diff --git a/extra_geom/__init__.py b/extra_geom/__init__.py
index 2619f85..9bd90dc 100644
--- a/extra_geom/__init__.py
+++ b/extra_geom/__init__.py
@@ -35,8 +35,11 @@
 
 __version__ = "1.7.1"
 
-from .detectors import (AGIPD_1MGeometry, AGIPD_500K2GGeometry, DSSC_1MGeometry, Epix10KGeometry, Epix100Geometry,
-                        GenericGeometry, JUNGFRAUGeometry, LPD_1MGeometry, PNCCDGeometry, agipd_asic_seams)
+from .detectors import (AGIPD_1MGeometry, AGIPD_500K2GGeometry,
+                        DSSC_1MGeometry, Epix10KGeometry, Epix100Geometry,
+                        GenericGeometry, JUNGFRAUGeometry, LPD_1MGeometry,
+                        PNCCDGeometry, agipd_asic_seams)
+from .special import jungfrau_module_data
 
 __all__ = [
     'AGIPD_1MGeometry',
@@ -44,6 +47,7 @@
     'agipd_asic_seams',
     'GenericGeometry',
     'DSSC_1MGeometry',
+    'jungfrau_module_data',
     'JUNGFRAUGeometry',
     'LPD_1MGeometry',
     'PNCCDGeometry',
diff --git a/extra_geom/special.py b/extra_geom/special.py
new file mode 100644
index 0000000..3a991e7
--- /dev/null
+++ b/extra_geom/special.py
@@ -0,0 +1,55 @@
+from warnings import warn
+
+import numpy as np
+
+from .detectors import JUNGFRAUGeometry as JF
+
+
+def jungfrau_module_data(data, use_mask=True, edge_pixels=1, trains=np.s_[:]):
+    """Get data for the given JUNGFRAU detector
+
+    The returned data has inter ASIC gaps added and masked applied if requested
+
+    ::
+
+        from extra_data import open_run
+
+        run = open_run(1234, 1)
+        jungfrau = run['HED_IA1_JF500K1/DET/JNGFR01:daqOutput']
+        data = jungfrau_module_data(jungfrau)
+
+    data: extra_data.SourceData
+        A SourceData object for a JUNGFRAU detector
+    use_mask: bool
+        if True, use the mask present in the data (if in use with CORR data)
+    edge_pixels: int
+        number of pixels to mask at the edges of detector ASICS
+    trains: slice
+        Select a subset of trains from this data collection.
+            Slice trains by position within this data::
+                np.s_[:5]  # first 5 trains
+            Or select trains by train ID, with a slice or a list::
+                from extra_data import by_id
+                by_id[142844490 : 142844495]
+                by_id[[142844490, 142844493, 142844494]]
+    """
+    jf = data.select_trains(trains)
+    data = jf['data.adc'].ndarray()
+
+    if use_mask:
+        if 'data.mask' in jf.keys():
+            mask = jf['data.mask'].ndarray().astype(np.bool_)
+            data = data * ~mask
+        else:
+            warn("No mask present in data")
+
+    if edge_pixels > 0:
+        asic = np.zeros((JF.frag_ss_pixels, JF.frag_fs_pixels), dtype=np.bool_)
+        asic[edge_pixels:-edge_pixels, edge_pixels:-edge_pixels] = True
+        edge_mask = np.tile(
+            asic, [JF.expected_data_shape[-2] // JF.frag_ss_pixels,
+                   JF.expected_data_shape[-1] // JF.frag_fs_pixels])
+        data = data * edge_mask
+
+    assembled_data, centre = JF.from_module_positions().position_modules(data)
+    return assembled_data, centre