René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"Nyquist plot rescaling is confusing","articleBody":"Hi - not sure I fully understand how the indentation around an imaginary axis pole is being done for the nyquist plot.\n\nLooking at freqplot.py L 1449:1555 I see:\n```\n# Add points for half/quarter circle around pole frequency\n# (these will get indented left or right below)\nsplane_contour = np.concatenate((\n    splane_contour[0:first_point+1],\n    (1j * np.linspace(\n        start_freq, p.imag + indent_radius, indent_points)),\n    splane_contour[last_point:]))\n\nbut this seems to make all the added points imaginary - what I would have expected to see would be a quarter/half circle with\nreal and imaginary parts -- something like:\n\n  if indent_direction == 'right':\n      direction = 1\n  else:\n      direction = -1\n  angle_start = [0 if start_freq == 0 else -np.pi/2][0]\n  angle_end = np.pi/2\n\n  splane_contour = np.concatenate((\n      splane_contour[0:first_point+1],\n      p + direction*indent_radius*np.exp(direction*1j * np.linspace(\n          angle_start, angle_end, indent_points)),\n      splane_contour[last_point:]))\n```\nwith the phase being from -pi/2:pi/2 (unless at origin) with a direction determined by the indent direction \nand np.exp(...) creating the small complex valued quarter/half circle centered at pole p (on imag axis)\n\nthis change seems to break the encirclement math that follows in the code, so I cannot provide a working example","author":{"url":"https://github.com/JonHowMIT","@type":"Person","name":"JonHowMIT"},"datePublished":"2025-03-24T13:24:55.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":9},"url":"https://github.com/1143/python-control/issues/1143"}