Dimmer Switches

[Trevor Smith]

Just a quickie, does having lights dimmed using a dimmer switch use less electricity than having them full brightness. Trevor Smith

[Andrew Gabriel]

Yes, but the reduction in electricity use is _very_ much less than the reduction in light output, almost to the point where it's less misleading to say No, i.e. don't bothing using this as a method to save electricity.

[Bruce]

So where does the energy go that isn't being used to make light? Heat dissipated in the dimmer itself?

[John Rumm]

Yes.

[John Rumm]

Some - dimmers usually run a little warm, but mostly in the lamp.

i.e. you can reduce the power to the lamp such that you are heating it only just to the point of it glowing visibly - however that will still use an appreciable amount of energy, and the lamp is still shining brightly, but mostly at IR wavelengths.

[Andrew Gabriel]

I guess a good way to think of it is a knob to turn the light bulb efficiency down.

[Frazer Jolly Goodfellow]

The energy doesn't go anywhere - literally. Dimmers work by electronically switching the supply off for part of each mains cycle. There's a small amount wasted in the circuitry.

Details here:

[Harry Bloomfield]

It does save some, but nothing like what you would expect judging by the light output. As the dimmer is turned down, the lamp becomes less and less efficient, producing a smaller proportion of light and a greater proportion of heat.

An efficient way to dim is to be able to switch lights off, or to switch in lower wattage lamps.

[ransley]

Yes it saves electricity, little is wasted in heat. I tested dimming and watts consumed to low watt bulbs years ago for fun so this is not by the numbers. It takes about 6 watts to fire a 100w bulb. a 100w bulb dimmed to consume 55 watts will be as bright as a 40w bulb. The dimmer is Triac switching and produces little heat, a few watts. A Rheostat dimmer converts it all to heat and are not what you buy. So you do save. to save more use a rated bulb, to save 75% more use a CFL. to save 75 to 90% use a T8. Rememner Incandecant bulbs are electric heaters that produce only 6 or so % of energy consumed into light that is visable to your eve. The rest or 94% of energy consumed is heat. If you electric cost per BTU is the same as fuels you dont waste money. But few places in the US have cheap electric, its usualy near a Hydro dam.

[John Rumm]

Well you could - but it always seems like a perverse point of view. They are called dimmer switches, not energy saving switches. If you want to control the brightness then they do a good job, and as a bonus you may save some energy as well. If you want to save energy then don't start from here. Either way, seems rather pointless complaining that the dimmer does what it says on the tin, and not what you might have hoped it would also do.

[Graham.]

I understand that a triac dimmer works by turning on for a proportion of each half-cycle. What I don't understand is if I set the knob to use exactly half of each half-cycle, why isn't that half the energy of the full-on condition?

[Dave Plowman (News)]

A small amount, yes. But the ratio of heat to light output from the bulb changes for the worse. A much more efficient way would be to have two lamps - large and small - and switch between them.

>

[John Rumm]

It would be half of the energy...

(Assuming the range of control on the knob actually goes down that far).

Note also the amount of visible light produced at half the energy would be significantly less than half as bright.

e.g. you may find the range of control on the dimmer is from 45 - 95% "on", and that corresponds to 5 - 98% brightness.

[john]

Graham. wrote: [...]

You're right, the answer isn't straightforward :) but as I've been knocking out similar calcs for the past week, I'll give it a go ... Yes logically, 1/2 way round the dial should give 50% power, cos we can easily see only 1/2 the wave getting through. Same reasoning with (say) the 25% or 67.89% positions. Problem is that the bulb only notices the total -heating effect- of the incoming voltage and this must be measured (using special kit) as the 'RMS voltage' *. This RMS value is decidedly not pro-rata with the knob position and kind of follows an "S" curve, with not a lot happening at the top and bottom 20% and most of the action ocurring in the middle bit. With 240V mains, that knob 50% position actually gives an RMS voltage of 70% of 240V (170Vrms), hence a fair bit higher than a common sense

120V. Quick measurement with a 40W bulb shows 42.0W at 100% knob, 31.1W at 50% knob, 15.0W at 25% knob. Even then, these voltages can't be easily related to bulb brightness, as that changes with summat like a voltage^3 law!. I.e we've two non linear systems in series. *'RMS voltage' = the square root, of the average, of the sum, of the squares, of the voltage of the waveshape being looked at!. This dimmer RMS voltage thing can be worked through using a calculator and a drawn single mains cycle on a squared sheet of paper but you'd need much more patience than me :(

[Andrew Gabriel]

and a third one - filament resistance goes down as filament temperature drops, which results in a higher power consumption at low voltage than you would expect for a fixed resistance load.

I did a test with a 500W halogen a few years ago. When dimmed to same light output as a 40W bulb, it was still consuming 300W, so we effectively dimmed down to roughly 1/10th* of the light output by reducing the power consumption by 40% and the efficiency by 75%. So you can get a rough feel that that when you use a dimmer, ratio-wise, it dims by reducing the efficiency at about twice the rate it reduces the power consumption.

• I didn't measure the relative light output of a 40W lamp verses a 500W lamp. It was probably more than 1:10, because a 500W lamp is much more efficient that a 40W lamp in the first place.

[Clint Sharp]

In message , Bruce writes

Negligible amounts of energy are wasted in the dimmer itself if it's an OK design.

The energy doesn't 'go' anywhere when you dim the light, a dimmer works by switching off the power for longer or shorter periods of the AC wave form so in essence it's working like a light switch and if your light is off then there's no energy used (apart from the tiny amount consumed by the dimmer circuitry).

Hopefully that made sense?

[Huge]

Because light bulbs are non-linear devices the curves of energy consumed and light output do not track one another. So a 50% reduction in energy consumed can be a 97% reduction in light output. You're looking at this from the wrong end - as someone has said, dimmers reduce the light output, not the amount of 'leccy consumed.

[Invisible Man]

Very little less. The bulb is turned on and off very fast and runs cooler. At cooler temperatures far more current flows during the periods the current is on.

A cold bulb has very little electrical resistance. When it is turned on a large current flows in the split second before it warms up. This is why bulbs blow when turned on.

[The Natural Philosopher]

heat in the bulb mostly.

[The Natural Philosopher]

Actually a typical bulb is 1-3% efficient.

2-5% for halogen, and CFLS somewhere around 10%. IIRC LEDS and striplights around 15%-25%. Sodium lamps up around that. Best IIRC is carbon arc lamps - searchlights etc.

Some LED technologies may rival striplights, but they are the best 'almost natural' light available.

Carbon arc is the best 'fully natural'.