James Newton, Host wrote:
> A 300'K lamp will have an in
> rush current about the same as its operating current,

Well yeah, considering that 300K is only 27C or 81F.  It shouldn't be much
of a revelation that when you don't change the temperature much from ambient
the resistance doesn't change much either.  In a lot of places you'd have to
cool the fillament to reach 300K.

> but a 2,500'K
> lamp will draw about 12 times the current when cold.
>
> ...
>
> So again, how do you know what the filament resistance / heat / current
> will be given a steady voltage?

The equations get pretty complicated because black body radiation is
significant at one end of the range and convection the dominant effect at
the other end.  In practise, the visible light from a bulb drops off very
quickly as the voltage goes down from the operating point.  I remember
playing with a variable transformer and a car headlight back in grade school
to see what happened when I lowered the voltage and at what point I could
just see the fillament glowing in a dark room.  If I remember right, that
was around 2/3 to 1/2 the rated voltage.

A few years ago I was working on a flight simulator project that needed to
control lots of lamps from a computer.  We used a PWM scheme, but found that
when trying to turn a bulb on full from cold, the current limit would kick
in and prevent the bulb from ever getting warm enough to increase the
resistance to bring the current below the trip point.  (It's an interesting
aside that incandescent bulbs exhibit hystersis and are bi-stable if current
controlled.  I've mused about building some device based on that
characteristic, but have never done so.)  We could fix it in software by
starting the PWM slowly with pulses short enough so that the current limit
didn't respond, but often enough to still raise the temperature.  This way
we could bootstrap the temperature to where we could apply full PWM duty
cycle without excess current.  However, this took too long to be acceptable
in some circumstances, so in the end we did this at system startup and kept
the fillaments partially warm all the time but well below visible
temperature.  If I remember right, we used about 10% of rated voltage and
applied a faster ramp of about 100mS to full brightness when requested.  It
also had the nice side effect of extending bulb life.


******************************************************************
Embed Inc, Littleton Massachusetts, (978) 742-9014.  #1 PIC
consultant in 2004 program year.  http://www.embedinc.com/products
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist