A better arrangement is shown in fig 5 where the on/off switch is replaced by a thyristor.
The circuit is switched on by pressing PB1 which triggers TH1 causing it to conduct. Even when PB1 is released, the current passing through TH1 keeps it latched on. While the LED is on, it draws charge from C1. As C1 discharges, its terminal voltage decreases which in turn decreases the LED current. As time passes and the current declines, the LED brightness fades. Eventually, when the current in the LED and therefore thyristor (as they are in series) drops below the thyristor holding current, the thyristor switches off. The circuit has reset ready for charging the next day. Using a white LED (rapid part no: 55-2482) and a 360R limiting resistor (R2), a full charge gives 15 minutes of good illumination with the LED remaining on for a further 15 minutes.
One of the nice features of the nightlight I originally published is that it was automatic; it switches on when the room lights are a switched off. Although the eco-nightlight circuit has a large light sensor, the solar cell, the limited available charge (in C1) makes it unsuitable for an automatic nightlight. This is because a voltage monitoring circuit would be needed to detect when room lights have been turned off.
The eco-nightlight has great potential as a school project because it is relatively inexpensive on components, can help address issues of sustainability and there is plenty of scope to explore circuit variations.
I am very grateful to:
Gareth Evans for reminding me of the existence of the Joule Thief circuit and for demonstrating its effectiveness when powered by ‘flat’ batteries.
My colleague, Paul Spence, for enquiring if the Joule Thief could operate from alternative power sources.
Download complete article Eco-nightlight (5 downloads)