from __future__ import print_function import time import numpy as np import pandas as pd import zhinst.utils import matplotlib.pyplot as plt from matplotlib import ticker, cm import matplotlib as mpl mpl.rcParams['text.color'] = '#009ee0' mpl.rcParams['axes.labelcolor'] = '#009ee0' mpl.rcParams['xtick.color'] = '#009ee0' mpl.rcParams['ytick.color'] = '#009ee0' device_id = 'dev4220' # replace with your device SN here apilevel_example = 6 err_msg = "This example only supports instruments with IA option." (daq, device, _) = zhinst.utils.create_api_session(device_id, apilevel_example, required_options=['IA'], required_err_msg=err_msg) zhinst.utils.api_server_version_check(daq) zhinst.utils.disable_everything(daq, device) # We use the auto-range example before the next point imp_index = 0 exp_settings = [['/%s/imps/%d/enable' % (device, imp_index), 1], ['/%s/imps/%d/mode' % (device, imp_index), 0], ['/%s/imps/%d/auto/output' % (device, imp_index), 1], ['/%s/imps/%d/auto/bw' % (device, imp_index), 1], ['/%s/imps/%d/bias/enable' % (device, imp_index), 1], ['/%s/imps/%d/bias/value' % (device, imp_index), 0], ['/%s/imps/%d/freq' % (device, imp_index), 1000], ['/%s/imps/%d/auto/inputrange' % (device, imp_index), 1], ['/%s/imps/%d/output/amplitude' % (device, imp_index), 0.1], ['/%s/imps/%d/output/range'% (device, imp_index), 10], ['/%s/imps/%d/model'% (device, imp_index), 0]] # Subscribe to the impedance sample node path. path = '/%s/imps/%d/sample' % (device, imp_index) daq.set(exp_settings) daq.sync() amp_list = [] the_list = [] c_list=[] '''for y in np.logspace(3, 6, 101): #freq daq.setDouble('/%s/imps/%d/freq' % (device, imp_index), y) for x in np.linspace(0.0, 2.0, 101): #bias/value, 2V is enough to turn on the LED daq.setDouble('/%s/imps/%d/bias/value' % (device, imp_index), x)''' # if sweep DC-offset, MFIA always auto-ranges, this is not good the instrument lifetime for x in np.linspace(1.0, 1.7, 201): #bias/value, 2V is enough to turn on the LED daq.setDouble('/%s/imps/%d/bias/value' % (device, imp_index), x) for y in np.logspace(3, 6, 201): #freq daq.setDouble('/%s/imps/%d/freq' % (device, imp_index), y) daq.sync() #synchronize settings daq.subscribe(path) sleep_length = 1.0 time.sleep(sleep_length) #this is to ensure the transient is not captured poll_length = 0.1 # [s] poll_timeout = 500 # [ms] poll_flags = 0 poll_return_flat_dict = True data = daq.poll(poll_length, poll_timeout, poll_flags, poll_return_flat_dict) #in the 19.05 API, replace with flat=True daq.unsubscribe('*') impedanceSample = data[path] amp_ = np.abs(np.mean(impedanceSample['z'])) the_ = np.angle(np.mean(impedanceSample['z']))*180/np.pi #convert radian into deg, phase can also be extracted from data directly c_ = np.mean(impedanceSample['param1']) ##this presumes MFIA is using R||C model amp_list.append(amp_) the_list.append(the_) c_list.append(c_) amp_array = np.array(amp_list).reshape(201,201) #convert into 2D matrix, and flip later x and y for easy plotting the_array = np.array(the_list).reshape(201,201) c_array = np.array(c_list).reshape(201,201) real_array = amp_array*np.cos(the_array*np.pi/180) imag_array = amp_array*np.sin(the_array*np.pi/180) '''This part can be used to show the original 2D amplitude and theta plots plt.close('all') X, Y = np.meshgrid(np.linspace(1.0, 1.7, 201), np.logspace(3, 6, 201)) fig0, ax0 = plt.subplots() cs0 = ax0.contourf(X, Y, np.transpose(amp_array), locator=ticker.LogLocator(), cmap=cm.PuBu_r) ax0.set_title('Impedance (Ohm)') ax0.set_xlabel('DC bias (V)') ax0.set_yscale('log') ax0.set_ylabel('Frequency (Hz)') ax0.autoscale() cbar = fig0.colorbar(cs0) plt.draw() plt.show() fig1, ax1 = plt.subplots() cs1 = ax1.contourf(X, Y, np.transpose(the_array), locator=ticker.LinearLocator(), cmap=cm.RuBu_r) ax1.set_title('Phase (Deg)') ax1.set_xlabel('DC bias (V)') ax1.set_yscale('log') ax1.set_ylabel('Frequency (Hz)') ax1.autoscale() cbar = fig1.colorbar(cs1) ''' ###here the data is excerpted only up to 1.56 V to avoid meaningless negative capacitance values at high DC bias X, Y = np.meshgrid(np.linspace(1.0, 1.56, 160), np.logspace(3, 6, 201)) amp_array_sub=amp_array[0:160] the_array_sub=the_array[0:160] c_array_sub=c_array[0:160] real_array_sub = real_array[::20] imag_array_sub = imag_array[::20] fig0, ax0 = plt.subplots() cs0 = ax0.contourf(X, Y, np.transpose(amp_array_sub), locator=ticker.LogLocator(), cmap=cm.PuBu_r) ax0.set_title('Impedance (Ohm)') ax0.set_xlabel('DC bias (V)') ax0.set_yscale('log') ax0.set_ylabel('Frequency (Hz)') ax0.autoscale() cbar = fig0.colorbar(cs0) plt.draw() plt.show() fig1, ax1 = plt.subplots() cs1 = ax1.contourf(X, Y, np.transpose(the_array_sub), locator=ticker.LinearLocator(), cmap=cm.PuBu_r) ax1.set_title('Phase (Deg)') ax1.set_xlabel('DC bias (V)') ax1.set_yscale('log') ax1.set_ylabel('Frequency (Hz)') ax1.autoscale() cbar = fig1.colorbar(cs1) plt.draw() plt.show() fig2, ax2 = plt.subplots() cs2 = ax2.contourf(X, Y, np.transpose(c_array_sub), locator=ticker.LinearLocator(20), cmap=cm.PuBu_r) ax2.set_title('Capacitance (F)') ax2.set_xlabel('DC bias (V)') ax2.set_yscale('log') ax2.set_ylabel('Frequency (Hz)') ax2.autoscale() cbar = fig2.colorbar(cs2) plt.draw() plt.show() ##Nyquist plot fig3, ax3 = plt.subplots() plt.plot(np.transpose(real_array_sub), -1*np.transpose(imag_array_sub)) #plot in 0.07V DC step plt.ticklabel_format(style='sci', scilimits=(0,0)) ax3.set_title(' Nyquist plot at different DC bias at a step of 0.07 V') ax3.set_xlabel('Real Z (Ohm)') ax3.set_ylabel('-Imaginary Z (Ohm)') plt.axis('equal') ax3.autoscale() ax3.set(xlim=(0, 6000000), ylim=(0,6000000)) #adjust to fit into the scale of the largest impedance range plt.draw() plt.show() daq.setDouble('/%s/imps/%d/bias/value' % (device, imp_index), 0) # to turn off the LED daq.setDouble('/%s/imps/%d/freq' % (device, imp_index), 1000) pd.DataFrame(amp_array).to_csv('amp.csv') #save for post processing. These are enough to reconstruct other impedance parameters pd.DataFrame(the_array).to_csv('the.csv') pd.DataFrame(c_array).to_csv('c.csv')