Solar-Powered High Efficiency Charger

This is a simple NiCd battery charger powered by solar cells. A solar cell panel or an array of solar cells can charge a battery at more than 80 % efficiency provided the available voltage exceeds the ‘fully charged’ battery voltage by the drop across one diode, which is simply inserted between the solar cell array and the battery. Adding a step-down regulator enables a solar cell array to charge battery packs with various terminal voltages at optimum rates and with efficiencies approaching those of the regulator itself. However, the IC must then operate in an unorthodox fashion (a.k.a. ‘Elektor mode’) regulating the flow of charge current in such a way that the solar array output voltage remains near the level required for peak power transfer. Here, the MAX639 regulates its input voltage instead of its output voltage as is more customary (but less interesting).

Circuit diagram:

Solar-Powered High Efficiency Charger Circuit Diagram

The input voltage is supplied by twelve amorphous solar cells with a minimum surface area of 100 cm2. Returning to the circuit, potential divider R2/R3 disables the internal regulating loop by holding the V-FB (voltage feedback) terminal low, while divider R1/R2+R3 enables LBI (low battery input) to sense a decrease in the solar array output voltage. The resulting deviation from the solar cells’ peak output power causes LBO (low battery output) to pull SHDN (shutdown) low and consequently disable the chip. LBI then senses a rising input voltage, LBO goes high and the pulsating control maintains maximum power transfer to the NiCd cells.

Current limiting inside the MAX639 creates a ‘ceiling’ of 200 mA for I out. Up to five NiCd cells may be connected in series to the charger output. When ‘on’ the regulator chip passes current from pin 6 to pin 5 through an internal switch representing a resistance of less than 1 ohm. Benefiting from the regulator’s low quiescent current (10 microamps typical) and high efficiency (85 %), the circuit can deliver four times more power than the single-diode configuration usually found in simple solar chargers. Coil L1 is a 100-µH suppressor choke rated for 600 mA.

Author: D. Prabakaran - Copyright: Elektor July-August 2004

Storage Battery Exerciser

A motorcycle or boat battery that is not needed over the winter is usually charged before being put away for the winter, after which it remains standing unused for months on end. As a result, it accumulates deposits of lead sludge, which can result in reduced capacity or even complete failure of the battery. If you don’t keep active, you rust! To avoid this, it’s necessary to keep the battery active even during the winter. This circuit does such a good job of exercising the battery that it doesn’t have to be recharged during the winter. It only has to be fully charged again in the spring before being used again. IC1.A is an astable multivibrator with an asymmetric duty cycle. The output is High for around 0.6 s and Low for around 40 s. IC1.B is wired as a comparator that constantly monitors the battery voltage. Its threshold voltage is set to 11.0 V using the trimpot.

Circuit diagram:

Storage Battery Exerciser Circuit Diagram

As soon as the battery voltage drops below this value, the comparator goes Low and D6 is cut off, allowing the second astable multivibrator IC1.C to oscillate at approximately 1.2 Hz. LED D7 then blinks to indicate that the battery must be charged. As long as the battery voltage is greater than 11 V, IC1.B is High. IC1.A is Low most of the time, and in this state D4 conducts and the inverting input of IC1.D is Low. This means that IC1.D is High most of the time, with T1 cut off. T1 only conducts during the 0.6-s intervals when IC1.A is High. In this state it allows current to pass through the lamp (12 V / 3 W), which forms the actual load for the battery. After this, darkness prevails again for 40 s. The average current consumption is approximately 5 mA. At this rate, a relatively new 40-Ah battery will take around one year to become fully discharged. However, this can vary depending on the condition of the battery, and it may be necessary to ‘top up’ the battery once during the winter.

Author: Ludwig Libertin - Copyright: Elektor Electronics

Solar Battery Protector Prevents excessive Discharge

This circuit prevents the battery in a solar lighting system from being excessively discharged. It's for small systems with less than 100W of lighting, such as several fluorescent lights, although with a higher rated Mosfet at the output, it could switch larger loads. The circuit has two comparators based on an LM393 dual op amp. One monitors the ambient light so that lamps cannot be turned on during the day. The second monitors the battery voltage, to prevent it from being excessively discharged. IC1b monitors the ambient light by virtue of the light dependent resistor connected to its non-inverting input. When exposed to light, the resistance of the LDR is low and so the output at pin 7 is low.

Circuit diagram:

Solar Battery Protector Circuit Diagram

IC1a monitors the battery voltage via a voltage divider connected to its non-inverting input. Its inverting input is connected to a reference voltage provided by ZD1. Trimpot VR1 is set so that when the battery is charged, the output at pin 1 is high and so Mosfet Q1 turns on to operate the lights. The two comparator outputs are connected together in OR gate fashion, which is permissible because they are open-collector outputs. Therefore, if either comparator output is low (ie, the internal output transistor is on) then the Mosfet (Q1) is prevented from turning on. In practice, VR1 would be set to turn off the Mosfet if the battery voltage falls below 12V. The suggested LDR is a NORP12, a weather resistant type available from Farnell Electronic Components Pty Ltd.

Author: Michael Moore - Copyright: Silicon Chip Electronics

10,000x With One Transistor

For a collector follower with emitter resistor, you’ll often find that the gain per stage is no more than 10 to 50 times. The gain increases when the emitter resistor is omitted. Unfortunately, the distortion also increases. With a ubiquitous transistor such as the BC547B, the gain of the transistor is roughly equal to 40 times the collector current (Ic), provided the collector current is less than a few milliamps. This value is in theory equal to the expression q/KT, where q is the charge of the electron, K is Boltzmann’s constant and T is the temperature in Kelvin.

For simplicity, and assuming room temperature, we round this value to 40. For a single stage amplifier circuit with grounded emitter it holds that the gain Uout /Uin (for AC voltage) is in theory equal to SRc. As we observed before, the slope S is about 40Ic. From this follows that the gain is approximately equal to 40I cRc. What does this mean? In the first instance this leads to a very practical rule of thumb: that gain of a grounded emitter circuit amounts to 40·I c·Rc, which is equal to 40 times the voltage across the collector resistor.

If Ub is, for example, equal to 12 V and the collector is set to 5V, then we know, irrespective of the values of the resistors that the gain will be about 40R(12–5) = 280. Notable is the fact that in this way the gain can be very high in theory, by selecting a high power supply voltage. Such a voltage could be obtained from an isolating transformer from the mains. An isolating transformer can be made by connecting the secondaries of two transformers together, which results in a galvanically isolated mains voltage.

Circuit diagram:

That means, that with a mains voltage of 240 Veff there will be about 340 V DC after rectification and filtering. If in the amplifier circuit the power supply voltage is now 340 V and the collector voltage is 2 V, then the gain is in theory equal to 40 x (340–2). This is more than 13,500 times! However, there are a few drawbacks in practice. This is related to the output characteristic of the transistor. In practice, it turns out that the transistor does actually have an output resistor between collector and emitter.

This output resistance exists as a transistor parameter and is called ‘hoe’. In normal designs this parameter is of no consequence because it has no noticeable effect if the collector resistor is not large. When powering the amplifier from 340 V and setting the collector current to 1 mA, the collector resistor will have a value of 338 k. Whether the ‘hoe’-parameter has any influence depends in the type of transistor. We also note that with such high gains, the base-collector capacitance in particular will start to play a role.

As a consequence the input frequency may not be too high. For a higher bandwidth we will have to use a transistor with small Cbc, such as a BF494 or perhaps even an SHF transistor such as a BFR91A. We will have to adjust the value of the base resistor to the new hfe. The author has carried out measurements with a BC547B at a power supply voltage of 30 V. A value of 2 V was chosen for the collector voltage. Measurements confirm the rule of thumb. The gain was more than 1,000 times and the effects of ‘hoe’ and the base-collector capacitance were not noticeable because of the now much smaller collector resistor.

Author: Gert Baars
Copyright: Elektor Electronics