4s-5V Buck Converter Rev 1 August 2, 2021 Testing
- Kevin Li
Overview
The 4s-5V Buck Converter Rev 1 PCBs were tested on August 2, 2021 in Engineering 5 at the University of Waterloo.
Purpose
The purpose of testing the 4s-5V Buck Converter Rev 1 PCB was to validate the functionality of the device and to analyze its efficiency when subjected to varied load current at different fixed input voltages. The parameters listed below were monitored:
Input Voltage
Input Current
Output Voltage
Output Current
Temperature
The first four items above were used to calculate input power, output power, and efficiency.
Test Equipment
Listed below are the items that were used for testing:
DC Power Supply
Electronic Load
Digital Multimeter (DMM)
Thermocouple
Wires
For automated testing, the items listed below are also included:
Computer
Cables (between computer and DC power supply, and between computer and electronic load)
Automated Testing Scripts
Test Setup
A model of the test setup is shown in the figure below:
As shown in the figure above, the input terminals of the 4s-5V Buck Converter Rev 1 PCB, “VBAT” and “GND”, are connected to the positive and negative terminals respectively of the DC power supply. The output terminals, “5V” and “GND”, are connected to the positive and negative terminals respectively of the electronic load.
For automated testing, the DC power supply and electronic load were connected to a computer that could compile the necessary scripts required to automate the process. The cables required to make these connections are dependent on the ports of the computer, DC power supply, and electronic load.
A thermocouple connected to a DMM was used to measure the temperature of important components, namely the buck converter IC and output inductor, when needed.
The test equipment and setup is shown in the figure below:
Test Procedure
Manual Test Procedure
The 4s-5V Buck Converter Rev 1 PCB was manually tested by performing load regulation tests at different fixed input voltages. The load current was varied from 0A to 5A in 0.5A increments. Once the load current reached 4A, the increment was reduced to 0.25A for greater precision in the potential event of board failure. These load regulation tests were performed at the minimum, nominal, and maximum input voltages expected to be produced by a 4s Li-Po battery, which correspond to 10V, 14.8V, and 16.4V respectively. At each step, the input voltage, input current, output voltage, and output current were recorded.
The steps for this procedure are outlined in chronological order below:
Configure the test equipment as specified in the “Test Setup” section.
Set the DC power supply to one of the key fixed input voltages (minimum, nominal, or maximum value).
Set the load current (output current) of the electronic load to 0A and record the input voltage, input current, output voltage, and output current.
Increase the load current in 0.5A increments and record the input voltage, input current, output voltage, and output current. When the load current reaches the 4A threshold, decrease the increments to 0.25A, or even 0.1A if failure is noticed between two specific load currents.
Repeat steps 2-4 at another input voltage of interest.
Automated Test Procedure
The 4s-5V Buck Converter Rev 1 PCB was automatically tested by performing a load regulation test at a fixed input voltage of 16.4V. The load current was varied from 0A to 5A in 0.05A increments. The “dc_dc_test.py” script (courtesy of A2D/Micah!) obtained here was executed to perform this test. All test results were automatically formatted to a .csv file.
The steps for this procedure are outlined in chronological order below:
Configure the test equipment as specified in the “Test Setup” section with the cable connections required for automation between the computer and the DC power supply and electronic load.
Execute the script and input values for the following parameters: Minimum Load Current, Maximum Load Current, Load Current Step Size, and Delay Between Load Steps.
Open and analyze the data found in the generated .csv file.
Results
Manual Testing Results
The data obtained from manual testing can be found in the “4s-5V Buck Converter Rev 1 August 2, 2021 Manual Test Data” spreadsheet attached below:
Manual testing yielded the following load regulation plots shown in the figure below:
The Load Current (A) vs. Output Voltage (V) plots are shown in the figure below:
The Load Current (A) vs. Input Current (A) plots are shown in the figure below:
It must be noted that for the manual load regulation test with the input voltage fixed at 16.4V, the buck converter IC began to undergo thermal shutdown mode at a load current of 4.83A. This was noticed when various parameters began to fluctuate up and down in magnitude. This observed phenomenon was consistent with the IC’s non-latch thermal shutdown protection. For non-latch protection, the IC will continuously attempt to re-operate once its temperature passes below the thermal shutdown temperature threshold (for the TPS565201, T_SDN = 172°C).
Fun note: For one of the boards, when manually attempting to read the temperature of the output inductor using a thermocouple, I accidentally shorted the input voltage and switch node nets. The pads flared up and the buck converter IC fried up.
Automated Testing Results
The data obtained from automated testing can be found in the “4s-5V Buck Converter Rev 1 August 2, 2021 Automated Test Data” spreadsheet linked below:
Automated testing yielded the following load regulation plot shown in the figure below:
The Load Current (A) vs. Output Voltage (V) plot is shown in the figure below:
The Load Current (A) vs. Input Current (A) plot is shown in the figure below:
Discussion
The efficiency of the 4s-5V buck converter remained above 88% within a load current range of 0A-4.5A. Consistent with the TPS565201 datasheet, the efficiency of the device decreased with increasing load current.
The output voltage of the 4s-5V buck converter was calibrated to have a nominal value of 5.1V instead of 5V. This was done to be considerate of the change in output voltage tolerance as load current differs. For all load currents between 0A and 4.5A, the output voltage remained above 5V. With this data acquired, its possible that a calibrated nominal output voltage value of 5.05V would be more suitable to achieve a strong 5V output voltage. Consistent with the TPS565201 datasheet, the output voltage decreased with increasing load current.
The input current was shown to increase linearly with load current. This is consistent with our understanding of the relationship between the two variables when both input voltage and output voltage are sufficiently constant:
Conclusion
The design fulfilled its desired functionalities within the appropriate load current range it is intended to be subject to.