Online TDMS viewer, no LabVIEW needed

How to open a TDMS file online

1. 1Open your .tdms file. Choose it, or drop it on the viewer at the top of this page. Its first group opens with its first channel plotted against its time. No file to hand? Click “Try a 12-million-sample TDMS file”: a week of made-up readings, made on your device.
2. 2See what’s inside. The menu in the file’s line lists its groups (a group with channels of different lengths or rates gives a table for each). “Variables” shows the file, its groups and channels with their type, length, first values and every property.
3. 3Plot, zoom and export. Tick channels in the list or in Variables to plot them, zoom with the scroll wheel, a pinch or a drag, and read their statistics. “Download all columns (CSV)” under “Do more” saves the table shown, with its time column.

A worked example you can check

This example uses the smaller sample: “Sample TDMS file”, under “Smaller samples”, opens a month of made-up UK weather, written the way LabVIEW writes a log: three groups, each with its own time base. To run the lines below yourself, download the sample TDMS file.

• Hourly (744 values, 5 channels): temperatures in London, Leeds and York, London’s humidity and pressure, every hour from 1 January 2026.
• Rain gauge (4,464 values, 2 channels): rain in London and Leeds every ten minutes, on its own time base.
• Daily (31 values, 4 channels): a timestamp channel, Date, which becomes the time axis, the day’s highest and lowest temperature and the weather in words.

It opens on Hourly, with London_Temperature plotted against the time its waveform properties give: from 1 January 2026, 00:00 UTC, every hour. In Variables, the channel reads float64 744 values, with its first values 5.44, 4.99, 4.74, 4.7, … · degC; click it for its properties:

wf_start_time = 2026-01-01 00:00:00 UTC · wf_increment = 3600 · wf_start_offset = 0 · unit_string = degC · description = Air temperature, London

The same file read with npTDMS, the Python library for TDMS files, gives the same channels and values:

from nptdms import TdmsFile

tdms = TdmsFile.read("megarows-sample-uk-weather.tdms")
for group in tdms.groups():
    print(group.name, len(group.channels()), "channels,", len(group.channels()[0]), "values")
london = tdms["Hourly"]["London_Temperature"]
print(london.properties["unit_string"], london[:3])
print(london.time_track(absolute_time=True)[:2])

Hourly 5 channels, 744 values
Rain gauge 2 channels, 4464 values
Daily 4 channels, 31 values
degC [5.44 4.99 4.74]
['2026-01-01T00:00:00.000000000' '2026-01-01T01:00:00.000000000']

Run on 2026-10-09 with npTDMS 1.12.1, NumPy 2.5.3, Python 3.12.15. A test opens the sample on every change to the site and checks that the viewer shows these groups, times and values. Download the sample (.tdms, 105 KB) to open it elsewhere too.

Groups, channels and their time

A TDMS file holds groups of channels, each with properties. The viewer opens each group as a table, its channels as columns, so a group of channels logged together plots on one time axis. Channels of a different length or rate inside a group get a table of their own, each on its own time, and channels you tick in Variables from different groups open together, each still on its own time axis.

The time comes from the waveform properties LabVIEW and DAQmx write (start time, start offset and increment): dates and times in UTC, or seconds from the start when the file gives no start time. A group without them uses a channel named like a time, or a timestamp channel, else the sample number. Units come from each channel’s unit_string property, and every property, of the file, of each group and of each channel, is in Variables.

Channels with different units read best apart: plot TDMS channels in stacked lanes, each with its own Y axis and range, on one time axis.

Big files and loggers

TDMS logs are often large and written in many pieces. The viewer reads where each channel’s values are without reading the values, so a big file’s groups and channels show at once; a channel’s values are read when you plot it, ask for its statistics or export it, and the data table reads only the rows on screen. Zooming reloads the full detail of the range in view, so a one-sample spike in millions stays visible.

Interleaved and big-endian files, TDMS 1.0 and 2.0, and DAQmx raw data with its scales all open the same way. A file a logger is still writing opens with everything it has written completely, and a note says the end was left out.

To see it with a big file, “Try a 12-million-sample TDMS file” makes a week of made-up sonic anemometer readings on your device (the wind in three directions and the temperature, 20 a second, 12,096,000 per channel), writes them as a TDMS file of seven segments, one a day, and opens it like any other: nothing is downloaded or uploaded. Zoom into any ten seconds and every reading is there, gusts and spikes included. Phones get one day, 1,728,000 per channel.