Incandescent lamp saver


WARNING!

This projects uses mains voltage which may be lethal! By following any designs or instructions you are doing so at your own risk!

Quite understandably it can get quite frustrating when your filament lamps fail when you switch them on. Nowadays filament lamps aren't in common use anymore but still, there are applications for filament lamps, like heavy machinery where mains LEDs and fluorescents cannot be used around rotating equipment due to the mains flicker.

Filament lamps generally will fail on startup. They exhibit two resistances. When cold there is a low resistance across the filament when power is applied there will be a lot of current flowing through the lamp this has the benefit of the lamp warming up quickly but it is what wears the lamp away quicker. When the lamp is hot the resistance is much higher because resistance increases when the temperature does. What this device does is gradually increase the power to the lamp to avoid the sudden inrush of current.

This circuit came up in the Electronics Australia magazine in the June 1986 issue. Understandably this article is very old mainly since incandescent lamps were so cheap that there wasn't much of a point. In the article, it shows a PCB layout that would've been etched at home. The PCB is to be glued onto the back of a switch plate. For people outside of Australia and New Zealand, the PCB layout is not suitable so you will have to make your own layout but it isn't a complicated circuit.

Electronics Austrlia June 1986 Article

Parts List
QTY Value Description
1 HPM 777WE Single-gang switch plate
1 2-Way terminal strip
1 Neosoid 17-132-10 Powdered iron ring-core
2M 0.5mm Enammeld magnet wire
1 SC141D / BT137 Triac
4 1N4007 1kV 1A Diode
1 2N4992 Silicon Bilateral Switch
1 BC547 NPN BJT Transistor
1 1N4746 18V 1W Zener diode
1 1uF 35V Electrolytic capacitor
1 220nF Greencap
1 100nF Greencap
1 100nF 300V.A.C. X2 Rating
1 1MΩ ½W
1 300kΩ ½W
2 100kΩ ½W
1 10kΩ ½W
1 10kΩ 1W

A problem arose when I came to gather all the parts to build this device. The issue is with making the coil, in 2020, it seems that the core that it specifies no longer exists on the Neosid website. The way I fixed that is that I took the product code and roughly figured out the modern code. On the older on it is 17-132-10, it looks fitting that is is the O.D, I.D, and height. What I did is that I took my deciphered product code and found a new core with similar dimensions and I calculated the turns for 100uH

New coil calculations
100uH
Powdered Iron
Core = T80-26
Grade - 26 - Yellow white
47 Turns

Schematic

PCB Layout
Scale to 41*69mm

Back
2020 OZFix