This Low Dropout Voltage (LDO) solar charge controller is a variation of the previously posted 12V LDO controller. It is optimized for charging a 6V lead-acid battery with a 9V solar panel. Minimum voltage drop is less than 1V. It uses a simple differential amplifier and series P channel MOSFET linear regulator. Voltage output is adjustable. It may also be applied in two or four cell lead-acid applications (4V & 8V).
It is not recommended for 12V applications.
6V Solar Charge Controller Specifications
Operation at lower current/power
While designed for 8A, 50W, it will function just as well at much lower current /power.
LDO Solar Charge Control Photos
Perf board—sorry, no circuit board artwork at the time of publication.
LM317LZ—many readers may not know that this component exists in the TO-92 package.
Bill of Materials
The input voltage exceeds the input voltage by 0.9V when charging at the maximum rate—the lower, the better. Low Dropout Voltage (LDO) is the catch phrase for anything under approximately 2V.
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 does not need current limiting.
Float Charge of Lead-Acid Batteries
This control charges the battery at a constant voltage and also maintains a charged battery (float charge). The float charge voltage specification is a little lower than the charge voltage, so to accommodate both voltages, a compromise is reached by simply reducing the voltage slightly—that is how ALL automotive systems operate. To obtain maximum charge in a 6V battery, set the control to 7 tp 7.4V.
To set the voltage, disconnect the battery and connect a 470Ω dummy load resistor across the output. The resistor is necessary to shunt potential MOSFET leakage current as well as the green LED current. The battery must be disconnected because the output voltage of the control cannot otherwise be set below actual battery voltage.
U1 is an LM317LZ TO-92 voltage regulator that is set to put out 3.1V. Low voltage zeners (below 6.2V) are too sloppy to use as voltage references, so the LM317 is used. Q1 & Q2 make up the classic differential amplifier that amplifies the difference between the reference voltage and the feedback voltage from the arm of potentiometer R6. The output is taken from the collector of Q2 and drives the gate of P Channel MOSFET Q3. Differential voltage gain is probably in the order of 100 to 200. For best performance, I selected Q1 & Q2 for matched hFE (approx 300). As the feedback voltage increases at the arm of R6, Q2 turns on harder and steals some of the emitter current away from Q1. The collector current of Q1 follows the emitter current and drops less voltage across R1 thus reducing Vgs of Q3 and turning it off. C2 provides frequency compensation to prevent the amplifier from oscillating.
Q3 (Fairchild NDP6020P) is a high current P-Channel MOSFET that has an Rds on of only 50mΩ. This is in the class of logic level controlled devices as it may be turned on fully with only a 4.5V gate to source voltage. To obtain these properties, a sacrifice is made in voltage rating. As a result, Q3 is rated for only 20V. Because Max Vgs is only 8V, a 6.2V zener (D1) protects the gate from potentially destructive voltage. Due to these voltage limitations, this control is not recommended for 12V applications.
When the battery reaches set voltage, Q3 starts to drop significant voltage and turns on Q5 which powers the Green LED.
Q4 is dormant unless the battery is connected reverse—should this happen, Q4 turns on and reduces the reference voltage input to zero thus turning Q1 & Q3 and preventing damaging battery current.
Blocking diode, D3, prevents the battery voltage from appearing across an inactive solar panel.
If this circuit appears too complex, strip off the unnecessary components and things get much easier—drop (or add later) R8-11, D2,4,5, Q4,5 thus saving 9 components.
This is a linear series regulator that dissipates significant power when the pass transistor is both conducting current and dropping voltage simultaneously—during maximum charge rate when the voltage drop is low, the heatsink runs warm—when the battery is fully charged and there is low charge current, the heatsink is cold—but when the battery starts to top off at maximum voltage, the heatsink runs very hot—such is the nature of a linear regulator. At 8A, Q3 drops 1.45V (assuming solar panel voltage is 9V). The remaining 0.55V is the D3 voltage drop. P = 8A * 1.45V = 11.6W. The heatsink is rated at 3.9°C/W, so heatsink temperature rise = 11.6W * 3.9°C/W = 45.2°C. Adding the 25°C ambient temperature results in a heatsink temperature of 70.2°C. While this may seem very HOT to the touch, it is still cool to the transistor that is rated for a junction temperature of 175°C.
D3 dissipates 4.4W @ 8A. This requires a heatsink. In an etched circuit, a large foil area is required to dissipate the heat. In the perfboard version, I added copper heatsink bars fabricated out of AWG #14 solid copper leads that were hammered flat to increase the surface area.
Testing the 6V LDO Solar Charge Control
My apparatus cannot simulate solar panel current above 6.6A. While the control is designed for 8A, it has not been actually tested at that level. Actual measurements indicate a voltage drop of 0.51V @ 4A and 0.64V @ 6.6A. Voltage regulation measures 80mV (NL to 6.6A). I do not know why the 12V LDO control performed much better in this regard.
For the future
Properties of the differential amplifier