Battery Testing Plan - MSXV

Purpose: Validate that our manufacturing procedure yields a module that can function in all realistic discharge scenarios

Tests:

• Module Resistance

  • See Single Cell Testing - Data Processing Guide for a more detailed overview of the intricacies of DCR measurements

  • Connect the eload to the module with a 10A capable connection

  • Measure the voltage of the pack at the busbars using the Keysight DMM

  • Apply a load of 5A with a slew rate of 0.25A/us with an eload, observe the immediate voltage drop of the pack.

    • Over time, chemical processes will cause the voltage to further sag over time. This must not be accounted for in the DCR measurement, so use the first measurement you observe.

  • Use Keysight multimeter to check resistance on terminals of battery module, with integration time of 20ms.

  • Test: Measure module internal resistance. Follow this tutorial: 18650 Li-ion Internal Resistance Explained & Measured

  • The cells in 4S8P configuration alone, based off of the measured internal resistance of ~22mOhm/cell, should have an IR of about 11mOhms

  • Cells should not have greater than a 60mV drop when loaded with 5A

  • Information gained: The measured internal resistance should tell use how much resistance is introduced from out spotwelds and busbars (indication of how good our spotwelds are)

• Nominal Current/Balance Test

  • Insert thermistor into the center of the module, in the top third

  • Measure and record the voltage of each cell to at least 1mV precision

  • Connect the eload with a 30A capable connection (2 thick alligator clips in parallel)

  • Apply a 30A load until the module drops below 12V (empty)

    • The temperature rise should be about 10C after 15 mins

  • Recharge the module to 14.5V

  • Measure the voltage of each cell, the imbalance should be the same as before, within 2mV

• High Current Test

  • Our modules are rated for a max discharge of 58.2A

  • Test: Draw 58.2A from module for 10 min

  • Information gained: Temperature rise from max discharge, visual observations of module in case anything unexpected occurs

Test set-up:

Attach module to module connections with M4 bolts, Belville disc springs, and regular washers in this stack up ( Nut - Washer - Busbar - Contact Grease - Busbar - Washer - Springs - Bolt ). Torque the M4 connections to 2.4 N*m.

This way we can see from the tests how good the connections between modules are (IR introduced) and if the module-module connections introduce any issues.

Open

Module w/ Module-Module connections attached

To validate the "completed" battery modules, it is essential to conduct a full charge-discharge test while monitoring key system parameters such as voltage and temperature. While the continuous discharge of modules at 40+A is unlikely, this validation procedure functions as a stress test.

Initiate the procedure by charging the module to 16.6V (below the maximum to provide a health-span buffer).

Charging

The modules will be charged to

Discharge Testing

Open

Equipment

• BK Precision 8600 E-load

• Rigol DL3031 E-load

• 6, 18AWG wires with banana connectors

• A2D DAQ w/ thermistors

• USB to USB B (Data-Transfer Cable)

• Windows Laptop

Software Setup

GitHub - kostubhagarwal/module_test_data_acquisition

Procedure

1. Connect A2D DAQ to Windows laptop.

2. Run serial_to_csv.py to check the functionality of the thermistors (only working thermistors will show readings on the terminal).

3. Pause (Ctrl-C) serial_to_csv.py, then complete the physical setup.

4. Connect working 6 thermistors to the battery module busing

   1. use electrical or kapton tape. ensure good contact!

   2. attach to 6 different central locations on the module as shown

5. Connect E-Loads to battery module as shown DONT MESS UP THE POLARITY OF TERMINALS

   1. two wires (one for each terminal) to connect BK Precision 8600 E-load to the battery module

   2. use four wires (double-up for each terminal, as wires are not rated for 30A) to connect Rigol DL3031 E-load to the battery module

6. Set BK Precision 8600 E-Load to 10A CC draw

7. Set Rigol DL3031 E-Load to 30A CC draw

8. Run serial_to_csv.py

9. Start both e-loads

10. Once modules reach 11V (to prevent draining module past safe spot) stop the e-loads

11. Terminate (Ctrl-C >> 2) serial_to_csv.py

12. Save the 'data.csv' file with a different name, as it will be overwritten during the next run of serial_to_csv.py.

13. Follow charging procedure to charge battery pack to 13.2V, this leaves cells at 3.3V which is a good storage voltage.

Storage Charge

Why?

Modules will be charged at 22A to 17V (4.25V per cell). We charge to 4.25V because it yields 105-110% of the rated energy storage BU-808: How to Prolong Lithium-based Batteries . Charging to this higher voltage is allowed because it is still within the spec of our LG M50s (4.2V +- 0.05V). The downside to charging to a high voltage is reduced cyclability, but this is irrelevant to our low cycle use-case.

Another concern is storage. Cell voltage and temperature have a great effect on recoverable capacity.

BU-808: How to Prolong Lithium-based Batteries

At 4.25V (105-110% charge), the recoverable capacity will be even lower.

Therefore, we will discharge test fully charged modules as soon as possible to avoid prolonged storage at high SOC. After a full discharge, modules will be recharged to a storage voltage of 14.8V (3.7V) per cell to minimize capacity loss and self-dischargeBU-702: How to Store Batteries.

How?

To discharge the modules, use the battery test mode on the RIGOL DL3031 E-Load.

1. Follow the instructions in Chapter 2: Application Function: Battery Test Function in the RIGOL User Manual to enter the Battery Test Mode.

2. Set discharge current to 8A (approx 0.2C per cell, the standard discharge in the cell datasheet to minimize stress on the module).

3. Set V_Stop to 14.3V

   1. We want to discharge to 3.7V per cell, which would be a module voltage of 14.7V. But keep in mind, voltage of a battery sags due to internal resistance. So why 14.3? Through experimentation, the internal resistance of a module is approximately 50mOhms. We are discharging at 8A. The voltage drop from the load would be V(drop) = IR = 8(0.05) = 0.4V.

4. Start Discharge