Light Flicker
An interesting issue to grapple with is the concept of "light flicker".
This is loosely defined as the impact on human vision perception by the impact
of a voltage event on a coiled filament incandescent lamp. More accurately
defined in IEC publication 868 and AS/NZ61000-3-3 and AS/NZ4376 and 4377
which describe an instrument to measure this parameter (called a "flickermeter")
If we do not have ready access to a "flickermeter" we must measure several
parameters and calculate the apparent "flicker".The
two important parameters
are RMS voltage dip and duration. To further complicate matters, the voltage
dipis not always constant during the start.
It is the voltage sag at the meter box ( PoCC or Point of Common Connection
)
that influences other local loads when the air-conditioner has its own dedicated
circuit. Thus the voltage sag experienced by other circuits connected to that
meter box ( lighting in particular) is not the same as experienced by the air-conditioner.
The sag at the air-conditioner is often much higher due to the additional run
of cable.
1. Line impedance.
Total impedance to the meter box is the sum total of all series
impedance's all the way back to the collective "power station". Obviously only
a small
part of this system is directly within our control. This part is also that part
that often
has the highest single contribution to the total impedance. It is therefore
conceivable
to upgrade the connection between the meter box and the pole.
2. Motor starting
currents. Without any form of soft-starter the DOL (Direct On Line)
starting of these motors could cause excessive light flicker. To meet local
power supply requirements and to ensure the customer is not irritated by poor
power quality a softstarter is required. It is important to select the type
of softstarter that provides the greatest reduction and simultaneously the quickest
starting time.
Attached is a copy of the graph of irritability vs voltage sag, with a vertical
line
representing five starts per hour which gives us a maximum of 5% voltage sag
before "irritability" is reached. (Note that if a greater number of starts per
hour is
expected the value of 5% should be decreased, however in NSW at least, 5% has
been nominated as a simplistic target to be achieved).
As a final point,
I should stress the importance of using the correct instruments and
interpreting the results carefully. For example the results of the above tests
when
recorded on a range of differing brands of tong (clamp) meters when set to recording
peak current values will often be totally inaccurate. Most meters can not read
accurately any event less than 500mS and are usually fooled by waveforms that
are not sinusoidal.