Lab.gruppen PLM Series Operation Manual - 12 Application Guide
68 PLM Series Operation Manual
referred to as the latency of a piece of equipment.
Ultimately, the delays may become large enough to be
audible and lead to phasing problems. For this reason,
it is desirable to use digital interconnections rather
than analog ones between the various components
of a digital audio system (mixer, outboard, crossovers,
amplifiers, etc.) wherever possible. Doing this will also
maximize audio quality, as other technical problems
(such as cable losses) will also be avoided.
Any audio signal in a digital audio system has a sample
rate associated with it. The two most common sample
rates are 44.1 kHz, used in consumer applications
such as CDs, and 48 kHz, used in professional and
broadcast applications. Multiples of these rates, e.g.
96 kHz, may also be encountered.
Care needs to be taken to ensure sample rate
compatibility between a digital audio source and the
equipment it is driving. Incompatibility will result in
silence! Much recent equipment employs universal
sampling rate conversion (generally referred to as
“SRC”) on inputs to resample the input signal if
necessary.
This leads to maximum flexibility in interfacing
different equipment, but is achieved at the expense of
increased overall latency, typically by about 2 ms.
It is desirable in live sound applications to keep the total
latency of the system to a minimum. One method of
doing this is to interconnect digital equipment digitally
- via an AES/EBU interface - wherever possible, thus
avoiding ADC & DAC delays Another is to ensure that
all digital equipment is set to the same sample rate,
thus avoiding SRC delays.
Sound engineers have traditionally corrected for
acoustic path differences – typically between different
speaker clusters – by adding delays to the various
amplifier feeds. In a wholly or partly digital system,
it is also necessary to take account of latency delays.
Having set a system up, including any outboard
processing, the system architecture (the “wiring”)
should not be altered once the speaker clusters have
been aligned as any alteration in latency may nullify
the time alignment. The PLM’s Dolby Lake system
permits delay to be added per-channel; adjustments
of a few milliseconds are usually adequate to achieve
coherence between clusters.
12.5.3 Interconnecting digital audio
Some interconnection methods commonly used with
analog audio may not possible when using AES/
EBU digital audio. If it is required to connect an AES/
EBU output to multiple AES/EBU inputs (such as
multiple PLMs), one of the following methods should
be employed:
Connect the source signal to the AES/EBU input
▸
on the first PLM. Connect the dedicated LINK
output on this PLM to the AES/EBU input on
the next. Continue to daisy-chain the PLMs in
this way until the last, using separate cables in
each case. Factory tests indicate that it may be
possible to daisy-chain up to 20 PLMs in this
manner, depending on environmental conditions,
cables type etc. This method has the advantage
of minimizing latency between units; however,
it does present a ‘single point-of-failure’ which
may be a consideration.
Use an AES/EBU distribution amplifier* with at
▸
least as many outputs as the number of PLMs
being driven. Connect the AES/EBU source
signal to the DA input, and use separate cables
to connect each DA output to the AES/EBU input
on each PLM.
*It may be possible to use a passive splitter instead
of a DA under some circumstances – see section
12.5.5.1.
12.5.4 Cable
It may be necessary to pay greater attention to the
choice of cable used to transmit AES/EBU digital audio
than with analog audio. Because the digital audio
signal is essentially high-speed data, cables suitable
for a far higher bandwidth are needed (typically 12
MHz instead of 20 kHz). It is also more important
that cable of the correct characteristic impedance
(110 ohm) is used. Impedance mismatching can
produce reflections within the cable, which in turn
can produce clocking and data errors.
Standard XLR microphone cables may work in
digital audio but this cannot be guaranteed under all
conditions. Cable types suitable for microphones may
have inadequate and unpredictable properties at the
high frequencies needed for data transmission.