🚀 go-pugleaf

`` Can someone versed in physics please explain this phenomenon?
Given an incandescent lamp with a dimmer switch. The lamp is on.  One
can then slowly turn the dimmer knob and reduce the light intensity
until the filament is just barely glowing.  If , however the lamp is
off, and one slowly turns the dimmer switch up, no matter how slow,
the light always comes on rather suddenly and much brighter than in
the reverse process.  Why is that so ?''

Here's my shot at it....

An alternating current dimmer uses a multi-layer semiconductor device
called a triac to control the brightness of the lamp.  It does this by
varying the `conduction angle' or percentage of the waveform of the
sine wave when it is conducting.  So if the triac conducts for half of
the waveform, you'll get something like half power.  Note that a triac
is either on or off, so that you don't waste power in resistive
elements.

The triac is triggered by driving current through its `gate.'  Once
triggered, the triac continues to stay on, regardless of the gate
current, until current no longer flows through the main terminals of
the triac (but this happens automatically with alternating current).
Usually this current is derived from a capacitor which charges through
a variable resistor.  Another semiconductor device, known as a
bilateral trigger (or something like that), is connected from the
junction of the resistor and the capacitor to the gate of the triac.
Once the capacitor charges up to a certain point, the bilateral
trigger `breaks over' and the capacitor gets discharged through the
path to ground going through the bilateral trigger and the triac's
gate, and the current flow triggers the triac.  The process gets
repeated 120 times a second for 60hz power.

I've built these, so I know the above description is correct.

My speculation of why the effect in question happens is as follows:

For a transistor or diode, there is a certain threshold of voltage
that must be applied to the junction being forward biased before
you'll get current to flow through that junction.  For silicon
devices, this threshold is .6 volts.  I assume that triacs also have
this junction threshold that must be overcome.  Since semiconductor
junctions also act as capacitors, I speculate that if the triac has
been conducting recently (e.g the past half-cycle), there will be some
residual charge on the gate junction that will lower the threshold for
it to fire.  If it hasn't been conducting, there will be no such
lowering of the threshold, and you'll need a higher voltage on the
gate before it fires.

I'm not certain that the above explanation is correct.  But it makes
sense to me.  You really want to ask an electrical engineer.
--
-Fred Gilham          gilham@csl.sri.com