Re: AC dimmer software development question

From: Jim Granville (no.spam_at_designtools.co.nz)
Date: 02/19/04 

Date: Thu, 19 Feb 2004 12:02:11 +1300

Meindert Sprang wrote:

> "rickman" <spamgoeshere4@yahoo.com> wrote in message
> news:40330918.14FF2F5@yahoo.com...
>
>>A lamp is excited by 120 Hz, not 60 (power waveform vs. voltage) and the
>>filament of a typical light bulb does not go out during cycles.
>
>
> Many years ago, I tried a "poor mans' dimmer" by switching a diode in series
> with a lamp, giving me off, 50% and 100%. I could clearly see the lamp
> flicker when running on only one half of the sinewave. So imagine what
> happens if you want to dim to lower intensity.

  If this is not for commercial deployment, you can do anything you like.

  If it is for a commercial product, then the OP needs to watch many
effects :

  Half-waving presents a DC load on distribution transformers, and is a
definite no-no.
  Your SW design should work to ensure BOTH half cycles conduction angles
are the same. Can be as simple as calculate a new value every whole
cycle, as well as care to ensure the zero crossing is balanced.

  There are also sub-mains flicker regulations, as well as bulb life
issues that would pretty much kill cycle-skipping schemes.

  Also, most mains SMPS now have power factor correction - which means
a pre-controller, in order to spread and reduce the very peaky bridge
diode currents you get with mains SMPS.

  Think about it:
The mains grids have a finite impedance, and transformers do not like DC
load components.
  Imagine : You ship large numbers of dimmers using any cycle-skip
scheme, and there is a risk some will phase-lock and impose your flicker
onto everyones lighting :)

Relevant Pages

  • Re: CRT randomly loses power?
    ... You can't even see flicker over about ... > norms are about 72-75Hz. ... and a few who can tell if it's 75 cycles and higher. ... frequency for merging successive visual images is a *nominal* frequency. ...
    (comp.os.linux.misc)
  • Re: CRT randomly loses power?
    ... You can't even see flicker over about ... > norms are about 72-75Hz. ... and a few who can tell if it's 75 cycles and higher. ... frequency for merging successive visual images is a *nominal* frequency. ...
    (comp.os.linux.hardware)
  • Re: CRT randomly loses power?
    ... You can't even see flicker over about ... > norms are about 72-75Hz. ... and a few who can tell if it's 75 cycles and higher. ... frequency for merging successive visual images is a *nominal* frequency. ...
    (comp.os.linux.setup)