LabVIEW
Convert TDMS to CSV
@Birdy830 wrote:
Hi, im looking to convert a TDMS file to CSV, can anybody help with this ?
Regards
Birdy830
So this isn't always easy but in some cases it is. If I ask to convert a Word file to Text can you do it? Well it depends, does my Word file have charts, pictures graphs? Does it have formatted text or hyperlinks? Then if you saved it as text alot of that information will be lost.
Same with TDMS. TDMS is not a flat file structure and saving it to a CSV which is essentially flat means you will lose some data. But just like in Word, if you aren't using any of those extra fancy features then you might not care if that information is lost.
TDMS has 3 levels, File >> Groups >> Channels. You start with just one file, but it may have multiple groups, which may have multiple channels, which generally have multiple samples. If you have just one group then you can save just that group to a CSV where you have channels as columns and samples of those channels as rows. But even if you only have one group, all of the properties of the File or Groups, or Channels will be lost unless you get creative on how to save that to a CSV too.
So so summarize, if you only have one group, and if you don't mind only saving the data and not the properties, then it is straight forward. Use the TDMS Open, then TDMS read providing the group, channels, and the data type to read (hopefully they are all the same so you can read them all at once). Then close the TDMS file. Then use the Write To Spreedsheet File giving it the 2D array of signals you read and use the comma as the delimeter.
- last edited on by
Content Cleaner
I would highly recommend Scout by Signal X. It is free, they just ask for you to register the software.
https://www.ni.com/en/support/downloads/tools-network/download.scout-tdms-editor.html
The TDMS viewer that comes with LabVIEW (under the TDMS palette) has a graph function. By default it won't load all the points on the graph. But it does allow you to say how much data to read, and starting from an offset. So you can load the first 100,000 values into a graph see what it looks like then load another 100,000 starting where the last graph left off.
It sounds like you have lots and lots of data. Do you really need all that much or would a subset work for a graph? I mean if you have 1 Million points of data you can't graph all of that at once and see each point. Your computer monitor is likely only 1920 X 1080 resolution or less. Maybe it would make sence to load one point out of every 100 or 1000 and graph that. This can be done using the TDMS Open and TDMS Read where you can specify the offset and number of samples to read. Put this in a for loop, or a while loop until the whole file has been read and graph the values loaded.
