3A Solar Charge Control Circuit Schematic

3A 6V/12V Solar Charge Control

This solar charge control combines multiple features into a single design: 3A current rating, low dropout voltage (LDO), range of voltage adjustment (accommodates 6 & 12V lead-acid batteries), reverse polarity protection, low parts cost ($5.90) and low parts count (14 components). High performance is attributed to the application of the common LM358 op amp and TL431 adjustable shunt voltage regulator.


  • Max solar panel rating (12V): 43W (open circuit solar panel voltage = 18 to 20V)
  • Max solar panel rating (6V): 22W (open circuit solar panel voltage = 9 to 10V)
  • Maximum input voltage: 36V
  • Output voltage range: 4.5 to 15V (continuously adjustable)
  • Max power dissipation: 17W (includes power dissipation of D3)
  • Typical dropout voltage: 0.7V @ 3A (less @ lower currents)
  • Maximum current: 3A (current limiting provided by solar panel characteristics)
  • Voltage regulation: 5mV (voltage change no load to full load)
  • Battery discharge: 100µA (most commercially units discharge at typically 5mA)
  • Reverse battery protection: Control shuts down if battery is inadvertently connected reverse

Solar Charge Control Circuit Schematic

3A Solar Charge Control Circuit Schematic

Other solar charge controls that I have posted at

Bill of Materials

3A Solar Charge Control BOM

Link to Excel file

Operation at reduced current/power

While designed for 3A maximum, it will function just as well at low currents. If applied below 200mA, the heatsink may be eliminated. If applied below 1A, D3 may be reduced in size.

Nominal vs. actual charging voltage

When the voltage is stated (e.g. 12V), this is a reference to the nominal battery voltage (voltage in name only). Actual battery voltage ranges from 10.5V (fully discharged) to (14V fully charged).

Float charge voltage vs. full charge voltage in lead-acid batteries

The float charge voltage is the charging voltage that may remain connected long term—this is approximately 7V for 6V batteries and 14V for 12V batteries—actual manufacturer’s recommendation may vary somewhat, so it may be helpful to check the actual specifications. For faster charging, the voltage may be set slightly higher (e.g. 7.4V for 6V batteries or 14.5V for 12V batteries)—this charges the battery more rapidly, but requires that the control be intelligent enough to reduce the voltage to the float charge level after charging is complete. Most charge controls (like this one) simply charge at the float charge voltage—all automotive electrical systems do this.

Mismatched solar panel application—charging 6V batteries from 18V solar panels

Normally, solar panels are designed for a specific battery voltage applications. For 12V applications, the solar panel open circuit voltage is generally 18 to 20V. Similarly, for 6V applications, the solar panel voltage open circuit voltage is generally 9 to 10V. Since the 9 to 10V panels are relatively uncommon, it is not unusual to use 18 to 20V panels for charging 6V batteries. However, in this case the power dissipation of the series regulator transistor is multiplied by a factor of approximately 5. To prevent thermal destruction of Q1, the current rating in such cases must be derated from 3A to 1A.

A potential work-around is the addition of a 3Ω, 25W resistor in series with the solar panel—this reduces the control input voltage thus maintaining the 3A current rating.

The characteristics of a linear series regulator

The dissipated power is simply the voltage drop times the current. When there is current, but little voltage, the power dissipation is low—when there is voltage, but little current, the power dissipation is also low—when both current and voltage are present simultaneously, there is substantial power. Such is the nature of a linear regulator. The power is maximum when the battery starts to “top off” at the set voltage.

Dropout Voltage

The input voltage exceeds the input voltage by 0.7V when charging at the maximum rate—the lower, the better. Low Dropout Voltage (LDO) is the catch phrase for anything under approximately 2V. This is an important detail for 6V systems—for 12V systems, it is not generally a big issue.

Current Limiting

Current limiting is provided by the solar panel—it is not a commonly understood fact that the solar panel tends to be a constant current device. For this reason, a solar panel can withstand a short circuit. Therefore, the control itself does not require a current limiting feature.

Voltage Reference

The Texas Instrument TL431 is an inexpensive programmable shunt regulator. In this circuit, it is applied at the minimum voltage (2.5V) and does not require the usual voltage divider resistors. The TL432 is very similar, but the pin-out is different. Both are available in a variety of package outlines, but only the TL431 is available in the popular TO-92 package.

Circuit Operation

R1 biases D1, the voltage reference diode. The 2.5V reference from D1 is compared with voltage feedback from the resistor divider. The op amp does all within its power to keep these two voltages identical. The ratio of R3 /R2 controls the proportional gain, and C1 is a compensation capacitor that blocks DC feedback, but responds to changes in output signal thus maintaining stability (prevents oscillation). Zener D2 prevents overvoltage at the gate of Q1—R4 limits op amp output current when D2 is conducting. C1 is the positive rail bypass capacitor. D3 prevents battery voltage from appearing across the solar panel and prevents unnecessary battery discharge when the solar cells are not generating power.

When the feedback voltage from the wiper of R6 drops below 2.5V, the output of U1A moves in the negative direction thus turning Q1 on. The increased current out of Q1 causes the battery voltage to increase and increases the voltage at the wiper of R6 until it is equal to the reference voltage.

It may seem like a waste to use a dual op amp when only a single is required, but the LM358 remains the least expensive and most available device. It also has an undocumented feature that provides reverse battery connection. When the battery voltage is reverse, the non-inverting input of U1 is driven below the negative rail (common)—when this happens, the output of the op amp swings to the positive rail thus turning off Q1 and protecting the circuit against this potentially damaging condition. While this ‘malfunction’ is perhaps well known in the engineering community, the application of this as a circuit trick is new to the world.


For the future

Reduced intelligence MPPT charge control dispenses with the microcontroller

Undocumented words and idioms (for our ESL friends)

work-around –idiom –noun, the solution of a problem that avoids redesigning key components—sometimes cumbersome


Join the conversation!

Error! Please fill all fields.
  • tschaerni

    I like the design and I build one for my self because I want to charge a 3S LiIo Pack with it (12.6V charge cut off voltage, and I’m also building a balancer for it) and I’m using a 40W Panel with a voltage around 13.6V. Because of that a linear charge controller would be the most efficient thing to use.

    I was thinking about efficiency, and how I can replace diode D3. I think it would work to use a second P-Channel Mosfet instead D3 (name it Q2). The Gate of Q2 would be on the same pin as Gate from Q1, drain of Q2 on drain of Q1 and source of Q2 would be the output. Like in this picture –>

    I think I will test this in the next days if I have the Time. But if it works (IMHO it should work) that would be a _really_ LDO Charge controller.

    – tschaerni

  • John

    Hello Jim,
    I have built the circuit and it does a great job of holding the voltage to 14V when the input voltage is between 19-23V however when the voltage drops below that threshold the output becomes too low to charge the battery. Do you have any suggestions for fixing this issue?

    Thank you

  • Jim Keith

    Recheck your circuit and make no assumptions–check out this article:

  • gauraveee710gmail-com

    i did above circuit, i am getting output voltage around 12-14 v (i am using 1M pot), but for every value of pot i am getting output current of controller zero.

  • emmaotsapa

    Yes it does.Thanks for the detailed answer,i appreciate.Only wish the answer came much sooner.

  • Jim Keith

    This design is largely empirical, based upon my experience. I simulated it my cranial simulator that is surprisingly good sometimes and surprisingly poor at other times.

    D1, voltage reference: Tying the adjustment pin back to the cathode, sets the voltage at 2.5V or minimum. It is biased close to minimum recommended current to conserve power (1mA).

    R1 sets the D1 bias current. This is set at the minimum input voltage for a 6V battery @ full charge (approx 7.4V + 0.6V diode drop). R1 = (8V – 2.5V) / 1mA. 4.7K is close enough–generally non-critical.

    C1 is a high frequency bypass capacitor to keep the op amp stable. The value is generally non-critical so an empirical value of 0.1uf was chosen.

    Zener D2 limits the gate to source voltage of Q1. With no load connected and high solar panel voltage, the Vgs otherwise well exceeds the maximum Vgs rating because the output of the output of the op amp drops to zero. This prevents MOSFET gate punch-through voltage failure. 12V was chosen to exceed the normally specified full ON Vgs of -10V.

    R4 limits the current through D2 during the above fault condition. It is empirically chosen to be high enough to limit the current to below about 1mA, but low enough so that the RC phase shift caused by R4 and the input capacitance of Q1 is much lower than the op amp response –driving an integrator with an integrator causes instability (oscillation).

    Q1 is selected for low Rdson (0.07Ω) and low cost (about $1). It is overkill, but works well.

    D2 is just a cheap 3A silicon diode that keeps the battery voltage from appearing across the solar panel when the sun is not shining –such can discharge the battery because the reverse leakage of the solar panel is not specified.

    Adjustable feedback voltage divider (R5, R6 & R7), sets the full charge output voltage. The values are calculated by Ohms law –the voltage at the wiper of R6 is 2.5V if operated within the limits of normal operation. (Abnormal operation would be battery open or short circuit etc.)

    U1A is an inexpensive op amp. The LM358 is perhaps the cheapest available and it just happens to be a dual, so the unused section is neglected. The error amplifier circuit has an open circuit voltage gain of perhaps 100,000, so the difference between the 2.5V reference and the wiper of R6 is extremely low. By putting C2 in the feedback loop, the amp runs at maximum gain –the op amp does everything it can to keep the pot wiper voltage equal to the reference voltage. The circuit has a proportional voltage gain of 3 that is set by (R3 + R2)/R2. If the proportional gain is too high, the circuit will oscillate. It is possible that R3 may be shorted –in this case the proportional gain would be unity. The values are empirically chosen as a place to start –since it worked OK, no resistance changes were required. The same goes for the value of integrator capacitor C2 that was empirically chosen at 0.1uf.

    Hope this answers your question.

  • emmaotsapa

    Thanks i have tested the circuit and it works fine,i am doing it as a school project.i would like to understand how the circuit was analysed.i.e each component value was determined and calculated.Pls a quick response will be appreciated.

  • shafqat

    sir plese tell about any two or equivlent for each f these P80PF55 (80A, 55V MOSFET P Channel) and 80SQ045 schotky doide. it is not avail here in pakistan
    i want to use this circuit for a 150 watt solar panel, please sir i need it urgently

  • emmaotsapa

    Please i would like to understand how the circuit was analyse.i.e how each component value was determined and calculated.

  • shafqat

    sir plese tell about any two or equivlent for ech f these P80PF55 (80A, 55V MOSFET P Channel) and 80SQ045 schotky doide. it is not avail here in pakistan

Looking for the latest from TI?