diff --git a/docs/examples/data/gen5_data_export_test_12-1-2025.xlsx b/docs/examples/data/gen5_data_export_test_12-1-2025.xlsx
new file mode 100644
index 0000000..eca9c17
Binary files /dev/null and b/docs/examples/data/gen5_data_export_test_12-1-2025.xlsx differ
diff --git a/mtphandler/readers/biotek.py b/mtphandler/readers/biotek.py
index 952b305..7b67777 100644
--- a/mtphandler/readers/biotek.py
+++ b/mtphandler/readers/biotek.py
@@ -61,11 +61,22 @@ def read_biotek(
 
     wavelengths = extract_integers(wavelengths_cell)
 
-    measurement_int_cell = str(df.iloc[15, 1])
-    if "add final component" in measurement_int_cell.lower():
-        measurement_int_cell = str(df.iloc[16, 1])
-
-    measurement_interval = parse_measurement_interval(measurement_int_cell)
+    measurement_interval = None
+    for index, row_value in df.iloc[:, 0].items():
+        if isinstance(row_value, str) and "start kinetic" in row_value.lower(): # this is the start of a kinetic read block
+            interval_candidate = str(df.iloc[index, 1])
+            try:
+                measurement_interval = parse_measurement_interval(interval_candidate)
+                break # Found a valid interval, stop searching. Note that there could be more kinetic intervals later.
+            except ValueError:
+                continue # Keep searching if this one fails
+    
+    if measurement_interval is None:
+        # Fallback to original logic if 'Start Kinetic' not found or parsed
+        measurement_int_cell = str(df.iloc[15, 1])
+        if "add final component" in measurement_int_cell.lower():
+            measurement_int_cell = str(df.iloc[16, 1])
+        measurement_interval = parse_measurement_interval(measurement_int_cell)
 
     plate = Plate(
         date_measured=timestamp,
diff --git a/tests/unit_tests/test_readers/test_biotek_epoch_2.py b/tests/unit_tests/test_readers/test_biotek_epoch_2.py
index 674f835..db01422 100644
--- a/tests/unit_tests/test_readers/test_biotek_epoch_2.py
+++ b/tests/unit_tests/test_readers/test_biotek_epoch_2.py
@@ -34,3 +34,43 @@ def test_read_biotek_epoch_2():
 
         if well.id == "D9":
             assert well.measurements[0].absorption[-1] == pytest.approx(0.226, rel=1e-2)
+            
+    # Test another BioTek plate example
+    """Tests that read_biotek can correctly parse the gen5 data file with dynamic kinetic interval detection."""
+    test_file_path = "docs/examples/data/gen5_data_export_test_12-1-2025.xlsx"
+
+    # Ensure the test file exists
+    #assert os.path.exists(test_file_path), f"Test file not found: {test_file_path}"
+
+    # Use the read_biotek function with the test file
+    # This should now run without raising a ValueError due to the fix
+    plate = read_biotek(path=test_file_path, ph=7.4)
+
+    # Assert that a Plate object is returned
+    assert isinstance(plate, Plate)
+
+    # Assert that the plate contains wells and measurements
+    assert len(plate.wells) > 0
+    assert len(plate.wells[0].measurements) > 0
+
+    # Get the first measurement from the first well
+    first_measurement = plate.wells[0].measurements[0]
+
+    # Assert that the time data is populated and has more than one point
+    assert len(first_measurement.time) > 1
+
+    # The expected interval from the file is 'Interval 0:08:00', which is 8 minutes.
+    # As currently implemented, `np.arange` uses this minute value directly
+    # to generate the time series, even though `time_unit='s'` is specified.
+    expected_interval_minutes = 8.0
+    actual_interval = first_measurement.time[1] - first_measurement.time[0]
+
+    # Assert that the actual interval matches the expected 8.0 minutes
+    assert actual_interval == pytest.approx(expected_interval_minutes)
+
+    # Note: There's a potential inconsistency in the read_biotek function
+    # where time points are generated using 'minutes' but `time_unit` is set to 's' (seconds).
+    # This test validates the numerical interval extracted, but a future enhancement
+    # might involve ensuring strict unit consistency (e.g., converting minutes to seconds
+    # before generating the time array if time_unit is 's').
+