Dimmable transfomers

Back in my "Disco Days" (~1980 :-), I was dimming neon transformers.

What I did was PWM by feeding an INTEGER number of AC cycles to the transformer primary using a TRIAC as the control element.

INTEGER is important to avoid saturation.

To dim you just skip cycles.

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
|  Analog/Mixed-Signal ASIC\'s and Discrete Systems  |    manus    |
|  Phoenix, Arizona            Voice:(480)460-2350  |             |
|  E-mail Address at Website     Fax:(480)460-2142  |  Brass Rat  |
|       http://www.analog-innovations.com           |    1962     |
             
I love to cook with wine.      Sometimes I even put it in the food.

The only problem most transformers would have with dimmers involves the symmetry of the dimmed waveform. The dimmer cuts off the early part of each half cycle to lower the average voltage. If the positive and negative trimmed half cycles do not match in total volt seconds, there is a net DC component in the transformer current and this can saturate the core (especially the toroidal ones, because they have effectively no gap at all in the magnetic flux path).

My guess is that dimable transformers have a tiny air gap in their core construction, just like class A output transformers do, so they can handle a bit more DC without saturating and becoming a short across the dimmer for a bit of each cycle. If you add a low value power resistor in series with the primary of an ordinary transformer you prevent dimmer damage and transformer heating, in the event of a little saturation. A large light bulb can perform this role.

It is also possible to design a phase control dimmer. with DC elimination, that balances the half cycle currents by active feedback.

Can anyone tell me what's the difference between a dimmable transformer and a standard chassis or torodial type?

Cheers

(snip)

(snip)

I'm not so sure about that. 10 volts DC across the primary resistance of a, say, 100 VA transformer (a few ohms) must represent a significant increase in the primary current and so, I^2*R losses. Now, if you get down to the size, regulation and winding resistance of the little transformers in a typical few watt, power limited wall wart, I would feel more confident that 10 volts wouldn't do much damage. This would make for an interesting experiment with a watt meter, an unloaded transformer and a floating adjustable DC supply in series with the line voltage.

I need to dim 12v halogen lamps for use in the garden. Upon enquiry here before, it was suggested that I purchase a dimmable transformer as I wish to dim the lamps. I've looked these up and they appear dearer than a comparable VA rated chassis or torodial transformer. My question is why? Is it something to do with the construction or electrical performance?

I read in sci.electronics.design that PigPOg wrote (in ) about 'Dimmable transfomers', on Wed, 2 Nov 2005:

In that context, a 'dimmable transformer' may be a switch-mode supply. OTOH, some people may be selling more or less ordinary (iron and copper) transformers as 'dimmable' because there is an urban myth that you can't use a phase-controlled dimmer on an ordinary transformer. This is only true if the transformer is lightly loaded or not loaded at all; phase-controlled dimmers of normal design don't like inductive loads.

Provided there is always a lamp load on the transformer, I see no reason why you can't run it from a phase-controlled dimmer. In fact, I've done temperature rise tests that showed no problem.

--
Regards, John Woodgate, OOO - Own Opinions Only.
If everything has been designed, a god designed evolution by natural selection.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

Popular misconception. Tungsten is lost much more rapidly from hot spots than from the rest of the filament. The cute regenerative mechanism that redeposits tungsten on the filament doesn't selectively redeposit it on the hot spots, and so doesn't increase the bulb life by itself.

The long life of halogens comes from the high gas pressure, which slows the diffusion of tungsten away from the hot spots, which *does* result in selective redeposition. The regenerative mechanism just keeps the envelope clean. Of course, you wouldn't be able to have the small, strong envelope to keep in the gas pressure if you couldn't keep it clean.

You can dim halogens if you like--the envelopes will darken slightly, but they clean up again once you turn them up.

Cheers,

Phil Hobbs

Thanks for the reply John.

There are a number of distributors for dimmable transformers and it's not possible to tell if these units are indeed SMPS or just, as you say, 'ordinary' transformers. My intention was to load a 200VA torodial transformer (as I have a spare) to 75 percent of it's rating (12V secondary) and phase-control the primary. I couldn't see any problems with this but the subject of dimmable units arose and it sowed the seeds of doubt in my mind - so I thought it best to seek advice here first.

Simon

Okay. That feels better.

Bad idea. Halogen lamps need to run at their normal temperature to prevent early burn out. If you dim them at all, you must dim them way down to below the temperature of a non-halogen lamp.

Since you want to dim them, use ordinary tungsten bulbs not halogens.

Ted

I'd still have qualms about dimming halogens anyway, regardless of the condition of the transformer or whatever supply.

Is there such a thing as a "dimmable halogen bulb"?

Thanks, Rich

OK, who wants to bother to do all of the arithmetic that would tell you the difference in, say, efficiency or component stresses, or something, between that and using a triac and timing the conduction cycle? I know, it's got the high harmonics, but doesn't turning even an integral sine wave on and off cause _some_ kind of harmonics?

And integral-cycle is so much harder to design. ;-P

Thanks! Rich

"PigPOg"

** Shame you were too stupid to provide any context.

You are now condemned to ambiguous replies.

........... Phil

"PigPOg" "Phil Allison"

** 12v halogen lighting "transformers" come in two kinds - iron core and electronic.

The electronic ones are feather light.

** 12v halogen transformers are legally controlled items - they must be made to a safety standard that other transformers are not required to meet.

You must NOT buy any old 12 volt tranny, especially a toroidal one and use it for 12 volt lighting.

** Most electronic lighting "transformers" cannot be dimmed with a standard wall dimmer - others often specify what kind of dimmer to use.

Read the instructions and specifications and ask the dealer.

Or, simply buy iron cored lighting grade transformers and use them with a dimmer specified & rated for such usage.

........ Phil

"PigPOg"

** They are never just "ordinary " transformers.

** That is a very dangerous and illegal thing to do where 12 volt lighting ( particularly outdoors) is involved. 12 volt lighting transformers are legally controlled items.

See my other post .

......... Phil

"Jim Thompson"

** I guess the flicker was an added bonus ?

The OP may not enjoy flickering garden lights.

.......... Phil

"John Woodgate" PigPOg

** That is no myth - it is for real. ** The worst case scenario can be when the tranny is heavily loaded ( cold lamps) and so presents an inductive load to the dimmer due to leakage inductance in series with the low resistance of a cold lamp load.

The inductive load will very likely result in "half waving" at some dimmer settings - hence a large DC component and then the smoke gets out of the transformer.

** Pulse fired triac dimmers ( using Diacs etc ) are quite unsuitable for inductive loads.

Only continuous fired ( hard fired ) triac dimmers are suitable & rated for use with transformer coupled lamp loads.

Typical, low cost wall plate dimmers are pulse fired.

........ Phil

With an AC mains supply, one type of dimmer is a choke in series with a number of filament lamps. As in theatrical stage lighting.

A choke consists of a coil of wire, in-and-out of which slides a laminated iron core. When the core is fully out, choke reactance is small enough to have no effect. When the core is fully in, choke reactance is sufficient to darken the lights.

--
Reg.