[evan_au]

The diagrams below show the sine-wave AC voltage waveform of a 240VAC mains supply.
- The peak voltage rises above 240V, but on average, it will deliver the same power as 240V DC
- You have to imagine this pattern continues for 100s or millions of cycles

The next large graph shows the current that would be drawn by a 100W incandescent light bulb.
- This is a resistive load; it gets more complex if the load is inductive or capacitive
- This is effectively the same current that would be applied to the light bulb if you set your dimmer to "maximum brightness"

The smaller graphs on the right show what happens if you turn the dimmer knob farther towards the "dim" setting.
- The TRIAC acts like a very fast power switch, which is initially OFF
- After a small delay, a timer circuit turns on the TRIAC switch, turning on the light bulb
- The TRIAC switch turns off again when the AC voltage crosses 0V
- Depending on how you set the dimmer knob, the current to the light bulb may be turned on almost all the time, or hardly at all.
- This allows you to adjust the brightness of the light from almost full brightness to almost off.
- Switching the light bulb on and off 100 or 120 times a second is not usually visible, because of human persistence of vision, and the thermal inertia of the light filament.

In the Wikipedia article, the timer circuit consists of:
- a variable resistor that you can turn
- a fixed capacitor
- a fixed Zener diode
- This allows you to adjust the switch-on angle from 0 to 180 degrees

In some countries, non-critical power loads (like off-peak water heaters) may be controlled by the electricity supplier by injecting signals on the power lines (you get a fee rebate for this off-peak power).
- The signals slightly move the zero-crossing point of the AC waveform
- This can result in a cyclic variation in the TRIAC switch-on point, and variation of the lamp brightness
- This is visible as a cyclic variation in lamp brightness while the AC signaling voltages are present.

The TRIAC current contains many vertical lines, which indicate the presence of high frequencies in the current, which were not present in the original smooth sine-wave AC voltage. These high frequencies are what destroyed the cheap power meter.

[homebrewer]

Every Fluorescent bulb contain a small amount of Mercury Hg 80, which evaporates when the enhanced start-up voltage strikes the electrodes., in a standard Fluorescent light circuit.  Without the recommended start-up voltage circuit the Hg 80 can not evaporate und will cause an destructive low frequency pulse in high current in the Triac circuit of a standard dimmer, designed for incandescent light bulbs.

[vhfpmr]

It's quite complicated isn't it? I could really do with a walk-through diagram to show what is happening at each stage of the waveform to understand how it works.

If you install a (free) copy of LTspice, there's a dimmer in the examples folder which you can play with to your heart's content: http://www.linear.com/designtools/software/

This is a plot of relative power as a function of triac trigger angle:

[syhprum]

I wonder if an incandescent lamp cap be considered as a purely resistive load, due to the low thermal inertia of the filament and the rapid change of resistivey of Tungsten with temperature the current is in no way proportional to the voltage this would seem to give a capacitive element to the impedance.
It would be interesting to feed a lamp via an inductor from a DC source and see if an oscilation could be produced
 

[evan_au]

It's true that the resistance of the filament is a fairly non-linear function of temperature: the resistance increases as temperature increases to its operating temperature. The temperature range is pretty wide from room temperature (around 300K) to operating temperature  around 3000K; the resistivity changes by a factor of 16 over this temperature range.
See: http://physics.usask.ca/~bzulkosk/Lab_Manuals/EP354/ep354_thermal-radiation-lab_Tungsten-Temp-Resistivity.pdf

But the pulses of electricity occur fairly close together in time (8-10ms, depending on country). As a teenager, I used a photocell to detect the change in light output from incandescent and fluorescent lamps due to mains frequency. But the variation in temperature and resistance of an incandescent lamp is not huge, once it reaches operating temperature.

The crucial effect in capacitance is that the current increases before the voltage increases (I leads V).
- It is the opposite for inductance: V leads I.
- Both are energy storage devices - they continue to deliver a current, even when the voltage has dropped to zero.

For an incandescent lamp, the pulse of voltage increases temperature, which decreases current a bit later (V leads I, like an inductor).
But when the pulse passes its peak, the current does not continue to increase while the voltage holds the same polarity, but starts to decrease with the voltage (unlike an inductor).
- When the voltage reaches zero, so does the current.
- This is the essential characteristic of a resistor, not an inductor or capacitor

It is true that every resistor has a tiny bit of capacitance and inductance. But we can say that the capacitance and inductance of a light bulb is pretty negligible compared to the 1kΩ resistance of a 60W/240V incandescent globe.