Dimmers are devices used to vary the brightness of a light. By decreasing or increasing the RMS voltage and hence the mean power to the lamp it is possible to vary the intensity of the light output.
Although variable-voltage devices are used for various
purposes, the term dimmer is generally reserved for those intended to
Dimmers range in size from small units the size of a normal
lightswitch used for domestic lighting to high power units used in
or architectural lighting installations. Small domestic dimmers are
generally directly controlled, although remote control systems (such as
X10) are available. Modern professional dimmers are generally controlled by a digital control system like DMX.
In the professional lighting industry changes in intensity are
called “fades” and can be “fades up” or “fades down”. Dimmers with
direct manual control had a limit on the speed they could be varied at
but this issue is pretty much gone with modern digital units (although
very fast changes in brightness may still be avoided for other reasons
like lamp life).
They are used instead of variable resistors because they have higher efficiency.
A variable resistor would dissipate power by heat (efficiency as low as
0.5). By switching on and off, theoretically a dimmer does not heat up
(efficiency close to 1.0).
One of the earliest recorded dimmers is Granville Woods's "Safety Dimmer", published in 1890; dimmers before that were liable to cause fires.
Early dimmers were directly controlled through the manual
manipulation of large dimmer panels, but this meant that all power had
to come through the lighting control location, which could be
inconvenient and potentially dangerous, especially with systems that
had a large number of channels, high power lights or both (such as a stage disco or other similar venues).
When thyristor dimmers came into use, analog remote control systems (often 0-10V lighting control
systems) became feasible. The wire for the control systems was much
smaller (with low current and lower danger) than the heavy power cables
of previous lighting systems. Each dimmer had its own control wires
which meant a huge number of wires leaving the lighting control
location and running to each individual dimmer. Modern systems use a
digital control protocol such as DMX to control a large number of dimmers (and other stage equipment) through a single cable.
In 1961 Joel Spira, founder of Lutron Electronics, invented the
first solid state dimmer, which switches the current on and off 120
times per second, saving energy and allowing the dimmer to be installed
in a standard electrical wallbox.
Types of dimmer
Early examples of a dimmer include a salt water dimmer. In a salt
water dimmer, there were two metal contacts in a glass beaker. One
contact was on the bottom, while the other was able to move up and
down. The closer the contacts to each other, the higher the level of
the light. Using salt water dimmers was a tedious and precarious task
that included filling the beakers with water, checking the
concentration of the salt, and raising or lowering the top contact.
Salt water dimmers were not efficient due to the evaporation of water
and the corrosion of the many metal pieces. These dimmers were
colloquially known as "pis pots", for obvious reasons. Many old theatre
electricians still recount stories of how they were initiated into the
art by being requested to "top up a pot" and receiving a shock, as
unbeknownst to them the pot was live...
Dimmers were also often based on rheostats.
These were inefficient; when set to the middle brightness levels, they
could dissipate as heat a significant portion of the power rating of
the load (up to 25% for resistive loads, more for temperature dependent
loads like lamps) so they were physically large and required plenty of
cooling air. Also, because their dimming effect depended a great deal
on the total load applied to each rheostat, the load needed to be
matched fairly carefully to the power rating of the rheostat. Finally,
as they relied on mechanical control they were slow and it was
difficult to change many channels at a time.
(often referred to as variacs) were then introduced. While they were
still nearly as large as rheostat dimmers, they were highly efficient
devices and their dimming effect was independent of the load applied so
it was far easier to design the lighting that would be attached to each
autotransformer channel. Remote control of the dimmers was still
impractical, although some dimmers (typically, for "house light" use)
were equipped with motor drives that could slowly and steadily reduce
or increase the brightness of the attached lamps. Whilst variacs have
fallen out of use for lighting they are still used in other
applications such as under/overvoltage testing of equipment due to the
fact they deliver a reasonably pure sine wave output and produce no
radio frequency noise.
Thyristor (and briefly, thyratron)
dimmers were introduced to solve some of these problems. Because they
use switching techniques instead of potential division there is almost
no wasted power, dimming can be almost instantanious and is easily
controlled by remote electronics. Triacs are used instead of SCR
thyristors in lower cost designs, but do not have the surge handling
capacity of back-to-back SCR's, and are only suitable for loads less
than about 20 Amps. The switches generate some heat during switching,
and can cause interference. Large inductors
are used as part of the circuitry to suppress this interference. When
the dimmer is at 50% power the switches are switching their highest
voltage (>300 V in Europe) and the sudden surge of power causes the
coils on the inductor to move, creating buzzing sound associated with
some types of dimmer; this same effect can be heard in the filaments of the incandescent lamps
as "singing". The suppression circuitry adds a lot of weight to the
dimmer, and is often insufficient to prevent buzzing to be heard on
audio systems that share the mains supply with the lighting loads. This
development also made it possible to make dimmers small enough to be
used in place of normal domestic light switches. European dimmers must
comply with relevant EMC legislation requirements; this involves suppressing the emissions described above to limits described in EN55104.
An alternative to the leading-edge dimming that is typically used
with SCRs is trailing edge dimming, where the falling part of the
waveform is cut rather than the rising part. This is most often used in
devices that use a switched-mode power supplies that need the front of the waveform complete so that it may cut itself.
A typical SCR based light dimmer which dims
the light through phase angle control. This unit is wired in series
with the load. Diodes (D2, D3, D4 and D5)
form a bridge which generates DC with lots of ripple. R and C form a
circuit with a time constant, as the voltage increases from zero (at
the start of every halfwave) C will charge up, when C is able to make
ZD conduct and inject current into the SCR the SCR will fire. When the
SCR conducts then D1 will discharge C via the SCR. The SCR
will shut off when the current falls to zero when the supply voltage
drops at the end of the half cycle, ready for the circuit to start work
on the next half cycle.
Sine-wave dimming promises to solve the weight and interference issues that afflict thyristor dimmers. These are effectively high power switched-mode power supplies. They rely on a new generation of insulated gate bipolar transistors (IGBTs) which are still relatively expensive.
See Lighting control console
Non domestic dimmers are usually controlled remotely by means of
various protocols. Analogue dimmers usually require a separate wire for
each channel of dimming carrying a voltage between 0 and 10 V.
Some analogue circuitry then derives a control signal from this and the
mains supply for the switches. As more channels are added to the system
more wires are needed between the lighting controller and the dimmers.
In the late 70s serial analogue protocols were developed. These
multiplexed a series of analogue levels onto a single wire, with
embedded clocking signal similar to a composite video signal (in the
case of Strand Lighting's European D54 standard, handling 384 dimmers) or separate clocking signal (in the case of the US standard AMX192).
Digital protocols, such as DMX512 have proved to be the answer since the late 80s. In early implementations a digital signal was sent from the controller to a demultiplexer,
which sat next to the dimmers. This converted the digital signal into a
collection of 0 to +10 V or 0 to -10 V signals which could be connected
to the individual analogue control circuits.
Modern dimmer designs use microprocessors to convert the digital
signal directly into a control signal for the switches. This has many
advantages, giving closer control over the dimming, and giving the
opportunity for diagnostic feedback to be sent digitally back to the
Dimmers are usually arranged together in racks, where they can be
accessed easily, and then cables are run to the instruments being
controlled. In architectural installations cables are run straight from
the dimmers to the lights. However venues such as theatres demand more
flexibility. The lighting rig may change dramatically for each show,
and occasionally during shows. Many theatres have cables run
permanently to sockets (called circuits) around the theatre, however
not all the sockets are needed for each show, so there will be fewer
dimmers than there are circuits. A patch bay usually sits next to the
dimmers enabling the dimmers to be connected to specific circuits. The
patch bay may also enable many circuits to be connected to one dimmer
and even series connection for low-voltage lamps. This patch bay is
known as the mains or hard patch. Analogue dimmers may also have a soft
patch to match output channels from the lighting controller to control
selected dimmers. Most new installations do not use patch bays, instead
they use a dimmer-per-circuit and patch dimmers into channels using a
computerized control consoles.
The design of most analogue dimmers meant that the output of the
dimmer was not directly proportional to the input. Instead, as the
operator brought up a fader the dimmer would dim slowly at first, then
quickly in the middle, then slowly at the top. The shape of the curve
resembled that of the third quarter of a sine wave. Different dimmers
produced different dimmer curves, and different applications typically
demanded different responses.
Television often uses a "Square" law, providing finer control in top
part of the curve, essential to allow accurate trimming of the colour
temperature of TV lighting. Theatrical dimmers tend to use a softer "S
curve" or linear curve. Digital dimmers can be made to have whatever
curve the manufacturer desires and may have a choice between a linear
relationship and selection of different curves, so that they can be
matched with older analogue dimmers. Sophisticated systems provide
user-programmable or non-standard curves, and a common use of a
non-standard curve it to turn a dimmer into a "non-dim", switching on
at a user defined control level.
Example dimmer curves:
Some types of incandescent (filament) lamps should not be switched to full power from cold, and doing so can shorten their life dramatically owing to the large inrush current that occurs. To soften the blow to the lamps slightly, dimmers may have a preheat
function. This sets a minimum level, usually around 5-10%, which is not
obvious to the audience, but stops the lamp from cooling down too much.
This also speeds up the instrument's reaction to sudden bursts of power
which operators of rock'n'roll style shows appreciate. The opposite of
this function is sometimes called top-set. This limits the maximum power supplied to an instrument, which can also extend its life.
The digital revolution
Modern digital desks can emulate preheat and dimmer curves and allow
a soft patch to be done in memory. This is often preferred as it means
that the dimmer rack can be exchanged for another one without having to
transfer complicated settings.
One measure of the quality of the dimmer is the "rise time". The
rise time in this context is the amount of time it takes within the cut
part of the waveform to get from the zero-point crossover to the start
of the uncut part of the waveform. A longer rise time reduces the noise
of the dimmer and the lamp as well as extending the life of the lamp.
Unsurprisingly, a longer rise time is more expensive to implement than
a short one, this is because the size of choke has to be increased.
Source: Wikipedia (All text is available under the terms of the GNU Free Documentation License and Creative Commons Attribution-ShareAlike License.)