Electrical Engineering - dimmer switch for leds

Author

dimmer switch for leds

Helen Lurie

2006-11-21, 3:25 am

I bought a light dimmer from IKEA.

If I connect my dimmer switch to a normal bulb and slide the controller
from off to on slowly, the light intensity slowly increases from off to
full brightness.

If I connect a string of LEDs to the dimmer and do the same sliding
from off to on, the bulbs first increase in brightness, then decrease
in brightness to almost off, then increase again to full brightness.
Why?

I'd appreciate any help you can offer.

Thanks
~Helen

(PS) I did a little research on how dimmer switches work, and they work
by turning on the incoming AC signal only when it reaches a certain
nonzero threshold, then keeping it on for the rest of the cycle. (The
technology is called TRIAC - see http://en.wikipedia.org/wiki/Triac ).
But, I don't see why this would cause the local minima in the led
current.

Andrew Gabriel

2006-11-21, 3:25 am

In article <1164093426.207981.116600@h48g2000cwc.googlegroups.com>,
"Helen Lurie" <HLurie@gmail.com> writes:
> I bought a light dimmer from IKEA.
>
> If I connect my dimmer switch to a normal bulb and slide the controller
> from off to on slowly, the light intensity slowly increases from off to
> full brightness.
>
> If I connect a string of LEDs to the dimmer and do the same sliding
> from off to on, the bulbs first increase in brightness, then decrease
> in brightness to almost off, then increase again to full brightness.
> Why?

Two possible reasons...
Simple triac dimmers have a minimum load requirement for
proper operation. It may be that the load presented by the
LED power supply is too low, or that it doesn't present a
resistive load and meet the minimum requirements throughout
the whole waveform. Add a regular lamp too, and see if that
fixes the problem.
Secondly, the LED power supply may well not be designed to
take a phase controlled input waveform and generate a
varying LED current from it. Does it claim to be dimmable?

--
Andrew Gabriel

Helen Lurie

2006-11-21, 3:25 am

Andrew Gabriel wrote:
> It may be that the load presented by the
> LED power supply is too low, or that it doesn't present a
> resistive load and meet the minimum requirements throughout
> the whole waveform. Add a regular lamp too, and see if that
> fixes the problem.

I will try this, good idea. Now to find a 2-pronged outlet extender...

> Secondly, the LED power supply may well not be designed to
> take a phase controlled input waveform and generate a
> varying LED current from it. Does it claim to be dimmable?

No, the LED string does not claim to be dimmable. I'm sure the LED's
AC->DC power supply is not intended to be used this way, but I just
want to figure out from a circuits point of view what the likely source
of this behavior is.

Thanks for your help.

EpsilonRho

2006-11-21, 9:25 am

SIMPLE dimmers need a resistive load (for example a bulb filament) and in
particular don't like to drive loads with conduction thresholds, like a
bunch of LED's in series . Test this: put in parallel to the LED cluster a
normal low wattage incandescent bulb.
I bet that in this case you will have a perfectly working dimmer, because
now, at least, some of the current passes through a resistive element (bulb
filament). An analysis of a typical simple dimmer circuit would easily
explain the reason

vandra

2006-11-21, 9:25 pm

to turn a diode into a resistive load, you should put a resistance with
it.

EpsilonRho wrote:
> SIMPLE dimmers need a resistive load (for example a bulb filament) and in
> particular don't like to drive loads with conduction thresholds, like a
> bunch of LED's in series . Test this: put in parallel to the LED cluster a
> normal low wattage incandescent bulb.
> I bet that in this case you will have a perfectly working dimmer, because
> now, at least, some of the current passes through a resistive element (bulb
> filament). An analysis of a typical simple dimmer circuit would easily
> explain the reason

vandra

2006-11-21, 9:25 pm

LinkBot