Bug: Keysight N5249A PNA-X network analyzer trace error

[bugusan]

Regardless of which qcodes release(tried with 4 different qcodes version) I use whenever I want to have trace data of network analyzer, N5249A, I get could not convert string to float: 'S21' error. My code is simple. I first set the required parameter for the network analyzer and then try to get trace data: data=qc.Measure(vna.trace).run(name='test')

my code is:

vna.sweep_mode('CONT')
vna.start(150.1e6)
vna.stop(350.1e6)
vna.power(-23)
vna.points(1601)# 1601
vna.if_bandwidth(1000)
vna.trace("S21")
data=qc.Measure(vna.trace).run(name='test') 

and it returns this error:

ValueError                                Traceback (most recent call last)
<ipython-input-8-48831a7c8d52> in <module>
----> 1 data=qc.Measure(vna.trace).run(name='test')
      2 
      3 # plot
      4 #plot = MatPlot(data.vna_magnitude, data.vna_phase)
      5 #plot.save()

~\anaconda3\envs\qcodes10\lib\site-packages\qcodes\measure.py in run(self, use_threads, quiet, station, **kwargs)
     83 
     84         # run the measurement as if it were a Loop
---> 85         self._dummyLoop.run(use_threads=use_threads,
     86                             station=station, quiet=True)
     87 

~\anaconda3\envs\qcodes10\lib\site-packages\qcodes\loops.py in run(self, use_threads, quiet, station, progress_interval, set_active, *args, **kwargs)
    745             if not quiet:
    746                 print(datetime.now().strftime('Started at %Y-%m-%d %H:%M:%S'))
--> 747             self._run_wrapper()
    748             ds = self.data_set
    749         finally:

~\anaconda3\envs\qcodes10\lib\site-packages\qcodes\loops.py in _run_wrapper(self, *args, **kwargs)
    794     def _run_wrapper(self, *args, **kwargs):
    795         try:
--> 796             self._run_loop(*args, **kwargs)
    797         finally:
    798             if hasattr(self, 'data_set'):

~\anaconda3\envs\qcodes10\lib\site-packages\qcodes\loops.py in _run_loop(self, first_delay, action_indices, loop_indices, current_values, **ignore_kwargs)
    882                     # log.debug('Going through callables at this sweep step.'
    883                     #           ' Calling {}'.format(f))
--> 884                     f(first_delay=delay,
    885                       loop_indices=new_indices,
    886                       current_values=new_values)

~\anaconda3\envs\qcodes10\lib\site-packages\qcodes\actions.py in __call__(self, loop_indices, **ignore_kwargs)
    165                 out_dict[param_id] = param_out
    166 
--> 167         self.store(loop_indices, out_dict)
    168 
    169 

~\anaconda3\envs\qcodes10\lib\site-packages\qcodes\data\data_set.py in store(self, loop_indices, ids_values)
    404          """
    405         for array_id, value in ids_values.items():
--> 406             self.arrays[array_id][loop_indices] = value
    407         self.last_store = time.time()
    408         if (self.write_period is not None and

~\anaconda3\envs\qcodes10\lib\site-packages\qcodes\data\data_array.py in __setitem__(self, loop_indices, value)
    339         self._update_modified_range(min_li, max_li)
    340 
--> 341         self.ndarray.__setitem__(loop_indices, value)
    342 
    343     def __getitem__(self, loop_indices):

ValueError: could not convert string to float: 'S21'

[astafan8]

I'm not very much familiar with that driver/instrument but i think vna.trace is just the particular S-parameter that you are interested in. The actual data is available in .magnitude and .phase parameters, see the example notebook here https://qcodes.github.io/Qcodes/examples/driver_examples/Qcodes%20example%20with%20Keysight%20Network%20Analyzer.html.

[bugusan]

I'm not very much familiar with that driver/instrument but i think vna.trace is just the particular S-parameter that you are interested in. The actual data is available in .magnitude and .phase parameters, see the example notebook here https://qcodes.github.io/Qcodes/examples/driver_examples/Qcodes%20example%20with%20Keysight%20Network%20Analyzer.html.

thank you for your comment. Actually, in that way, I could get data but since I am sweeping voltage from some other instruments, I have 2D loop, to be able to make all my instrument work at the same time I need VNA to work via qcodes.Measure function

[astafan8]

to be able to make all my instrument work at the same time I need VNA to work via qcodes.Measure function

i'm not sure i understand why. I wanted to suggest that instead of data=qc.Measure(vna.trace).run(name='test') one needs to use data=qc.Measure(vna.magnitude).run(name='test') because vna.magnitude parameter acquires the data while vna.trace parameter jsut allows you to select which S parameter you're interested in.

---

anyway, did this issue get resolved? if so, please close it :)

[bugusan]

I agree on using vna.magnitude. However, this time I get unhashable type: 'numpy.ndarray' error.

data=qc.Measure(vna.magnitude).run()

ypeError                                 Traceback (most recent call last)
<ipython-input-35-4a14ea72942f> in <module>
----> 1 data=qc.Measure(vna.magnitude).run()
      2 

~\anaconda3\envs\qcodes10\lib\site-packages\qcodes\measure.py in run(self, use_threads, quiet, station, **kwargs)
     75         """
     76 
---> 77         data_set = self._dummyLoop.get_data_set(**kwargs)
     78 
     79         # set the DataSet to local for now so we don't save it, since

~\anaconda3\envs\qcodes10\lib\site-packages\qcodes\loops.py in get_data_set(self, *args, **kwargs)
    656         """
    657         if self.data_set is None:
--> 658             data_set = new_data(arrays=self.containers(), *args, **kwargs)
    659             self.data_set = data_set
    660 

~\anaconda3\envs\qcodes10\lib\site-packages\qcodes\loops.py in containers(self)
    466                 # note that this supports lists (separate output arrays)
    467                 # and arrays (nested in one/each output array) of return values
--> 468                 action_arrays = self._parameter_arrays(action)
    469 
    470             else:

~\anaconda3\envs\qcodes10\lib\site-packages\qcodes\loops.py in _parameter_arrays(self, action)
    548             for j, (vij, nij, lij, uij) in enumerate(zip(sp_vi, sp_ni, sp_li, sp_ui)):
    549                 sp_def = (shape[: 1 + j], j, setpoints, vij, nij, lij, uij)
--> 550                 if sp_def not in all_setpoints:
    551                     all_setpoints[sp_def] = self._make_setpoint_array(*sp_def)
    552                     out.append(all_setpoints[sp_def])

TypeError: unhashable type: 'numpy.ndarray'

[astafan8]

@bugusan a side question - do you absoltely need to use qcodes.Measure? Couldn't you use qcodes.data_set.measurements.Measurement context manager as described in this doc https://qcodes.github.io/Qcodes/examples/DataSet/Performing-measurements-using-qcodes-parameters-and-dataset.html ?

[astafan8]

I couldn't figure out why the "unhashable type: 'numpy.ndarray'" occurs, perhaps, Loop and Measure are not smart enough about parameters which return numpy arrays of data. You could ask this question on our Slack channel, perhaps other people from our community know - do you know how to access the Slack channel?

[bugusan]

would you please advise me of how to access the Slack chann?

[astafan8]

@bugusan just send an email to Mikhail.Astafev(at)microsoft.com and I will invite you to the Slack channel