/fix_add_heat.cpp
+++ b/src/GRANULAR/fix_add_heat.cpp
@@ -21,7 +21,6 @@
 #include "error.h"
 #include "input.h"
 #include "memory.h"
-#include "update.h"
 #include "variable.h"
 
 using namespace LAMMPS_NS;
@@ -125,7 +124,6 @@ void FixAddHeat::post_force(int /*vflag*/)
   int *mask = atom->mask;
   double *heatflow = atom->heatflow;
   double *temperature = atom->temperature;
-  double dtinv = 1.0 / update->dt;
 
   if (vstyle == ATOM) {
     if (atom->nmax > maxatom) {
@@ -142,7 +140,7 @@ void FixAddHeat::post_force(int /*vflag*/)
       if (mask[i] & groupbit)
         heatflow[i] = 0.0;
 
-  double vtmp, dt;
+  double vtmp;
   if (vstyle == CONSTANT) vtmp = value;
   if (vstyle == EQUAL) vtmp = input->variable->compute_equal(var);
   for (int i = 0; i < atom->nlocal; i++) {
@@ -150,11 +148,11 @@ void FixAddHeat::post_force(int /*vflag*/)
       if (vstyle == ATOM) vtmp = vatom[i];
 
       if (style == ADD) {
-        heatflow[i] += dtinv * vtmp;
+        heatflow[i] += vtmp;
       } else if (style == LINEAR) {
-        heatflow[i] += dtinv * prefactor * (vtmp - temperature[i]);
+        heatflow[i] += prefactor * (vtmp - temperature[i]);
       } else if (style == QUARTIC) {
-        heatflow[i] += dtinv * prefactor * (pow(vtmp, 4.0) - pow(temperature[i], 4.0));
+        heatflow[i] += prefactor * (pow(vtmp, 4.0) - pow(temperature[i], 4.0));
       }
     }
   }