12V Halogen bulb dimmer

Hi guys,

I installed a row a halogen bulbs in my house. Each has a 12V transformer.
The electrician installed the row of bulbs in parallel with an incandescent light. Curious fact is that the dimmer does not work properly if I install only the halogen bulbs but it works ok if there is a one (or more) incandescent bulbs in parallel.

My theory is that the dimmer works ok only with a resistive load (the incandescent bulb) so it has issues if I install the halogen bulbs only (would see pure inductive load). Am I correct in my hypothesis?

I'm ok with leaving one incandescent bulb in parallel, but I wonder if that is ok and whether I should instead look for a 12V-halogen bulb dimmer (assuming it exists).

Thanks

Hi,

Your assumption is correct, only one detail not:
The transformer with the halogen bulbs only will not be purely inductive.
A purely inductive load is not able to give real power, but lighting up a halogen bulb is real power.

The load is complex, with real and inductive part.

Klaus

Also worth bearing in mind that many halogen 'transformers' are actually high frequency SMPS. They are definitely not resistive loads and also need a certain amount of input power before starting to operate at all.

Brian.

Yes Betwixt, I was going to add that running the 12V SMPS from lower then its rated input voltage could damage it.
Frank

I think the lower-than-rated input POWER will damage a halogen bulb. The principle of halogen bulb operation consists in keeping halogen gas circulation between the filament and glass bulb. This circulation cools down the filament for a long life.
Reducing the power reduces the halogen gas flow and cooling, so the bulb overheats and dies sooner. In some halogen bulbs one can see the light color changing if input power(yes, defined by voltage or pulse duty cycle) is varied.

Ok, the model of the transformer in my lamps is Anchorn AET-C060B1. I could not find any data sheet on the website, but they are called "electronic transformer" so it looks like they are likely SMPS circuit. Can anyone confirm?
I have two follow up questions:

1) Why would a lower voltage damage the regulator? I'm assuming that, if you build a SMPS circuit, you would also integrate an under-voltage protection so that it starts only when the voltage is high enough to power the logic/control circuit (maybe I should not assume anything). The power sections shouldn't be at risk for lower-than-rated voltage, a part from working at low efficiency maybe. Is there anything else I'm not considering?

2) Considering jiripolivka's comment: how does one dim an halogen bulb if reducing the power is not an option? I though halogen bulbs are routinely dimmed

Thanks.

Car headlights are halogen and some cars dim them for daytime running lights except Chrysler (Fiat now?). Chrysler (and their Jeep) cars use full power high beams as daytime running lights that blind oncoming drivers in the daytime, especially when it is cloudy or raining.
I notice recently that Volkswagen and Kia cars also blind oncoming drivers with daytime running lights that are high beams that are much too bright.

Low cost dimmers rely on variable R, one or two fixed C phase shift in series with the load to trigger a diac or Triac gate. THe Series inductive excitation current of the transformer reverses the phase shift and in the case of the Halogen transformer conduct all the time or in the case of the AC LED, never conduct or have excessive hysteresis.

The advantage of these simple circuits is the Triac clamps the voltage across the trigger circuit, thus significantly reducing the power loss and cost of the parts. A slightly more expensive trigger circuit must be generate a trigger independent of load type and work directly from the line voltage and thus need extra protection from voltage transients and means to prevent power loss. Some are leading edge trigger at variable phase and shut off below holding current , while others are switch on at zero crossing voltage and use variable trailing edge cut-off.

An analysis of the dynamic impedance ratios of the gate vs the trigger source will lead to a more detailed explanation. THe Halogen bulb presenting a R/10 cold resistance is almost shorting the small transformer until hot and thus the apparent series impedance to trigger circuit is the mutual inductance.

I feel like I'm getting there with the understanding: thank you guys!

I did some parallel research and it seems that, in general, electronic transformer are dimmable, but one should use trailing edge modulation instead of the leading edge used with regular transformer or straight incandescent lights (check this out: https://www.lutron.com/europe/ResourceLibrary/362219.pdf). One could argue where is the difference, you are still chopping the voltage at the input... I read somewhere else that the reason is that the steep rising edge of the voltage will generate a high inrush current into the electronic transformer input cap (they are a mainly capacitive impedance) which I guess can damage it eventually. This also explain why my dimmer hums more than usual when I regulate the light at about 50% intensity.

Now this is all good, but, if really electronic transformer are dimmable, I don't understand why I need to have at least one incandescent bulb in parallel to my series of 5 halogen bulbs to make it work. Is it a matter of minimum load? Btw, should my electrician have connected the 5 halogens in series???

The Anchorn 'transformer' is basically a switch mode supply with no reservoir capacitor so it tries to operate at HF but only for the part of the cycle passed to it through a conventional triac dimmer control. I have replaced several of them, the ones rated at 120W which drive 6 x 20W halogen capsule lamps in parallel. They are unreliable and after looking inside them at the charred PCB, not really worth repairing. They overheat very badly.

Brian.

betwixt said:

The Anchorn 'transformer' is basically a switch mode supply with no reservoir capacitor so it tries to operate at HF but only for the part of the cycle passed to it through a conventional triac dimmer control. I have replaced several of them, the ones rated at 120W which drive 6 x 20W halogen capsule lamps in parallel. They are unreliable and after looking inside them at the charred PCB, not really worth repairing. They overheat very badly.

Brian.

Click to expand...

Hi Brian, do you happen to have the/a schematic available? I'm curious to see it...

I did trace a schematic but I can't find it now. I've moved home since I did it and the schematic probably got discarded in the clear out. From what I can remember, there is a large ferrite ring transformer at the output end with just a few turns of thick secondary wire and maybe a hundred or so turns on the primary. The circuit is basically a standard electronic ballast but has no reservoir capacitors at it's input so it starts and stops on a 'per cycle' basis according to the waveform the dimmer passes to it.

After maybe two years they fail and when opened the PCB is very darkened and the components all look well cooked. I tried replacing all electrolytic capacitors and checking what else I could but to no avail. They seem to run for a few seconds then either flicker rapidly or just go dead until cooled off again.

There are alternative 'transformers' which so far have been reliable but the boxes are slightly taller, thankfully there is just enough room for them in the ceiling fittings.

Brian.