Klark Teknik HELIX DN9340E, Руководство оператора

Страница: 73 / 78

Application Notes
63
16.3. What is AES/EBU?
AES/EBU refers to the professional digital audio transmission system, jointly specified by the Audio
Engineering Society (AES) and the European Broadcast Union (EBU), which allows the transmission of
two channels down a shielded twisted pair cable using time division multiplexing (TDM) with one
sample from each channel being transmitted within the sample period of the system. The system
was published as the AES3 standard.
By virtue of TDM, one XLR cable carrying AES/EBU data can replace two regular analogue connections
- hence only single in/out connectors are required on the AES/EBU Unit. The clock for the data
transmission, which is embedded in the incoming data using a process known as bi-phase mark
encoding or Manchester encoding, can also be used as the clock source for the master clock within a
unit equipped with an AES/EBU interface. This is the preferred mode of operation, as it guarantees
that the unit is synchronised to the incoming data stream.
For large digital transmission systems using AES/EBU interfaces, such as those encountered in
broadcast and studio installations, a clock signal operating at the sampling frequency can be
distributed to all units separately to the AES/EBU signals. This is generally known as 'word clock' and
allows all connected units to be synchronised on a sample-accurate basis. Word clock is most
commonly connected using 75 Ohm, BNC connectors.
All DN9000E series Helix units feature sample rate converters (SRCs) on their AES/EBU inputs and, in
the case of the DN9340E Helix Dual EQ and DN9344 Quad EQ, SRCs on their AES/EBU outputs as
well. This allows the units to interface to AES/EBU systems operating up to 96kHz, including the
Midas XL8 Live Performance System.
The SRCs may be bypassed in 44.1 kHz and 48 kHz AES/EBU synchronised transmission systems
(48kHz only in the case of the DN9848E System Controller), which eliminates propagation delays
introduced by the SRCs and reduces the input-to-output latency of the signal paths to that incurred
by the actual digital signal processing algorithms.