diff --git a/energytool/base/parse_results.py b/energytool/base/parse_results.py
index 0121889..00dbe7b 100644
--- a/energytool/base/parse_results.py
+++ b/energytool/base/parse_results.py
@@ -141,7 +141,9 @@ def get_output_variable(
     results = eplus_res.loc[:, mask]
 
     if drop_suffix:
-        new_columns = [re.sub(f":{variables}.+", "", col) for col in results.columns]
-        results.columns = new_columns
+        strip_pattern = re.compile(
+            ":(?:" + "|".join(re.escape(v) for v in variable_names_list) + ").+"
+        )
+        results.columns = [strip_pattern.sub("", col) for col in results.columns]
 
     return results
diff --git a/energytool/system.py b/energytool/system.py
index 53cc435..fc8f29c 100644
--- a/energytool/system.py
+++ b/energytool/system.py
@@ -244,7 +244,9 @@ class Sensor(System):
         (list of strings). Default is '*' meaning all the available variables.
     """
 
-    def __init__(self, name: str, variables: str, key_values: str | list[str] = "*"):
+    def __init__(
+        self, name: str, variables: str | list[str], key_values: str | list[str] = "*"
+    ):
         super().__init__(name=name, category=SystemCategories.SENSOR)
         self.variables = variables
         self.key_values = key_values
@@ -257,13 +259,16 @@ def pre_process(self, idf: IDF):
         )
 
     def post_process(self, idf: IDF = None, eplus_results: pd.DataFrame = None):
-        results = get_output_variable(
-            eplus_res=eplus_results,
-            key_values=self.key_values,
-            variables=self.variables,
-        )
-        results.columns = results.columns + f"_{self.variables}"
-        return results
+        per_variable_results = []
+        for variable in to_list(self.variables):
+            res = get_output_variable(
+                eplus_res=eplus_results,
+                key_values=self.key_values,
+                variables=variable,
+            )
+            res.columns = res.columns + f"_{variable}"
+            per_variable_results.append(res)
+        return pd.concat(per_variable_results, axis=1)
 
 
 class SimplifiedChiller(System):
diff --git a/tests/test_outputs.py b/tests/test_outputs.py
index 202edd5..67b674e 100644
--- a/tests/test_outputs.py
+++ b/tests/test_outputs.py
@@ -127,14 +127,20 @@ def test_get_output_zone_variable(self):
             ),
         )
 
-        pd.testing.assert_frame_equal(
-            toy_df.iloc[:, [0, 4]],
-            get_output_variable(
-                eplus_res=toy_df,
-                key_values="Zone1",
-                variables=[
-                    "Equipment Total Heating Energy",
-                    "Ideal Loads Supply Air Total Heating Energy",
-                ],
-            ),
+        # Real Sensor usage passes variable names that themselves start with
+        # "Zone" (e.g. "Zone Ideal Loads Supply Air Total Heating Energy"), so
+        # this must mirror that instead of variable names with no relation to
+        # the column's own key prefix - otherwise the drop_suffix regex bug
+        # (corrupted char-class silently failing to match) goes unnoticed.
+        multi_variable_result = get_output_variable(
+            eplus_res=toy_df,
+            key_values="Zone1",
+            variables=[
+                "Zone Other Equipment Total Heating Energy",
+                "Zone Ideal Loads Supply Air Total Heating Energy",
+            ],
+        )
+        assert list(multi_variable_result.columns) == ["ZONE1", "ZONE1"]
+        assert list(multi_variable_result.iloc[0]) == list(
+            toy_df.iloc[0, [0, 4]]
         )
diff --git a/tests/test_system.py b/tests/test_system.py
index 0439aec..9906bd4 100644
--- a/tests/test_system.py
+++ b/tests/test_system.py
@@ -133,6 +133,50 @@ def test_sensor(self):
             rel=0.05,
         )
 
+    def test_sensor_multi_variable(self):
+        """A Sensor built with a list of several variables must return one
+        distinctly-named, non-collapsed column per variable (regression test
+        for the silent multi-variable column collapse bug)."""
+        test_build = Building(idf_path=RESOURCES_PATH / "test.idf")
+        test_build.add_system(
+            sys.Sensor(
+                name="IDEAL_LOADS",
+                variables=[
+                    "Zone Ideal Loads Supply Air Total Heating Energy",
+                    "Zone Ideal Loads Supply Air Total Cooling Energy",
+                ],
+                key_values="*",
+            )
+        )
+
+        result = test_build.simulate(
+            parameter_dict={},
+            simulation_options={
+                "epw_file": (RESOURCES_PATH / "Paris_2020.epw").as_posix(),
+                "outputs": "SENSOR",
+                "verbose": "v",
+            },
+        )
+
+        heating_cols = [
+            c
+            for c in result.columns
+            if c.endswith("_Zone Ideal Loads Supply Air Total Heating Energy")
+        ]
+        cooling_cols = [
+            c
+            for c in result.columns
+            if c.endswith("_Zone Ideal Loads Supply Air Total Cooling Energy")
+        ]
+
+        assert len(heating_cols) == 4
+        assert len(cooling_cols) == 4
+        assert not set(heating_cols) & set(cooling_cols)
+        # the two variables must carry distinct, non-collapsed values
+        assert not np.allclose(
+            result[heating_cols].sum().to_numpy(), result[cooling_cols].sum().to_numpy()
+        )
+
     def test_heater_simple(self, idf):
         gas_boiler = sys.HeaterSimple(
             name="Main_boiler",