It's taken me a while to get around to looking at one of these things. It's a compact electronic power supply that takes mains voltage in and puts out a rough equivalent of 12V AC.
I say rough equivalent, because the output consists of a 100/120Hz modulated high frequency hump with spikes in the region of about 30V that average out to 12V as seen by a simple tungsten load.
LED lamps may tolerate being connected to these things, but may suffer damage. I always recommend using 12V DC LED drivers for your 12V LED lamps.
These units will not operate correctly if under-loaded. They require a minimum current to ensure stable oscillator start up on each half wave. If replacing lamps in ceiling lights with LED and they suddenly start flashing and flickering as a group, there's probably one of these involved. The fix is either to replace one or more of the lamps with tungsten ones again to provide a stable load, or to swap out the power supply for a 12V DC LED driver - which is the best option.
Because these units "ride the sinewave" they are dimmable and also have near unity power factor as the current is spread across the full sinewave.
It's best to avoid using these electronic transformers on the same circuits as traditional magnetic ones, as the switch-off voltage spike from the transformers may damage them. It's also best to avoid placing them where they will be covered with thermal insulation, as heat is an issue. They have a history of going bang. In the long term, a traditional properly rated core and coil transformer will outlast these electronic supplies by a huge margin.
The start up circuitry in this unit is odd. It looks like it relies on the rising positive voltage to pass a controlled pulse of current through the 10nF capacitor when the diac triggers, and then relies on the low impedance of the start transistors base circuit to avoid accidental triggering with noise.
If you enjoy these videos you can help support the channel with a dollar for coffee, cookies and random gadgets for disassembly at:-
http://www.bigclive.com/coffee.htm
This also keeps the channel independent of YouTube's advertising algorithms allowing it to be a bit more dangerous and naughty.

14 thoughts on “Inside an electronic halogen lamp driver, with schematic”
  1. Avataaar/Circle Created with python_avatars nsakpe samuel says:

    Please am looking for an alternative to power halogen lamp inside microscope

  2. Avataaar/Circle Created with python_avatars Mérnökállat says:

    Have you seen these in 6 V and 24 V verisons? They sometimes pop up on online sales, 6 and 24 V halogen downlights seem much less common than 12 V ones. 32, 36, 42 and 48 V spotlights are so uncommon that even the classic iron core 50 Hz transformer is hard to find.

  3. Avataaar/Circle Created with python_avatars Mérnökállat says:

    I tried a 20-105 W electronic halogen transformer witha 10 W lamp and it worked. If it power a 5 W lamp its lit dimly. The miltimeter cant measure the HF AC output so I used a diode and a capacitor to measure peak voltage. With no load it was about 6 V in 1 direction and about 12 V in the other direction. With a load its about 17 V in both directions.

  4. Avataaar/Circle Created with python_avatars Paul T says:

    Thanks for explaining this. I took one of these out of a vintage makeup lamp. It has a large capacitor and inductor for smoothing and two 1N diodes where the diac would be. I didn’t know why they convert AC to DC back to AC again, but then I learned about the need for feedback, isolation, and dimming. I enjoyed your explanation of the double crossover circuit. Now know there isn’t much use outside of driving a halogen bulb. Maybe it is useful for parts!

  5. Avataaar/Circle Created with python_avatars Nigel Pearson says:

    My tablet corrected my text just before sending. VA and PF

  6. Avataaar/Circle Created with python_avatars Nigel Pearson says:

    I measured a 65 VS unit made by Ring. A solid looking unit. 20 watt halogen load. 12.2 V rms once a window suiting 48 kHz was found. This also captured a good looking square wave. The odd looking Forth bridge wave was seen. Ring say OF 0.99 !

  7. Avataaar/Circle Created with python_avatars Erdem K. says:

    It's looking like a electronic CFL circuit with push-pull / half bridge configuration.

  8. Avataaar/Circle Created with python_avatars Gnome Cagnome says:

    Thank y sir!
    It's possible use this for drive a little neon tube at 12v?
    For the science can try?

  9. Avataaar/Circle Created with python_avatars Manar House says:

    Hi

    Is the little transformer in the PCB an ordinary 50/60 Hz or it workes on higher frequency?

    Waiting for your highly informative answer.

  10. Avataaar/Circle Created with python_avatars Derplays03 says:

    A fusible resistor. It’s a component which limits inrush current and acts as a fuse.

  11. Avataaar/Circle Created with python_avatars Eric Finck says:

    Hey Clive. When I was probably about 12 or so, i remember finding an old desk at the curb with a halogen light setup. I actually hooked up the transformer to a TV flyback and to my surprise it drove it beautifully! I dont think I've seen anyone on YouTube trying it out.

  12. Avataaar/Circle Created with python_avatars Genius No Copyright Music says:

    Can you explain me the circuit of halogen lamp transformer 220v AC to 12 v Ac . My transformer not giving 12v AC output

  13. Avataaar/Circle Created with python_avatars Ipsum the Purple Wuss - Loud Nation says:

    You can connect a flyback transformer to the driver and make a high voltage supply, im gonna make, its kind of like a ZVS driver ⚡

  14. Avataaar/Circle Created with python_avatars Allan Copland says:

    Don't you hate this sort of shitty SMPS PSU. No mains input common-mode filtering. Total junk!. Broadband radio-frequency noise sourse.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.