Modern high-energy density batteries such as lithium-ion is potentially dangerous when not used or charged properly. Battery safety is paramount. To ensure the device is safe and reliable, thorough stress testing of the charging circuitry is a critical task. Batter safety is crucial in many systems ranging from consumer mobile devices, to airplanes, to tractors and agricultural vehicles. Poorly-charged batteries can become excessively hot and may even catch fire, reinforcing the need for high reliability stress testing. Battery-powered devices need a charger to recharge its battery on a regular basis,and that charger must be rigorously tested to ensure it is reliable over many usage situations, from normal to extreme, safely handling all possible conditions. In addition, if the charger is powered from an USB port, it must also meet all USB specs.
Unfortunately, using real batteries can be difficult for design engineers to use to test a system’s functionality. For example, it takes several hours to charge a battery, and for thorough charger testing over many charging cycles is cumbersome. Add to that the need to test under different temperatures and in multiple usage conditions, so testing with real batteries can be a very time consuming task. A battery simulator (or emulator), will speed up these tests, and perform others not possible using actual batteries. A simulator can be a very useful equipment for saving time as well as delivering functionality during the system development and production phases by providing the required power, voltage, and current to the system under test with no apparent distinction between a real battery and the emulator.
How it works
A full-function battery simulator is in reality a power supply with the ability to sink and source current, using a special type of voltage supply that can operate in either two- or four-quadrants. In contrast, a conventional power supply can only source, but not sink, current and can only operate in the first quadrant. Figure 1 shows the four quadrants of operation. The normal, source-only, voltage supply uses only one output transistor which is meant for sourcing current and it is operating only in the first quadrant.
Figure 2 depicts a simplified diagram of a conventional power supply. The simulator on the other hand possesses two output power transistors: one to source current and one to sink it. And, it can rapidly move from sink to source current without creating a glitch. A simulator is operating in the fourth quadrant.
Figure 3 depicts the simplified equivalent circuit of an emulator. An example of a simulator is the TS250/TS200. Although they are being called Modulated Power Supply and Waveform Amplifier respectively, they can in fact operate in all four-quadrants and its output can be either DC or AC.
Figure 4 shows the simplified equivalent circuit of the TS250 being used as a battery simulator presented in Figure 5. The TS250 has all of the capabilities of a battery simulator. It can source and sink current the same way a true battery does. Its DC-Offset control is for adjusting the output DC voltage, in which it emulates battery voltage changes.
A battery simulator can set to any voltage in a matter of seconds vs a real battery, which takes hours to reach a desirable voltage, streamlining bench testing. A simulator can replicate an overcharged battery, which can’t be done safely with a real battery. Similarly, a simulator can emulate a dead battery (0V) easily, also difficult to do with a real battery. Furthermore, if there is a problem in the charging system, a simulator can effortlessly sweep the battery voltage from high-to-low and low-to-high so that the issue can be quickly identified.
The purpose of a simulator is to validate on the bench the performance of chargers to make sure they are reliable and are able to safely recharge batteries. As discussed before, an actual battery will take a considerable period of time to discharge as well as to charge, up to 6 hours or more to complete a charging and discharging cycle. The charger’s performance and reliability must analyze throughout the charging cycle. An emulator can imitate a charged or discharged battery in just seconds. By using an emulator, you have the ability to mimic a battery at any voltage by simply adjusting the output voltage control. You can measure the charging current at low voltage (under 3 volts for a lithium cell), nominal voltage (3 volts to ~4.2 volts), and high voltage (over 4.2 volts) to analyze its full-charged output.
Lithium-ion battery charging methodology
A battery simulator equipment is typically employed to test the charger’s behavior over the entire voltage. A good example of this is the lithium-ion battery, which has a standard operating voltage of 3.0 to 4.2 volts, but can reach as low as 0V if it has been completely discharged. Lithium-ion battery most commonly uses a CC/CV (constant-current and constant-voltage) charging methodology. At low battery voltage (below 3.2 volts), it needs to be trickle-charged with a low current. The battery is can’t handle a higher charging current when the voltage is too low, so for safety reasons the trickle-charge current is typically one-tenth of normal. Figure 6 shows the charging voltage and current profile.
The next charging current stage is when the battery voltage has reached a safe level for normal current, usually called the rapid-charging stage. Typically the rapid charge starts when the battery voltage reaches about 3.2V. The charger will transition to higher charging current, usually about ten times higher than trickle charge. This is the normal charging current, and the charger regulates a constant current independent of the battery voltage. majority of the battery charging time is in the constant-current stage. The charger will maintain the charging rate until the battery reaches about 4.2V.
While the battery is being charged, its voltage continues to rise. When the voltage reaches about the “float” voltage, typically 4.2V, the charger enters a constant-voltage charging mode. The battery continues to charge, but the charging current is gradually decreased. The float voltage is the maximum battery voltage when fully charged.
While charging at a constant voltage, the battery charging current is gradually decreased as the battery fills. When the battery is full, the current reduces to a level typically about 1/20th of the normal charging current, or may stop charging completely. Figure 6 shows the lithium-ion battery charging techniques by plotting the voltage and current vs time, showing the battery voltage and current charging profile.
Charger stress tests
A well-designed charger must to be able to charge a battery correctly under normal conditions. More importantly it must able to safely handle a number of corner charging cases. Figure 5 uses the TS250 as an example of how a simulator can be connected to a charging system. The output of the TS250 battery emulator is attached to the battery connectors of the system or device. The simulator takes the place of the battery pack. To test the charger, change the DC OFFSET control to change the emulated voltage while recording of the charging current. Move the output voltage of the simulator from low-to-high and high-to-low to show how the charger reacts to the changes.
A charger must also be able to charge a completely drained battery. Set the battery simulator to 0V and test the charger to see if it enters the trickle charge state. The charging current should be small about one-tenth of the normal charging current. Then slowly increase the simulator voltage while monitor the charging current. It should stay in the trickle charge current level until the battery simulator voltage reaches a safe level (i.e. 3.2V), depend on the charger design. At this point the charger should start rapid charging with a normal current, typically about 1C. 1C is referring to the charging current being the same as the battery capacity. For example, for a 1000mAh (mili-amp hour) battery, 1C charging rate is equal to 1000mA.
To further stress test the charger, adjust the battery voltage back-and-fourth between about 2.9V and 3.3V to cause the charging to switch back-and-forth between trickle charge and rapid charge mode. Monitor both the charger input current and the charger output current (to the battery simulator) behavior. Look for any unusual behavior such as current spikes, both input and output, during the transition.
If the charging power is from a USB port, the charger must also meet all USB-related specification as well. The standard USB generally has two current levels available: 100mA and 500mA. If the system under test cannot successfully communicate with the USB host, it only allows a 100mA draw from the USB port. So the charging current should be limited to 100mA for battery voltages between 0V and ~3.2V while the device under test is off. Only after the system determines that 500mA is available, then charging can be as high as 500mA, but should never exceed that amount. Watch the USB port current draw to make sure it does not exceed 500mA, especially during mode transaction between trickle charge and rapid charge.
The bulk of charging is in the constant-current rapid charge mode. The battery voltage is between 3.2V and 4.2V during this stage. As the battery voltage is approaching the 4.2V full-voltage, the charging current is gradually reduced. In effect the charger is changing from CC mode to CV mode. Use the battery emulator to change the voltage back-and-forth between about 4.1V and 4.2V to force the charger to switch between CC and CV modes and back. Make sure the charger is behaving correctly by monitoring the input and output currents. If the charger is powered by a USB port, make sure the input current never exceed 500mA.
For overvoltage testing, intentionally set the simulated battery voltage to 4.3V. This is the over-voltage condition. Now turn on the charger by applying input power. Make sure the charging never charge at this voltage, not even for 1ms. When it comes to temperature faults, most lithium batteries have a temperature sensor built-in. You may stress test the charger by generating temperature fault signal to the charger. The charger must stop charging immediately for either hot or cold. You should repeat the temperature fault test for multiple battery voltages and charging modes. You may do that by using the simulator to set the voltage at different levels and modes: 0V – 2.9V trickle charge, 2.9V to 3.2V (transition), 3.2V to 4.1V (CC mode), 4.1V to 4.2V (transition and CV mode), 4.3V (over-voltage).
You can emulate an old and worn battery by simulating the battery internal resistance. You can do that by adding a resistor in series with the simulator. An old battery typically has an effective internal resistance that is 5-10 times higher than a new battery. So set the resistance accordingly. High charging current can heat up the resistor. When choosing a resistor, make sure it can handle the power dissipation. You may repeat the above discussed tests with an emulated old battery. Anther fault condition is the battery is not installed or disconnected (broken wire, momentarily disconnect, etc.). The charger should be able detect the battery is missing and safely handle the situation.
The first-case testing is when the battery is missing and power applied to the charger. Depends on the device design, the charger should not be turn on. Monitor the voltage at the charger out (normally connected to the battery terminal). The voltage at the battery connection node (but no battery) should not ramp up to high voltage. High voltage at the connector node may damage other circuitry connected to the same node. The voltage should not be oscillating back-and-forth between high and low voltages.
The second test case is for a battery initially installed, but later disconnected. The charger should gracefully power down. Make sure the battery connector node voltage does not overshoot excessively or oscillating. The third test case is for the battery re-connected after being discounted. The charger should start charging when battery is available again.
Thorough charger testing is an important task during product development to ensure safe operation, normal testing with an actual battery is time consuming. A battery simulator help reduces the test time tremendously. For more information visit www.accelinstruments.com/Applications/TS200/Battery-Simulator-AppNote.html