Using Python to automate measurements

As a system and application engineer, I have saved countless hours through measurement automation with software such as LabVIEW. Although I've used it to build measurement applications, I've started to replace LabVIEW with Python for basic lab measurements where I don't need develop an easy-to-use GUI for others to use. When I just need to quickly take some measurements, Python lets me save them in an easy-to-read format and plot them.

To understand why, let's look at the main advantages of Python and discuss a working example of a Python application. The best way to convey the convenience and power of Python is to describe a complete, working Python automation script, such as the one I used to automate the measurement of a VR's (voltage regulator's) load-regulation curve (load regulation is the variation of the output voltage as the output current—the load—increases).

VRs are divided into two categories: zero-droop regulators and droop regulators. Zero-droop regulators have zero output resistance; the output-voltage setpoint shouldn't change with increasing output currents. On the contrary, droop regulators are said to have a 'loadline', which means they're designed to have a specific equivalent output resistance. The regulator used for this example has a zero-current output voltage of 1 V and a programmed loadline of 2.5 mΩ. Figure 1 shows the test setup.

The load current (the VR output current) is applied using a Chroma 63201 electronic load. The output current is measured by acquiring the voltage across a calibrated 4-mΩ shunt resistor. Both voltage and current are acquired using a Keysight 34970A DAQ (data-acquisition system), and both the DAQ and the electronic load communicate to a computer over a GPIB link. The goal of our measurement is to verify that the output voltage is within specification across a range of output currents; figure 2 shows the application's flowchart.

The next pages describe the code I used to make these measurements. There, you'll find links to download the code as a text file.

Basic code structure
Below, you can find the first part of the automation script code listing. In Python, comments are preceded by #:

import numpy as np # 1
import pandas as pd # 2
import visa, time # 3

chroma = visa.instrument('GPIB::2') # 4
daq = visa.instrument('GPIB::9') # 5

results = pd.DataFrame() # 6
loads = np.arange(0,20+2,2) # 7

for load in loads: # 8
# Measure the current and the voltage
# Save the results

Lines 1 to 3 import libraries that contain methods used later in the code:

Numpy is a package used for scientific computing. In this example, Numpy is used to generate the array of output-current values.
Pandas (a library for data manipulation and analysis) creates a very powerful data structure to store the results of our measurements.
Visa is the PyVISA library that we use to control our instruments.
Time is a handy library that we need to generate some time delays.