ECN Forum Forums General Topics of the Trade General Discussion Area dimmer on transformer

A good explanation of how a phase angle control works (with waveforms) is available here:

http://ubasics.com/adam/electronics/doc/phasecon.shtml

The steep wavefronts generated by chopping the sinewave like that generate a LOT of RF noise, which needs to be filtered out to prevent radio interference.

Every incandescent dimmer I have looked at had specific warnings NOT to use it to control inductive loads, such as the primary of a transformer or ballasted HID/fluorescent lighting. Presumably, the fast switching of the current through the load would generate high voltage spikes that will damage the TRIAC. There are similar types of controllers intended for use with motors (fan speed controls) which might be a better choice if you want to control a LV lighting transformer.

The best choice for varying the output of a transformer is to use a variable autotransformer or "Variac" on the input. These produce a variable voltage, clean sine wave output like a rheostat would, but without the terrible efficiency/heat dissipation problems of a resistive control. They are somewhat large and expensive, though.

I usually use Lutron Diva dimmers for ballasted fixtures. I find I'm trying to lay my hands on the $80 one used exclusively on electronic ballasts more often than the model used in magnetic ballasts. I'm guessing the main difference between the two is the frequency they are tuned for.

I have had a customer hire me to remove a pair of Lutron Smart dimmers (incandescent) because the RF they generated interfered with their habit of listening to AM radio in the kitchen.

Okey Dokey, I was wrong
The transformer does reflect the input to the output almost exactly. I guess I should have known that because of the way we couple audio. I am not sure why I thought it would smooth out the signal. ... sorry.

http://esteroriverheights.com/electrical/dimmer_on_transformer.jpg

I guess the next question is, does that really matter if the load is a light bulb? I can't imagine it hurts the transformer in any way.

Oops, I was wrong about the frequency thing too. And I should have known better because I had read this page a while back.

http://www.lutron.com/technical_info/pdf/LutronDimmingBasics.pdf

A good reason not to use an ordinary light dimmer in series with a power transformer is that if the Triac is not conducting symmetrically on both positive and negative swings of the AC sine wave, then a DC component will be created.
As a transformer primary usually has a low resistance, a significant DC current will flow causing heating of the winding as well as core saturation. This in turn reduces inductance of the winding.
It's this principle that is behind saturable reactors used for power control.
The best way to deal with the situation is to use a Variac, or a light dimmer designed for transformer loads, which will be designed not to cause a DC component, as well as incorporating the correct RC network to deal with inductive spikes created.

Where is the DC going to come from ... unless you consider zero volts a DC level? When current is flowing in the triac it is AC although not a pretty sine wave.

Assuming a Triac is not switching symmetrically then the area under the curve will not be the same on both sides of the zero axis. This is what creates a DC component. As an example, say the triac is being triggered at 90 degrees on the positive cycle and 85 degrees on the negative cycle. More current will flow into the load on the negative cycle than on the positive cycle. So, there will be a slight negative DC offset to the AC waveform.
You can prove this by observing such a waveform on an oscilloscope with the input set to DC coupling.

Cheap/poor QC dimmers may be unintentionally providing some rectification to the output signal if the triacs are not providing equal gating for + and - current, just like if you clipped the waveform with a diode, only not as severe. So, instead of getting, say, equivilent to +/-60V, you get +65 & -55V: superimposed 60VAC+5VDC.