Solar Charge Controller Circuit Design

A solar panel produces a voltage that depends on the light and a current that depends on the load, and the product of the two is the power delivered. A charge controller sits between the panel and the battery and tries to keep that product as large as possible while charging the battery within its limits. Doing both well is what separates a controller that works from one that merely connects.

What a Solar Charge Controller Does

The controller takes the panel output, converts it to a voltage suited to the battery and manages the charging process. The simplest version connects the panel to the battery through a switch and stops when the battery is full, which works but wastes most of the panel capability during the bulk of the charge.

A better controller operates the panel at its maximum power point and converts the result to a lower or higher voltage as the battery requires. That conversion allows a panel with a high voltage and a low current to charge a battery at a lower voltage and a higher current, which reduces the losses in the wiring between them.

The controller also protects the battery. Overcharging a lead acid battery drives off water and shortens its life, while over discharging damages the plates. Both limits are implemented in the controller, and the accuracy of those limits determines how much of the battery capacity can actually be used.

Panel Characteristics and Maximum Power Point

A silicon panel behaves approximately as a current source in parallel with a diode. The current is set by the light, and the voltage is limited by the diode characteristic, so the output curve has a flat region where the current is nearly constant and a knee beyond which the current collapses.

maximum power point is the point on that curve where the product of voltage and current is largest, and it sits near the knee. It moves with the light level and with the temperature of the cells: a cold panel has a higher maximum power voltage, and a hot panel a lower one. A fixed operating voltage therefore only matches the panel under one set of conditions.

Tracking the maximum power point is done by perturbing the operating voltage and observing the change in power. If a small increase in voltage increases the power, the controller continues in that direction; if it decreases, the direction reverses. The perturbation has to be slow enough that the converter settles between steps, and fast enough to follow a passing cloud.

Converting the Panel Voltage

A buck converter is used when the panel voltage is above the battery voltage, which is the usual arrangement for a nominal twelve volt panel charging a twelve volt battery at a higher operating voltage. A boost converter is used when the panel voltage is below the battery, as with a single cell panel and a higher voltage battery.

The converter has to work over a wide input range, because the panel voltage moves with light and temperature, and over a wide output range, because the battery voltage moves with state of charge. The control loop has to remain stable across that whole space, and the components have to be selected for the worst case of each, not for the nominal condition.

Conversion efficiency matters because every watt lost is a watt that does not reach the battery. The switching device, the inductor and the diode are the three main contributors, and the choice between a schottky diode and a synchronous switch is usually decided by the efficiency at the typical operating current rather than at the maximum.

Battery Charging Profiles

battery charging follows a profile that depends on the chemistry. A lead acid battery is charged at a constant current until the voltage reaches a defined level, held at that voltage until the current falls to a small fraction of capacity, and then maintained at a lower float voltage. Each stage has a voltage limit that has to be accurate, because it sets the trade between charge completeness and battery life.

Lithium chemistries require a different and more precise approach. The charge is a constant current stage followed by a constant voltage stage, with a tight voltage tolerance and a defined termination current. Exceeding the voltage limit damages the cell, which is why the tolerance is usually specified as one percent or better.

charge profile temperature compensation is important for lead acid batteries, whose voltage limits fall as the temperature rises. Without compensation, a battery charged in a warm enclosure is undercharged and one charged in the cold is overcharged. A sensor on the battery, rather than on the board, is the only way to apply the correction correctly.

Sense Paths and Measurement Accuracy

The controller makes its decisions from measurements of the panel voltage, the battery voltage and the current in each path. Every one of those measurements has an error, and the errors combine to define how well the charge profile is followed. A divider with ordinary resistors and a reference with a poor temperature coefficient can shift the charge voltage by more than the tolerance the battery allows.

Measuring current requires a sense element in series with the path, and its resistance has to be low enough not to waste power while still producing a signal large enough to measure. A four terminal connection keeps the resistance of the solder joints out of the measurement, and the layout around the sense pair has to be symmetric for the same reason.

panel voltage measurement is affected by the switching converter, because the panel current is pulsed. The measurement has to be synchronised with the switching cycle, or taken with a filter that averages over a whole cycle, otherwise the reading depends on the duty cycle rather than on the panel condition.

Load Control and Low Voltage Disconnect

Many controllers also manage a load output, disconnecting it when the battery falls below a defined voltage and reconnecting it when the battery has recovered. The thresholds need hysteresis, or the load will cycle on and off repeatedly as the battery recovers under no load and sags again under load.

The disconnect threshold has to be low enough to use most of the battery capacity and high enough to avoid damage. It also has to account for the voltage drop in the wiring, which is significant at the currents a load draws, and for the temperature of the battery. Measuring at the battery terminals rather than at the controller terminals removes one source of error.

A load output that switches a large current requires its own protection. A shorted load draws current limited only by the battery, and a fuse or an electronic current limit is needed to protect the wiring. The connector and the traces have to be rated for the current, and the return path should be separate from the measurement ground.

Layout and Protection

The power stage should be laid out with a small switching loop and a short return path, following the same rules as any converter. The panel and battery connectors carry the full current, so they should be placed at opposite ends of the power stage with the switching components between them, which keeps the high current paths short and the sensitive measurement paths away from them.

Thermal design is dominated by the switching device and by the sense elements. A converter running at ninety percent efficiency with a hundred watts of input dissipates ten watts, and that heat has somewhere to go. Our guide to PCB thermal management covers the copper and via arrangements that make that possible.

Protection should cover reverse panel connection, reverse battery connection and lightning induced transients on the panel wiring, which is often the longest cable in the installation. A series element with a clamp at the connector is the usual arrangement, and the devices have to be chosen so that they do not conduct at the maximum operating voltage. Component tolerances across the temperature range are covered in component tolerance and reliability.

Verification and Common Faults

Verify the tracking by measuring the panel voltage and current while the light changes, and confirm that the operating point follows the maximum power point rather than sitting at a fixed voltage. A controller that holds a constant panel voltage is either not tracking or has a measurement that is dominated by the switching ripple.

Verify the charge profile by recording the battery voltage and current through a full charge. The transitions should happen at the specified voltages and the current should fall to the termination value before the controller switches to float. A charge that terminates early indicates a voltage measurement error, while one that continues too long indicates a current measurement error.

A controller that works well in the morning and poorly in the afternoon is usually suffering from temperature drift rather than from a logic error. Checking the measured panel and battery voltages against a reference meter at both temperature extremes separates a measurement problem from a control problem, and the reference path issues described in power integrity are worth reviewing at the same time.

FAQ

Do I need maximum power point tracking on a small system? It pays for itself whenever the panel voltage is significantly above the battery voltage, because a direct connection wastes most of the available power in the bulk charge stage.

How accurate must the charge voltage be? Within one percent for lithium chemistries and within a few percent for lead acid, including the temperature coefficient of the reference and the divider tolerance.

Why does my battery never reach full charge? Check the charge voltage measurement first, then the termination current threshold. A divider error or a warm reference is the most common cause.