You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi Scott,
First of all, thanks for your great work!
I'd like to report a problem when a channel is used for triggering, but not shown on the oscilloscope screen.
Steps to reproduce: I used a DS1104Z (latest FW: 00.04.04.SP4) to sample 2 channels, used CH2 for triggering on the slope of a rectangular signal, then turned off CH2 (the yellow CH2 light is off, CH2 is not shown on the screen).
Then I did a single shot capture, then saved the .wfm file.
Now when I read the file using RigolWFM, CH1 and CH2 are interleaved.
I would have expected to only get CH1.
Just saw this report today. Sorry. Unfortunately, there is not much that can be done because I don't know where the triggering information is in the .wfm file. If you reverse engineer something I would be happy to implement it.
A workaround is to save both the trigger and signal channel. The problem is that when the trigger channel is not enabled, there is no way to know that it is in the file or not.
Hi Scott,
First of all, thanks for your great work!
I'd like to report a problem when a channel is used for triggering, but not shown on the oscilloscope screen.
Steps to reproduce: I used a DS1104Z (latest FW: 00.04.04.SP4) to sample 2 channels, used CH2 for triggering on the slope of a rectangular signal, then turned off CH2 (the yellow CH2 light is off, CH2 is not shown on the screen).
Then I did a single shot capture, then saved the .wfm file.
Now when I read the file using RigolWFM, CH1 and CH2 are interleaved.
I would have expected to only get CH1.
The effect looks like this:


The .wfm file is attached:
NewFile1.wfm.zip
Thanks
Michael
The text was updated successfully, but these errors were encountered: