Why is It Crucial to Determine the Battery's Remaining Capacity As Precisely As Possible?
In the current day, it has been discovered advantageous to use cleaning systems, robotics, military equipment, and medical equipment. Additionally, it has been discovered that measurement tools such portable equipment (like the universal cosmetics) follow the development pattern. However, because all portable devices require a battery for electricity, there hasn't been much advancement.
When it comes to estimating the system's remaining runtime, the battery can be considered the most challenging component of the supply chain. There is a need to do more critical activities due to the constant growth of portable apps, such as using a mobile phone for account management; medical equipment must be completely deployed for the monitoring of crucial data.
The Techniques for Assessing Battery Capacity
There are two different sorts of monitoring techniques employed nowadays. While the second kind is based on voltage measurement, the first type is based on the current integral (current integration). The former is based on a reasonable assumption that there is a flow of discharge whenever the battery is fully charged. One can determine the residual capacity in this scenario. When the battery is freshly recharged and is known to be fully charged, the current integration approach performs exceptionally well.
The employment of this technique in numerous current battery monitoring techniques has proved successful.
The Flaw in Using the Current Integral to Measure Battery Capacity
It is critical to emphasise that this approach has some limitations. The battery's lack of functionality in usage mode is one of its flaws. The self-discharge brought on by the internal chemical reaction will be seen whenever the battery hasn't been used for a few days after charging or if a few recharge and discharge cycles haven't fully charged the battery. The only way to fix it is by a specified equation because there are no means or procedures for self-discharge. Depending on the charging state (SOC), temperature, previous charge/discharge cycles, and other factors, different battery models self-discharge at varying rates.
Another issue with this approach is that the overall power value can only be updated when the battery is entirely depleted right after a full charge. The true capacity of the battery may have started to considerably degrade before the metre updates the actual battery value if the number of complete discharges over the battery life span is insufficient. As a result, the monitoring overestimates the energy supply throughout these cycles.
1. A chemical reaction and a battery's voltage change
2. The battery's transient voltage response is brought on by the intricate electrochemical interaction of the battery itself.
3. The main phases in charge transfer from a lithium-ion battery's electrodes (the steps of other batteries are similar).
Before reaching the surface of the particle, the charge must pass through an electrochemically active substance (anode or cathode) that stores energy in the form of electrons. These electrons are then stored in the electrolyte as ions. The time constant of the battery voltage response is connected to these chemical processes.
The available power will alter as the temperature and discharge speed change, even if the battery charge is updated under the same conditions.
Monitoring the Voltage of the Remaining Capacity
One of the early techniques is the voltage-based method. It is only necessary to measure the voltage differential between the battery's two levels. The method is based on the known relationship between battery voltage and remaining charge. Although it appears simple, the monitoring process is actually quite difficult. There won't be a direct relationship between the battery voltage and the power during the measurement unless no load is being applied. When a load is applied, the internal impedance of the battery causes a pressure drop that distorts the battery voltage (which occurs in the majority of situations where the user is concerned about capacity).
Furthermore, the battery's normal relaxation process can continue cause a consistent change in voltage within a few hours even after the load has been removed. For a variety of reasons, the pressure drop correction method based on battery impedance knowledge is still problematic.
Depending on how quickly the load is removed, the battery voltage will gradually fall over time at various rates. The lithium-ion battery's voltage relaxation during various charging states when the load is added.
We presume that the load voltage may be rectified by removing the IR pressure drop in order to account for the mistake of the voltage-based battery monitoring, and that the current SOC can then be calculated using the corrected voltage value. The R-dependence value's on the SOC is the first issue that needs to be addressed. In a virtually fully discharged state, if the average value is applied (at this time the impedance is more than ten times the state of the charge). The SOC error will be estimated to be 100%. Utilizing numerous voltmeters based on the SOC under various loads is one method to address this issue. Temperature has a significant impact on impedance as well (for every 10°C drop in temperature, impedance increases by 1.5 times).
The battery often responds in a transitory manner. This implies that the effective R-value is load time dependent, therefore it follows that we can use the internal impedance as an ohm resistor without taking time into account. This is because, even if the voltmeter takes into account the correlation between R and Soc, the change in load will result in significant inaccuracy. The range of transient errors varies from 50% in the discharge state to 14% in the charging process because the slope of the SOC (V) function depends on the SOC.
The complexity of the situation is heightened by the differences in impedance among various batteries. Even recently manufactured batteries will have low-frequency DC impedance variations of about 15%, which significantly impacts high load voltage correction. For instance, a 2Ah battery's normal DC impedance is roughly 0.15 for the typical 1/2c charge and discharge current. The worst case scenario will result in a 45mV voltage difference between the batteries, and the related SOC calculation error is 20%.
Finally, an impedance-related maximum issue develops as the battery ages. It is common knowledge that the impedance rise is much smaller than the battery's. The usual lithium-ion battery has 70 charges and discharge cycles, a single rise in DC impedance, and a loss in voltage of only 2%–3% throughout a cycle with no loads.
The voltage-based method appears to function properly with the new battery pack, however if the aforementioned criteria are ignored, a fatal error (50%) is produced after the battery pack reaches 15% of its service life (the estimated 500 charge and discharge cycles).
Battery Monitoring: Voltage and Current Techniques
The next-generation power monitoring algorithm developed by TI makes use of the advantages of the current approach and the voltage method. The business has carefully thought through this seemingly natural, yet previously complex scheme: mixing current and voltage ways to apply the most effective method in various circumstances. This method may produce accurate SOC estimation without load and power supply in a relaxation condition due to the exact correlation between the open-circuit voltage and SOC. By utilising the non-working phase, the approach also enables it simple to determine the precise "beginning position" of the Soc (any battery-powered device will have a working period).
Because the device can identify the precise SOC when it is attached, this method eliminates the necessity for self-discharge correction in the absence of a working period. The current integration approach can also be employed when the gadget is powered on and the battery is charged. Since the Coulomb metre number (coulomb-counting) has been monitoring the changes in the SOC from the start of the process, the approach does not require a complicated and inaccurate adjustment for pressure drop under load.
Can the Charge Be Completely Updated Using This Method?
Yes! The charge can be updated using the procedure. Depending on how much of the SOC was used prior to the load, how much SOC was used after the load (both were determined by voltage measurements in a relaxation state), and how much charge was transported between the two. As a result, it is simple to calculate the total power that, in the case of a particular charge change, corresponds to the SOC change. However, despite the volume of transmission and the starting situation, this is still possible (without full charge).
As a result, there is no longer a need to update the charge in certain circumstances. As a result, it aids in reducing yet another flaw in the integration method as it stands. This approach not only offers a remedy for SOC's problems, but it also gets around the impact of battery impedance and serves other goals. The entire power that corresponds to the "no-load" condition, such as the maximum charge that can be used, can be updated using this manner. The power at zero load will also decrease due to the reduction in IR, and when the load is applied, it takes less time to achieve the end-voltage value.
However, as was already established, the impedance is battery-dependent and rises sharply with battery age and longer charge and discharge intervals. It is therefore unnecessary to store it in the database. In an effort to address this issue, TI has created a specific class of integrated circuit (IC) that enables real-time impedance measuring, which allows for continuous database updating. This provides a remedy to the discrepancies between the battery's impedance and its age. The correct prediction of the voltage situation under the stated load is made possible by the ongoing update of impedance data.
The method can be used in a number of situations to help reduce the predicted error rate of electricity consumption to less than 1% and, more critically, to achieve high accuracy throughout the battery's life cycle.
Plug and play is another benefit of the adaptive algorithm. Since this information will be gathered through real-time measurement, the implementation of the algorithm does not need to specify the link between impedance, Soc, and the temperature of the database. It is no longer necessary to build the database for self-discharge correction, but it is still necessary to define the database for open-circuit voltages and SOC (including temperature) relationships.
But rather than specific design elements of the battery model, this link is acquired by the chemical properties of the positive and negative polar systems (e.g. electrolyte, separator, the thickness of active material, and additive). The V (soc,t) relationships of the majority of battery manufacturers are the same since they all employ the same active components (LiCoO2 and graphite).
When the voltage of batteries from various manufacturers is compared, particularly when there is no load, it can be deduced that their voltage value is extremely close to the variance of 5mV. The SOC inaccuracy is only 1.5 percent in the worst-case situation as well. In contrast to the hundreds of databases that are already utilised for various battery models, the development of a new battery needs the construction of a new database. This guarantees that the Power metre solution's implementation across a variety of devices and the database do not depend on the batteries being used. Even batteries made by various manufacturers do not require reprogramming. As a result, the accuracy and dependability of the battery monitoring IC Plug and Play were also increased in line with this.