Fix: Keysight_34465A overload value 9.9e37 to be np.inf instead

[lakhotiaharshit]

In case the DMM tries to measure the voltage value outside the its range, it overloads and throws 9.9e37 instead of throwing an error. Here we substitute this number with np.nan.

Side note: The error query does not throw any error even though the instrument display shows in bold that instrument is overloaded.

• Test on the instrument

To reproduce the error:

@jenshnielsen @astafan8

[codecov]

Codecov Report

Merging #1929 into master will decrease coverage by <.01%.
The diff coverage is 50%.

@@            Coverage Diff             @@
##           master    #1929      +/-   ##
==========================================
- Coverage   68.73%   68.72%   -0.01%     
==========================================
  Files         155      155              
  Lines       19791    19798       +7     
==========================================
+ Hits        13603    13607       +4     
- Misses       6188     6191       +3

[jenshnielsen]

thanks @lakhotiaharshit

Does this work if you read out a buffer with multiple datapoints?

[jenshnielsen]

What happens with negative voltages. Is that even supported. Does it return a negative value?

[lakhotiaharshit]

thanks @lakhotiaharshit
Does this work if you read out a buffer with multiple datapoints?

Yes, I think I have fixed that part in the code. I am also planning to test it on the actual instrument in the evening to have the final confirmation. Do not know about the negative voltage though.

[astafan8]

Looks good! let know once the final testing on the instrument is done.

wrt, negative values - would be nice to catch it in this PR as well, but i'd say it's a nice-to-have for now.

[lakhotiaharshit]

What happens with negative voltages. Is that even supported. Does it return a negative value?

Yes, fixed the PR to support negative voltage overload also.

[jenshnielsen]

So the question that I didn't really raise but is relevant when you consider positive and negative values is. Should this not return +- inf rather than nan ?

[astafan8]

looks great! and thanks for the tests!

[jenshnielsen]

Looks great but something seems to be up with the tests.

[lakhotiaharshit]

Looks great but something seems to be up with the tests.

Individually all tests are running fine however when we run together all the test after the test below are failing. I do not know the reason at the moment.

@pytest.mark.parametrize("val_volt", ['10, 9.9e37, -9.9e37'])
def test_get_timetrace(driver_volt):
driver_volt.timetrace_npts(3)
voltage = driver_volt.timetrace.get()
assert (voltage == np.array([10.0, np.inf, -np.inf])).all()

[lakhotiaharshit]

Following Mikhail's nice suggestion. I have ignored that particular test at the moment. The is marked not to run and a small reason is added. I will open a new card in qcodes debt so that at later point we can fix that test.