In my extensive work with lithium-ion batteries, I have consistently observed that the accuracy and consistency of performance testing are highly sensitive to the connection methods employed. Lithium-ion batteries, as the cornerstone of modern energy storage systems, require precise evaluation to ensure reliability and safety. This study delves into how specific connection parameters—namely, the voltage sense line placement, washer dimensions, and fastener torque—affect key performance metrics such as capacity, temperature rise, and DC internal resistance. The findings underscore that meticulous control over these connection details is not merely a procedural formality but a critical factor in obtaining reliable data for lithium-ion battery characterization.
The widespread adoption of lithium-ion batteries in electric vehicles, portable electronics, and grid storage necessitates rigorous testing protocols. Performance parameters like DC internal resistance and thermal behavior are vital indicators of a lithium-ion battery’s health and efficiency. Commonly, a four-terminal (Kelvin) connection method is preferred over a two-terminal approach because it theoretically eliminates the influence of lead resistance. However, in practice, the implementation of the four-terminal method introduces nuances related to contact resistance and measurement geometry that can significantly skew results. My investigation aims to quantify these effects and provide guidelines for standardizing connection practices in lithium-ion battery testing.
The fundamental principle behind the four-terminal method for a lithium-ion battery involves separating the current-carrying and voltage-sensing paths. The current leads deliver the charge or discharge current, while the voltage sense lines measure the potential difference directly at the terminals of the lithium-ion battery. The measured DC internal resistance, \( R_{DC} \), is then calculated using Ohm’s law from the instantaneous voltage change, \( \Delta V \), during a current pulse, \( \Delta I \):
$$ R_{DC} = \frac{\Delta V}{\Delta I} $$
However, this calculation assumes that the voltage is measured at the exact point where the current enters and leaves the lithium-ion battery cell. Any impedance in the connection interface, known as contact resistance \( R_{contact} \), contributes to additional voltage drop and Joule heating. The total power loss \( P_{loss} \) at a connection point with current \( I \) is:
$$ P_{loss} = I^2 R_{contact} $$
This power dissipation directly influences the temperature rise of the lithium-ion battery during testing. Therefore, minimizing and controlling \( R_{contact} \) is paramount for accurate thermal and electrical assessment of any lithium-ion battery.
To systematically evaluate these effects, I designed a series of experiments using different lithium-ion battery cell formats, including prismatic hard-case and pouch-type cells. The primary variables were the voltage sense line attachment point, the diameter of the washers used in bolted connections, and the applied fastener torque. Each test sequence was repeated to ensure statistical significance, and data was recorded using high-precision data acquisition systems. The core test procedures for capacity and DC internal resistance are summarized in the following tables.
| Step | Action | Details |
|---|---|---|
| 1 | Standard Charge | Constant current (CC) charge at the target current until the cell voltage reaches \( V_{max} \), then constant voltage (CV) charge until the current decays to 0.05C. Follow by a rest period of 0.5 hours. |
| 2 | Standard Discharge | Constant current (CC) discharge at the target current until the cell voltage reaches \( V_{min} \). Follow by a rest period of 0.5 hours. |
| Step | Action | Parameters |
|---|---|---|
| 1 | Standard Charge | 1C CC charge to \( V_{max} \), then CV charge until current ≤ 0.05C. Rest for 1 hour. |
| 2 | SOC Adjustment | Adjust the State of Charge (SOC) to 90%, 50%, and 20% (accounting for pulse energy). |
| 3 | Rest | Rest for 30 minutes. |
| 4 | Discharge Pulse | Apply a constant current discharge pulse of \( I_{discharge} = 2C \) for 10 seconds. Record voltage and current. Abort if any cell voltage reaches \( V_{min} \). |
| 5 | Rest | Rest for 30 seconds. |
| 6 | Charge Pulse | Apply a constant current charge pulse of \( I_{charge} = 2C \) for 10 seconds. Record voltage and current. Abort if any cell voltage reaches \( V_{max} \). |
| 7 | Rest | Rest for 10 minutes. |
| 8 | Cycle | Repeat steps 5 through 10 until all SOC points are tested. |
The first critical factor I examined was the placement of the voltage sense lines. In a typical test setup for a lithium-ion battery, the sense lines can be attached using alligator clips to different points: directly on the cell’s terminal busbar (the ideal reference point), on the connecting bolt head, or on the load cable from the test equipment. Using a 150Ah lithium iron phosphate (LFP) prismatic lithium-ion battery, I performed capacity and DC internal resistance tests with the sense lines at these three distinct locations.
| SOC Level | Sense Line on Cell Busbar (mΩ) | Sense Line on Connecting Bolt (mΩ) | Sense Line on Load Cable (mΩ) |
|---|---|---|---|
| 90% | 0.492 | 0.589 | 0.687 |
| 50% | 0.507 | 0.606 | 0.702 |
| 20% | 0.588 | 0.686 | 0.780 |
The data reveals a substantial deviation. At 50% SOC, the measured DC internal resistance of the lithium-ion battery was 0.507 mΩ when sensed at the busbar, but it increased to 0.702 mΩ when sensed at the load cable—a difference of 0.195 mΩ. This discrepancy stems from the inclusion of parasitic resistances in the measurement path. To quantify this, I simultaneously measured the voltage drop across different segments during a 135A discharge of the lithium-ion battery.
| Cell ID | Voltage Drop Between Bolt and Busbar (mV) | Voltage Drop Between Load Cable and Busbar (mV) | Voltage Drop Between Busbars (mV) | |||
|---|---|---|---|---|---|---|
| Positive | Negative | Positive | Negative | Positive | Negative | |
| Cell 1 | 2.035 | 1.550 | 2.540 | 2.085 | 0.360 | 0.420 |
| Cell 2 | 3.525 | 2.450 | 4.205 | 3.345 | 0.370 | 0.655 |
The voltage drop between the load cable and the busbar is the largest, confirming that attaching sense lines to the load cable incorporates significant contact and lead resistances into the measurement for the lithium-ion battery. This directly impacts other metrics. For instance, during a 1C capacity test on the same lithium-ion battery, the average discharge voltage platform was 26mV higher, and the charge voltage platform was 71mV lower when sense lines were on the busbar compared to the load cable. This can be modeled by considering the total measured voltage \( V_{meas} \) as:
$$ V_{meas} = V_{cell} + I \cdot R_{parasitic} $$
where \( V_{cell} \) is the true terminal voltage of the lithium-ion battery, and \( R_{parasitic} \) is the sum of resistances between the sense point and the cell terminal. For accurate characterization of a lithium-ion battery, sense lines must be attached as close as physically possible to the active cell terminals.
The second factor investigated was the size of the washers used in bolted connections for the lithium-ion battery. The contact area between the terminal, washer, and current lug profoundly affects the current density and thus the contact resistance and localized heating. I tested a 49Ah ternary (NMC) pouch lithium-ion battery under 2C charging conditions, using washers with diameters of 12mm, 16mm, 20mm, 25mm, and 30mm. The temperature rise on the negative tab was monitored.
The temperature profiles clearly indicated that smaller washers led to higher temperature rises. The peak temperature approached 34°C with a 12mm washer and 33°C with a 16mm washer. In contrast, washers of 20mm, 25mm, and 30mm diameters resulted in similar, lower peak temperatures around 31°C. This can be explained by the contact resistance model where the constriction resistance \( R_c \) is inversely proportional to the effective contact area \( A_c \):
$$ R_c \propto \frac{1}{A_c} $$
For a lithium-ion battery tab of approximately 20mm width, a washer diameter exceeding 20mm does not substantially increase the conductive contact area, explaining the plateau in performance. The heat generation rate \( \dot{Q} \) at the interface for a lithium-ion battery under current \( I \) is:
$$ \dot{Q} = I^2 R_c $$
Therefore, using an appropriately sized washer that maximizes contact area with the lithium-ion battery terminal is crucial to minimize parasitic heating and obtain accurate temperature rise data for the lithium-ion battery itself.
The third and equally critical parameter is the fastener torque applied to the bolted connection of a lithium-ion battery. Torque determines the contact pressure, which directly influences the microscopic conformity of the mating surfaces and thus the contact resistance. Using a 72Ah LFP prismatic lithium-ion battery with M6 bolts, I measured the contact resistance between the cell busbar and the load cable using a micro-ohmmeter at various torque levels.
| Applied Torque (Nm) | Measured Contact Resistance (mΩ) |
|---|---|
| 0.1 | 0.155 |
| 1 | 0.041 |
| 3 | 0.035 |
| 5 | 0.030 |
| 8 | 0.028 |
| 10 | 0.027 |
| 12 | 0.029 |
The contact resistance decreases sharply as torque increases from a very low value, reaching a minimum of 0.027 mΩ at 10 Nm for this specific lithium-ion battery connection. Beyond this optimum point, at 12 Nm, the resistance slightly increases. This non-monotonic behavior is classic in joint physics: insufficient torque leads to a small real contact area, while excessive torque can cause plastic deformation of soft materials (like aluminum busbars on a lithium-ion battery), potentially degrading the contact or reducing the area. The relationship between contact resistance \( R_{contact} \) and applied force \( F \) (related to torque) is often empirically modeled for such interfaces:
$$ R_{contact} = k \cdot F^{-n} $$
where \( k \) and \( n \) are constants dependent on the materials and surface finish of the lithium-ion battery terminal and connector.
To see the operational impact on the lithium-ion battery, I performed 1C discharge capacity and DC internal resistance tests at different torque settings. The results for the lithium-ion battery are tabulated below.
| Torque (Nm) | Initial Temp. (°C) | Final Temp. (°C) | Temp. Rise (°C) | V-drop Pos. (mV) | V-drop Neg. (mV) |
|---|---|---|---|---|---|
| 1 | 28.9 | 37.29 | 8.39 | 11.993 | 12.554 |
| 3 | 28.54 | 36.18 | 7.64 | 3.955 | 4.452 |
| 5 | 28.05 | 35.34 | 7.29 | 1.817 | 1.972 |
| 10 | 28.35 | 35.15 | 6.80 | 1.428 | 1.408 |
| 12 | 28.33 | 35.62 | 7.29 | 1.527 | 1.554 |
| Torque (Nm) | DC Internal Resistance (mΩ) |
|---|---|
| 1 | 1.354 |
| 3 | 1.251 |
| 5 | 1.201 |
| 10 | 1.178 |
| 12 | 1.199 |
The data shows a clear trend: both the temperature rise of the lithium-ion battery during discharge and its measured DC internal resistance decrease as torque increases to an optimum value (10 Nm), then show a slight degradation. The voltage drop across the connections follows the same pattern. This confirms that the contact resistance significantly affects the perceived performance of the lithium-ion battery. The total measured internal resistance \( R_{meas} \) of the lithium-ion battery system is the sum of the cell’s true internal resistance \( R_{cell} \) and the connection resistances \( R_{conn} \):
$$ R_{meas} = R_{cell} + R_{conn} $$
where \( R_{conn} = R_{contact, pos} + R_{contact, neg} + R_{leads} \). Inconsistent torque application leads to variable \( R_{conn} \), causing poor reproducibility in testing lithium-ion batteries.
Expanding on the theoretical implications, the thermal dynamics of a lithium-ion battery under test can be modeled more comprehensively. The temperature increase \( \Delta T \) of a lithium-ion battery cell due to Joule heating from connection losses over a time period \( t \) can be approximated by:
$$ \Delta T \approx \frac{I^2 R_{contact} t}{m C_p} $$
where \( m \) is the mass and \( C_p \) is the specific heat capacity of the lithium-ion battery or its components. This equation highlights why even milliohm-level contact resistances can cause measurable temperature differences during high-current testing of lithium-ion batteries.
Furthermore, the interplay between these factors must be considered. For instance, using a large washer with insufficient torque may not yield the full benefit of increased area if the contact pressure is too low to penetrate surface oxides on the lithium-ion battery terminal. A holistic approach is needed for connecting to a lithium-ion battery. Based on my findings, I propose a standardized connection protocol for testing lithium-ion batteries: 1) Voltage sense lines must be attached directly to the cell’s terminal busbars using a low-resistance method (e.g., spot-welded tabs or dedicated sense points). 2) Washer diameter should be selected to match or slightly exceed the width of the lithium-ion battery’s terminal tab to maximize contact area without overhang. 3) A specific, optimized fastener torque must be determined for each cell format and connection hardware combination and strictly adhered to for all tests on that lithium-ion battery type. This torque should be validated by measuring the connection resistance to ensure it is at the minimum plateau of the resistance-torque curve.
In conclusion, my investigation demonstrates that the connection methodology is a significant source of error and variability in the performance evaluation of lithium-ion batteries. The placement of voltage sense lines can cause deviations in DC internal resistance measurements exceeding 0.195 mΩ for a lithium-ion battery. The size of washers and the applied fastener torque directly govern contact resistance, which in turn affects the recorded temperature rise and capacity voltage profiles of the lithium-ion battery. These effects are non-negligible and must be controlled to achieve accurate, repeatable, and comparable test results for lithium-ion batteries. As the demand for precise lithium-ion battery grading and modeling grows, standardizing these connection details becomes increasingly critical for engineers and researchers working with lithium-ion batteries across the industry. Future work could involve developing automated connection systems with integrated force and resistance feedback to ensure perfect reproducibility for every lithium-ion battery test.