diff --git a/src/pandablocks/connections.py b/src/pandablocks/connections.py
index 2f6440ec..059f4d0e 100644
--- a/src/pandablocks/connections.py
+++ b/src/pandablocks/connections.py
@@ -223,14 +223,14 @@ def receive_bytes(self, received: bytes) -> bytes:
                     is_multiline = False
                 else:
                     # Check a correctly formatted response
-                    assert line.startswith("!"), (
-                        "Multiline response %r doesn't start with !" % line
-                    )
+                    assert line.startswith(
+                        "!"
+                    ), f"Multiline response {line} doesn't start with !"
             else:
                 # Single line mode
-                assert not self._lines, (
-                    "Multiline response %s not terminated" % self._lines
-                )
+                assert (
+                    not self._lines
+                ), f"Multiline response {self._lines} not terminated"
                 to_send += self._update_contexts([line])
         return to_send
 
diff --git a/src/pandablocks/hdf.py b/src/pandablocks/hdf.py
index 1ab31270..8008478f 100644
--- a/src/pandablocks/hdf.py
+++ b/src/pandablocks/hdf.py
@@ -80,12 +80,16 @@ class HDFWriter(Pipeline):
             per file created.
     """
 
-    def __init__(self, file_names: Iterator[str], dataset_names: Dict[str, str]):
+    def __init__(
+        self,
+        file_names: Iterator[str],
+        capture_record_hdf_names: Dict[str, Dict[str, str]],
+    ):
         super().__init__()
         self.file_names = file_names
         self.hdf_file: Optional[h5py.File] = None
         self.datasets: List[h5py.Dataset] = []
-        self.dataset_names = dataset_names
+        self.capture_record_hdf_names = capture_record_hdf_names
         self.what_to_do = {
             StartData: self.open_file,
             list: self.write_frame,
@@ -101,8 +105,11 @@ def create_dataset(self, field: FieldCapture, raw: bool):
         else:
             # No processor, datatype passed through
             dtype = field.type
+        dataset_name = self.capture_record_hdf_names.get(field.name, {}).get(
+            field.capture, f"{field.name}.{field.capture}"
+        )
         return self.hdf_file.create_dataset(
-            f"/{self.dataset_names.get(field.name, field.name)}.{field.capture}",
+            f"/{dataset_name}",
             dtype=dtype,
             shape=(0,),
             maxshape=(None,),
@@ -196,7 +203,9 @@ def scale_data(self, data: FrameData) -> List[np.ndarray]:
 
 
 def create_default_pipeline(
-    file_names: Iterator[str], dataset_names: Dict[str, str], *additional_downstream_pipelines: Pipeline
+    file_names: Iterator[str],
+    capture_record_hdf_names: Dict[str, Dict[str, str]],
+    *additional_downstream_pipelines: Pipeline,
 ) -> List[Pipeline]:
     """Create the default processing pipeline consisting of one `FrameProcessor` and
     one `HDFWriter`. See `create_pipeline` for more details.
@@ -204,13 +213,17 @@ def create_default_pipeline(
     Args:
         file_names: Iterator of file names. Must be full file paths. Will be called once
             per file created. As required by `HDFWriter`.
-        dataset_names: A dictionary of dataset names to use for each field.
+        capture_record_hdf_names: A dictionary of dataset names to use for each field.
+            The keys are record names, the values are another dictionary of
+            capture type to dataset name.
         additional_downstream_pipelines: Any number of additional pipelines to add
             downstream.
     """
 
     return create_pipeline(
-        FrameProcessor(), HDFWriter(file_names, dataset_names), *additional_downstream_pipelines
+        FrameProcessor(),
+        HDFWriter(file_names, capture_record_hdf_names),
+        *additional_downstream_pipelines,
     )
 
 
@@ -257,6 +270,7 @@ async def write_hdf_files(
         HDFDataOverrunException: if there is a data overrun.
     """
     counter = 0
+
     end_data = None
     pipeline = create_default_pipeline(file_names, {})
     try: