diff --git a/mne/viz/tests/test_utils.py b/mne/viz/tests/test_utils.py
index 8cb99d5224f..5f8932c9219 100644
--- a/mne/viz/tests/test_utils.py
+++ b/mne/viz/tests/test_utils.py
@@ -98,7 +98,8 @@ def test_auto_scale():
 
     for inst in [raw, epochs]:
         scale_grad = 1e10
-        scalings_def = dict([('eeg', 'auto'), ('grad', scale_grad)])
+        scalings_def = dict([('eeg', 'auto'), ('grad', scale_grad),
+                             ('stim', 'auto')])
 
         # Test for wrong inputs
         assert_raises(ValueError, inst.plot, scalings='foo')
diff --git a/mne/viz/utils.py b/mne/viz/utils.py
index 62312cebdd4..8eeb7a15fc4 100644
--- a/mne/viz/utils.py
+++ b/mne/viz/utils.py
@@ -1058,7 +1058,6 @@ def _compute_scalings(scalings, inst):
         A scalings dictionary with updated values
     """
     from ..io.base import _BaseRaw
-    from ..io.pick import _picks_by_type
     from ..epochs import _BaseEpochs
     if not isinstance(inst, (_BaseRaw, _BaseEpochs)):
         raise ValueError('Must supply either Raw or Epochs')
@@ -1066,11 +1065,12 @@ def _compute_scalings(scalings, inst):
         # If scalings is None just return it and do nothing
         return scalings
 
-    ch_types = _picks_by_type(inst.info)
-    unique_ch_types = [i_type[0] for i_type in ch_types]
+    ch_types = channel_indices_by_type(inst.info)
+    ch_types = dict([(i_type, i_ixs)
+                     for i_type, i_ixs in ch_types.items() if len(i_ixs) != 0])
     if scalings == 'auto':
         # If we want to auto-compute everything
-        scalings = dict((i_type, 'auto') for i_type in unique_ch_types)
+        scalings = dict((i_type, 'auto') for i_type in ch_types.keys())
     if not isinstance(scalings, dict):
         raise ValueError('scalings must be a dictionary of ch_type: val pairs,'
                          ' not type %s ' % type(scalings))
@@ -1096,15 +1096,13 @@ def _compute_scalings(scalings, inst):
         data = inst._data
     if isinstance(inst, _BaseEpochs):
         data = inst._data.reshape([len(inst.ch_names), -1])
-
     # Iterate through ch types and update scaling if ' auto'
     for key, value in scalings.items():
         if value != 'auto':
             continue
-        if key not in unique_ch_types:
+        if key not in ch_types.keys():
             raise ValueError("Sensor {0} doesn't exist in data".format(key))
-        this_ixs = [i_ixs for key_, i_ixs in ch_types if key_ == key]
-        this_data = data[this_ixs]
+        this_data = data[ch_types[key]]
         scale_factor = np.percentile(this_data.ravel(), [0.5, 99.5])
         scale_factor = np.max(np.abs(scale_factor))
         scalings[key] = scale_factor