Channel Editor

The Channel Editor is where recordings are prepared before analysis. It combines two jobs in one window: fixing up channel metadata (names, units, physical quantities, calibration values, measurement DOFs) and performing data preparation on the samples themselves (cropping a run-up out of a long recording, joining several files, removing a DC offset, rescaling a channel against a calibrator, deriving a difference channel, checking for clipping).

Data preparation is non-destructive. Every operation that changes samples — crop, gain, DC removal, invert, zero, calibrate, channel math, concatenate, export channels — asks you for a new file name and writes a new file. Your original recording is never modified. The only operation that writes back into the original file is Save Changes, and it only updates header metadata, never the recorded samples.

Opening the Channel Editor

There are three ways in:

Start Menu → Editor → Channel Editor — opens the editor empty; load files with the Add Files… toolbar button.
Right-click a file in the file tree → Edit with → Channel Editor — opens the editor with that file already loaded.
Add Files… inside the editor — loads one or more further files; several files can be open side by side in the same table.

Supported inputs are HEAD time-data .hdf and Deepsona .dpsa recordings.

Window Layout

Area	Contents
Toolbar (top)	Save Changes, Add Files, Delete selected, Clear all, Move up / down, Export channels, Concatenate files, Bulk rename, Export / Import CSV, Copy calibration diagnostics. A progress bar and a status message appear at the right.
Channel table (centre)	One row per channel of every loaded file. Editable metadata cells, read-only file information cells.
Preview (bottom)	Time Domain waveform on the left and FFT vs. Time (Spectrogram) on the right, for the selected channel row.
Tool panel (right)	Four sections — Extract, Repair, Calibrate, Quality. Appears once a channel row is selected.

The right-hand tool panel stays hidden until you select a channel row. It is also hidden for transfer-function files, because the sample-level operations do not apply to them.

The Channel Table

Column	Editable	Description
No.	no	Position of the channel within its file, starting at 1. It follows the current row order, so it renumbers when you move rows.
Name	yes	Full channel name.
Abbr	yes	Short name used in compact plot legends and tables.
Title	yes	Descriptive title.
Sampling Rate	no	Sampling rate of this channel in Hz.
Nbr of Samples	no	Number of recorded samples.
Quantity	yes	Physical quantity, chosen from a list that starts with the common ones (sound pressure, vibration acceleration, speed of rotation, force, voltage…) followed by a separator and the full alphabetical list.
Unit	yes	Unit of measure. The available units follow the selected Quantity.
Calibration	yes	The sensitivity / calibration factor stored in the file header.
DOF	yes	Measurement degree of freedom, in the form `point:axis`, e.g. `3:Z+`. Type it, or click the `…` button for a picker.
*	no	An amber asterisk marks a row with unsaved metadata edits. Edited rows also get a pale yellow background.
File Name / Directory	no	Where the channel came from. Hover for the full path.

Changing Quantity resets Unit to that quantity's default unit. Always set the quantity first and the unit second, otherwise your unit selection will be overwritten.

Reordering channels

Drag a row to a new position, or use the Move selected channel up / down toolbar buttons. The new order is not written anywhere by itself — it takes effect when you use Export channels to file, which writes the channels out in the displayed order.

The DOF picker

Click the … button in the DOF cell to open DOF Picker. Filter the point list by typing in the search box, select a measurement point, then pick one of the six axis buttons X+ X− Y+ Y− Z+ Z−. The picker needs a measurement point library, so open the Geometry module and load or import your points first; otherwise the picker reports that no measurement points are available.

Editing Metadata and Saving Changes

Edit the Name, Abbr, Title, Quantity, Unit, Calibration or DOF cells. Edited rows are marked with an asterisk.
Click Save Changes in the toolbar.
The header of each affected file is updated in place and the asterisks clear.

Only the header fields listed above are written. The recorded samples are untouched, so metadata editing is safe to repeat and cannot corrupt a measurement.

Bulk rename

Click Bulk rename channels to rename many channels at once. The dialog applies to the selected rows, or to all rows if nothing is selected, and offers four fields:

Field	Effect
Prefix (added to the front)	Text inserted before the existing name.
Suffix (added to the end)	Text appended after the existing name.
Find (regex, optional)	A regular expression matched against the existing name.
Replace with	Replacement text. Capture groups from the Find pattern are available as `$1`, `$2`…

Find and replace runs first, then the prefix and suffix are attached. For example, Find ^Mic (\d+)$ with Replace M$1 turns Mic 3 into M3. Bulk rename only changes the rows in the table — click Save Changes afterwards to write the new names into the files.

CSV and clipboard

For large channel lists it is often faster to edit metadata in a spreadsheet:

Export to CSV writes the whole table (No., Name, Abbr, Title, Unit, Quantity, Calibration, DOF, Sampling Rate, Samples, File) to a UTF-8 CSV file.
Import from CSV reads a file back in, starting at the currently selected row. The expected column order is Name, Abbr, Title, Unit, Quantity, Calibration, DOF.
Ctrl+V on the table pastes tab-separated cells straight from Excel, in the same column order, starting at the selected row.

Imported and pasted values land in the table as ordinary edits — the asterisk column shows what changed, and nothing reaches the files until you click Save Changes.

Selecting a Region

Most data-preparation operations act on a time window. Select the channel row you want to work on — the waveform and spectrogram appear at the bottom — then define the window in one of three ways:

Drag across the Time Domain waveform.
Type the Start and End times, in seconds, in the Extract section of the right-hand panel.
Click Auto Ramp to have the run-up or coast-down located automatically.

The selected length is shown underneath as Δ n s. Reset clears the selection. Right-click the waveform for Replay Selection, Crop to New File and Reset Selection.

Auto Ramp

Auto Ramp finds the main speed ramp in the recording and sets Start and End to its endpoints. It looks for a speed signal in this order: a pulse / tacho channel in the same file, or — failing that — the selected channel itself if its unit or quantity indicates rotational speed. The speed trace is lightly smoothed, and the ramp endpoints are placed where the curve leaves and re-enters a narrow band around its minimum and maximum. The status bar reports what was found, for example Run-up ramp: 800→6000 rpm, [12.340, 71.220] s.

Auto Ramp only sets the selection — it writes nothing. Review the highlighted region on the waveform, adjust the Start / End boxes if you want a little margin, and then crop.

Extract: Cropping to a New File

Select the channel row and define the time window.
Click Crop to New File (or right-click the waveform and choose it).
Accept or edit the suggested file name — it is built from the source name plus the time window, e.g. Run01_crop_12.34-71.22s.hdf.
Click Save. The new file is written and immediately loaded into the table, so its channels appear as additional rows.

Cropping keeps all channels of the file over the chosen time window — it trims in time, not in channel count. The source file is not modified.

Repair: Gain, DC, Invert, Zero

The Repair section fixes common recording defects. Each button writes a new file with a descriptive suffix and then loads it.

Operation	What it does	Region	File suffix
Gain × + Apply	Multiplies the channel by the entered factor. Use this to correct a wrong sensitivity by a known ratio.	Selection if one exists, otherwise the whole channel.	`_gain`
DC	Subtracts the mean value, removing a DC offset.	Selection if one exists, otherwise the whole channel.	`_removedc`
Invert	Negates the channel — fixes a sensor mounted or wired the wrong way round.	Selection if one exists, otherwise the whole channel.	`_invert`
Zero	Sets the selected region to zero — silences a disturbance such as a dropped tool or a spoken comment.	Requires a selection.	`_zero`

Gain, DC and Invert fall back to the whole channel when no region is selected. If you only meant to treat a section, make sure a selection is active before you click.

Calibrate

If a recording contains a calibrator tone, the channel can be rescaled to the correct absolute level:

Select the channel row.
Drag out a region that lies entirely inside the steady calibrator tone. Avoid the switch-on and switch-off transients.
Enter the calibrator level in the Level box — the default is 94 dB.
Click Calibrate.

The RMS of the selected window is measured, the gain needed to bring it to the target level is computed, and that gain is applied to the whole channel, which is written to a new file with the suffix _cal. Sound channels are referred to 20 µPa; all other channels are referred to 1.0 in their own unit. The status bar reports both the target and the measured level, e.g. Calibrated to 94 dB (measured 86.03 dB, ×2.505).

This is a sample-level rescale. It is different from the Calibration column in the table, which is a metadata field describing the sensor sensitivity. Use the column when the file's stored sensitivity is simply wrong; use Calibrate when you want the recorded waveform itself brought onto an absolute scale using a calibrator.

Channel Math

Channel Math… derives a new channel from two existing channels of the same file. It is the standard way to obtain a pressure difference, a sum signal or a scaled copy.

Select a row of the file you want to work with, then click Channel Math….
Choose channel A (primary channel) and B (second channel).
Choose the Operation: A - B, A + B, A * B, A * scale or (A + B) / 2.
For A * scale, enter the factor in Scale.
Enter the New channel name and, optionally, a New unit.
Click Create and choose a destination file name.

The derived channel is appended after all existing channels in the new file. All channels must share one sample rate; where A and B differ in length, the result is as long as the shorter of the two.

Concatenating Files

Concatenate files end-to-end into one new file joins several recordings of the same configuration into a single long file — for example several consecutive run-ups recorded as separate takes.

Click the concatenate button in the toolbar.
In the file dialog select at least two files. The order in which you select them is the order in which they are joined.
Choose the output file name (suggested: concatenated.hdf).

A short fade is applied at each joint so the transition does not click. Pulse channels are joined without fading so that speed information stays intact.

All input files must have the same channel count, the same sample rate, and a single sample rate per file. HDF and Deepsona files cannot be mixed in one concatenation — convert to a common format first. Transfer-function files cannot be concatenated.

Deleting, Reordering and Exporting Channels

Removing unused channels and writing out a tidy file is a two-step operation:

Select the rows you no longer need and click Delete selected channel(s), or use the same command from the row context menu. Confirm the prompt. The rows disappear from the table.
Optionally reorder the remaining rows by dragging or with the Move up / down buttons.
Click Export channels to file. Choose a name (suggested: source name plus _channels). The remaining channels are written to a new file, in the displayed order.

Deleting rows only removes them from the list in the editor. Nothing is written until you export. If you close the editor or clear the list without exporting, the deletions are lost — and equally, your original file still contains every channel.

Export works on one file at a time: select rows belonging to a single source file, or load just that one file and export everything that remains.

Quality: Clipping Check

Click Check Clipping to test the selected channel — the selected region if there is one, otherwise the whole channel — for samples sitting at the recording's full-scale limit. The report shows the channel name, the region examined, the number of samples, the verdict, the count and percentage of full-scale samples, and the peak, minimum and maximum values.

Clipping is reported when a meaningful number of samples sit at the extreme values — a handful of isolated samples is not flagged. For full amplitude statistics (RMS, kurtosis, skewness, statistics over time), use the Statistic stage in the analysis Pool instead.

Replay

Right-click the waveform and choose Replay Selection to listen to the selected region through the default audio device. This is the quickest way to confirm that a suspect section really is a disturbance before you zero or crop it.

Typical Workflow: Preparing a Run-Up for Order Analysis

Right-click the recording in the file tree and choose Edit with → Channel Editor.
Check the metadata: set the correct Quantity first, then Unit, for every channel. Fix any wrong Calibration values. Click Save Changes.
Select the microphone channel and click Check Clipping to confirm the recording is usable.
Click Auto Ramp. Review the highlighted region on the waveform and widen the Start / End values slightly if you want some margin.
Click Crop to New File and save as e.g. Run01_crop_12.34-71.22s.hdf. The cropped file is loaded automatically.
On the cropped file, select the channels you actually need for the analysis, delete the rest, and click Export channels to file.
Drag the resulting file into the analysis Pool.

Notes and Limitations

Sample-level operations require a single sample rate per file. Multi-rate recordings must be prepared elsewhere.
Transfer-function files support metadata editing only — the right-hand tool panel is hidden for them.
Pulse and CAN channels pass through sample operations unchanged, so speed and bus information survives cropping and repair.
The output of every write operation is loaded back into the table automatically. After a few operations the table can get long; use Clear all to start fresh (you will be warned if there are unsaved metadata edits).
Copy calibration diagnostics puts a plain-text report of the raw calibration fields of each loaded HDF channel on the clipboard. Use it when you need to send an exact description of a file's calibration state to support.
Status messages in the toolbar are shown in English in all interface languages.