Skip to main content

​
Description

With this manipulator you import measurement data from MF4 (ASAM MDF4) files into a notebook. You can import every channel in the file, or filter the import down to just the channels you need.

​
Application

MF4 files recorded by test benches and data loggers often contain hundreds or thousands of channels, but a typical analysis only uses a handful of them. You are likely to start a notebook with this manipulator to bring in your raw measurement data, and the channel filter lets you import only the signals you care about — which means faster loads and lower memory use on large files.
Not all customers have access to channel filtering yet. The channel-selection and CSV import options described below are being rolled out gradually, and some users have been given early access. If you do not see a Filter channels? option after selecting your file, this feature has not been switched on for your workspace yet — please contact your Monolith representative if you would like it enabled. The rest of the MF4 import works as normal.

​
How to use

The importer dialog consists of the sections shown in the image below.
  1. Import type option.
  2. Selected files are displayed here.
  3. Click this button to open a file/folder selection dialog.
  4. Filter channels? — tick this to choose a subset of channels (see below). It appears once a file is selected.
  5. Assign a name to the dataset which is created from the imported file(s). All other manipulators refer to the imported dataset via this name.
[Screenshot: the MF4 importer showing (1) import type, (2) selected files, (3) Add/Remove Datasets, (4) the “Filter channels?” checkbox, and (5) the dataset name field.]

​
Selecting your file

Click Add/Remove Datasets to open the file selection dialog, choose your .mf4 / .mdf file from the File Manager, and confirm your selection. Selecting one or more files reads the available channels from the file(s).

​
Importing multiple files

You can select more than one file in a single import. The files are combined into one dataset by stacking their rows — the rows of one file follow the rows of the next — and a filename column is added so you can see which file each row came from.
The files do not have to contain the same channels. The combined dataset holds the union of every channel found across the selected files, and where a file does not contain a particular channel those rows are left blank for that column. Channels that share the same name across files are stacked into a single column, and rows are never aligned on time — so two files that happen to share the same timestamps produce one row per file rather than a single merged row. (This differs from the Tabular Loader, which requires all selected files to share the same columns.)
If the same channel name appears in more than one group within a single file, each occurrence is kept as a separate column — the duplicate is renamed with a numeric suffix (for example rpm and rpm_0) — so no data is overwritten.

​
Filtering channels

By default, leaving Filter channels? unticked imports every channel in the file — this matches the previous behaviour. Tick Filter channels? to reveal the Channels to use picker. The picker lists all channels found across the selected file(s).
  • Tick the channels you want to import directly in the list.
  • Leave the picker empty to import all channels.
[Screenshot: the “Channels to use” picker open, showing the list of channels with a count of how many are selected.]
The names shown in the picker are the raw channel names from the MF4 file. The column names created in the notebook are normalised (sanitised) during import, so they may differ slightly from the names shown here.

​
Importing a channel list from a CSV file

If you always work with the same set of channels, you can keep their names in a CSV (or .txt) file and load them in one step instead of ticking each one by hand. This is useful for long, standardised channel lists shared across a team. Inside the Channels to use picker:
  1. Click Select .csv and choose your CSV file from the File Manager.
  2. Click Apply .csv. The names in the file are matched against the channels in your MF4 file and the matching channels are selected for you.
The CSV simply needs to contain the channel names (one per line, or one per cell). Names must match the channel names in the MF4 file exactly. [Screenshot: the “Select .csv” and “Apply .csv” controls inside the Channels to use picker.] Depending on how well the names in your CSV match the channels in the file, you will see one of the following:

​
Data file requirements

  • File format: MF4 / MDF4 (.mf4, .mdf).
  • CSV/TXT used for the channel list: plain text, UTF-8 encoded, containing channel names only.
  • Channel names in the CSV must match the raw channel names in the MF4 file exactly (the names shown in the picker).

​
Advanced options

Click Show advanced options in the importer to control:
Resampling step lengthResamples and unifies the timestamps across channels. If the timestamps are not the same across all channels in the original MF4 file, leaving this unset can lead to an out-of-memory error.
Include metadataImports the file’s metadata as a separate dataset, which you can name.
Last modified on July 23, 2026