update forked #4
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
gives same results than ANSYS, but one test is failing. Need to correct doc equation as well.
but still same problem z <-> y
add the keyword argument rcond=None to np.linalg.lstsq function in order to suppress the FutureWarning occuring since numpy version 1.14.0
uses numpy asser_allclose method instead of pure == and set a absolute tolerance of 1e-3. This decrease in float values precision is due to the correction of power waves for complex char impedance.
Modified the z/y/s transformation function to deal with either power-waves (default) or pseudo-waves definition of the scattering parameter. A new parameter 's_definition' has also been added to the Network class to define the chosen definition. To be tested.
replaced @ operators with npy.matmul Fixed file encoding for Python 2 compatibility
replaced @ with npy.matmul
…3.6 (?) So I've increase a little bit the absolute tolerance from 1e-3 to 1.1e-3.
…r-waves definition
- renamed s_definition to s_def - propagate s_def in Network.copy - propagate s_def in Network.renormalize and rf.renormalize_s - suppressed rf.renormalize_s_pw - propagate s_def in calibration EightTerm.renormalize - decrease absolute tolerance in calibration test (lower than before all these modifications) -
- Modified `connect` to allow networks with frequency subsets - Changed the `__getitem__` to avoid interpolation with explicit frequency indices. - Updated the `__getitem__` to allow slicing using Frequency objects. - Updated the subset slice to use a copy_subset function
The command did not set the name for the frequency extender without the single quotation mark. Works now: ``` In [1]: from skrf.vi.vna import ZVA In [2]: vna = ZVA(address='192.168.1.83', visa_library='@py', interface='SOCKET') In [3]: vna.echo = 1 In [4]: vna.scpi.query_converter_name() :SENS1:FREQ:CONV:DEV:NAME? Out[4]: 'ZC220' In [5]: vna.scpi.set_converter_name(1, 'ZC330') :SENS1:FREQ:CONV:DEV:NAME 'ZC330' In [6]: vna.scpi.query_converter_name() :SENS1:FREQ:CONV:DEV:NAME? Out[6]: 'ZC330' ```
- keep previous implementation (as s_def = 'traveling') - add some test in s_def to raise ValueError is s_def is not set correctly - add S_DEFINITIONS and S_DEF_DEFAULT constants into constant.py, in order to be able to easily change the s-param definition everywhere for sure - added test in test_network for s parameter conversion when Network is created from Z or Y parameters directly
+add small real part on z0 in case of pure imaginary values to prevent numerical errors
+ add a related test case
This fixes:
[...]/skrf/io/touchstone.py in is_from_hfss(self)
417 '''
418 status = False
--> 419 if 'exported from hfss' in str.lower(self.comments):
420 status = True
421 return status
TypeError: descriptor 'lower' requires a 'str' object but received a 'NoneType'
... when comments is None. Return False instead.
ZVA: add missing single quotation mark for CONV name
is_from_hfss(): fix TypeError when comments is None. Thank you.
Presently, Circuit algo works with complex impedance but give the same results that previous s-param implementation of scikit-rf, that is sort of "traveling" waves, but not power-waves. So changed doc to precise that z0 should be real to avoid confusion.
Fix bug in write_touchstone (Issue Number 325)
Correct s-parameters calculations for complex characteristic impedance
- was made before in the work on complex z0, but disappear of the final changes... (?)
due to PR#289, the port order has changed in the reference networks used in Circuit tests
Suppress Numpy warning with lstsq in calibration.py
Integrates the content of PR proposal #289.
Release 0.15.1
Release 0.15.1 (2nd try)
Fix few minor things related with equations too close from the text
Documentation Fix
Added me
since networkx is used in `Circuit`
All 1-port Network were previously filtered in the Circuit builder in order to filter out "port" Networks, which is obviously not correct if one has to deal with 1-port Network in the Circuit. Instead, now only 'port' Networks are sorted out, and 1-port Networks are kept. Also add new tests for various 1-port devices.
fix issue 335
- Added Circuit tutorial and associated figures - Added examples on matching from Pozar book (LC and stub) - corrected reference in Time Domain - use notebook image support in One Port Tiered Calibration.ipynb instead of IPython SVG - Added svg file containing Network symbols - fix missing os import in tutorial/Networks
Documentation Improvements
See [BIPM - SI prefixes](https://www.bipm.org/en/measurement-units/prefixes.html)
Return kHz when network.frequency.unit = 'khz'
Fix bug associated to issue 341
The data from "simple_touchstone.s2p" now also checked for if the correct S-parameter is returned.
In TouchStone 1.0 files, the data order for 2-port data is: (11) (21) (12) (22) while for a higher number of ports, the data is arranged as follows (11) (12) (13) ... (1n) This is now honored in the get_sparameter_data() function's return values, if the filename ends on "s2p". The internal representation of the class is unchanged.
TS v1 2port data order
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.