Performing battery run-down tests

Of all of the specifications of a battery-powered device, the battery run time is key. Every user wants their device to run longer on a charge. Therefore, to know the run time of your device is important, not only to represent the device to the end user, but as a means of knowing if improvements in power management within the device are successful at increasing battery run time.

Battery run time is determined by a run-down test. In this, you run your device, starting with a full battery and measure the time it takes until it stops working. The time that is measured is the battery run time. While this is easily said, it is not so easily done, as there are many variables in this test.

Some engineers have considered using a power supply to simulate the battery during a run-down test. To date, this has not been practical. While it may be a hassle to ensure battery state consistency (type and charge), using a power supply to simulate a battery introduces additional variables and test errors.

A standard power supply will not act like the battery as it will never run down, so the run-down test will never reach an end condition. Instead, the power supply will need specialised features to be a battery emulator.

Part of the emulation is to have a controllable output resistance and to have good transient response with respect to current pulses being drawn by the device. But to fully emulate a battery, the power supply’s output voltage will need drop off as charge is pulled from the power supply into the device during the run-down test.

This simulation of draining the battery from full to discharged is challenging and sophisticated battery models must be employed. If the battery model is not adequate, the results obtained when using the power supply will not match the results obtained when using the battery.

Until a good battery modelling solution becomes available, the best way to do a run-down test is to use the real battery, as this will give the exact same results that the end user will experience.

When designing a device, you will want additional insight into what is happening during the test, beyond just measuring the run time. This will additionally mean measuring the current being drawn from the battery and the voltage on the battery simultaneously.

By plotting voltage and current versus time, a complete picture of a battery run-down is achieved (Figure 1).

Figure 1: Battery run-down test results.

To simultaneously measure battery voltage and current flowing between the battery and the device, you will need two DMMs, a 2-channel data logger or a 2-channel digitiser.

To measure the battery voltage is almost trivial, as the voltage will change slowly, so a DMM or data logger making measurements as slow as once a second should be fast enough to capture the slowly decaying voltage waveform.

But measuring the current can be a much bigger challenge. In many battery-powered devices, sophisticated power management schemes are used to increase run time.

These schemes turn subsystems on and off for hundreds of microseconds as needed within the device to conserve power. The result is a rapidly changing current waveform that can range from microamperes to amperes.

DMM’s make integrated measurements that can take hundreds of milliseconds so they cannot capture a rapidly changing current waveform.

Another issue with using a DMM is burden voltage. When a DMM is configured as an ammeter, the current to be measured is flowed through a calibrated current shunt inside the DMM. The DMM measures the voltage drop across the shunt and calculates the current.

The voltage drop inside the DMM reduces the available voltage at the DUT and hence places a burden on the circuit. This burden voltage can be hundreds of millivolts (Figure 2).

Figure 2: DMM presents burden voltage when measuring current.

When trying to measure a rapidly changing waveform over a long period of time, a digitiser seems like the best choice. They have sufficiently wide bandwidth to capture rapidly changing waveforms, but they don’t directly measure current so a current shunt must be used.

If the dynamic range of current to be measured is microamperes to amperes, what size shunt should be selected? If the shunt is sized to measure the lowest current accurately, there will be a large voltage drop across the shunt during the high current events and this will place an unbearable burden voltage on the circuit.

If the shunt is sized for the high current, at low currents there may not be enough voltage for the digitiser to measure accurately.

Therefore, you may need to compromise on low-level current measurement accuracy by selecting a shunt that is suitable for high currents (with acceptable burden voltage drop) and low currents (with a voltage drop that may be at the low limits of the digitiser’s ability to measure).

In June 2010, Agilent introduced the N6781A battery drain analyser and turnkey software for performing run-down tests. This can be configured as a zero-burden ammeter, meaning there is zero voltage drop across the instrument as it measures current flow between the battery and the device.

It offers a feature called seamless ranging so that it can instantly and automatically change range and measure currents from microamperes to amperes at a speed of 100,000 samples per second without losing any data during the range change, making it suitable for measuring dynamic currents during run-down tests.

Furthermore, it can simultaneously measure the voltage across the battery. With the control and analysis software, a battery run-down test can be quickly set up and run-down measurements captured and plotted without writing any software.