AI data not being recorded by Lyse

[Alex M.]

Hello all,

I am trying to measure the high-voltage of a 5 millisecond square wave generated by a digital channel on an NI_PCIe-6361 DAQ using an analog input channel on the same DAQ. To do this, I am using the acquire() method to measure the voltage during the entire 20 ms shot I have programmed, with the square wave lasting from 5 to 10 ms.

However, I am running into some problems when I try to access that measurement from the data frame in lyse. When the shots are loaded into the lyse GUI, I can see in the GUI data frame all the meta data about the shots except for the measurement of the voltage of the square wave. It seems that the data is not being recorded. As a sanity check, I was able to create the desired signal in an oscilloscope, so I know the square wave is being generated.

For reference, my experimental setup consists of a prawnblaster, one clockline of which clocks the 6361 DAQ. I am using a Rigol function generator to produce pulses at a frequency of 10 MHz as an external reference to synchronize the prawnblaster and the AI channel to avoid drift. All other aspects of shot generation and labscript are working flawlessly, it is just the issue with the voltage data seemingly not being recorded.

I have attached my connection table and experiment script I am using. Any help is greatly appreciated!

best,

Alexander

[dihm....@gmail.com]

Alexander,

Have you used H5Viewer to actually look at the generated shot files? I suspect the data is in there and just not showing in lyse as you expect. In fact, AI traces are not saved to the lyse dataframe by design (the dataframe is an actual pandas dataframe, which doesn't like nesting array structures into individual entries). To access input traces with lyse, you should use the Run.get_trace functionality (and cousins). Of course, the traces can also be had by using h5py directly and pulling from the correct group.

-David