2 the feedback destroyer in the shark – Behringer SHARK DSP110 User Manual

Page 17

Advertising
background image

17

SHARK DSP110

2.2 The Feedback Destroyer in the SHARK

The SHARK identifies feedback by splitting the entire frequency spectrum (20 Hz to

20 kHz) into sections of 1/60 of an octave and determining the level of each of these

bands. The values calculated are then referenced to the level of the overall signal. The

resulting level difference determines whether or not a filter is set. The SHARK allows

you to adapt these decisive parameter to your needs: within a range from 1 through

100 you can edit the feedback detection sensitivity. The standard setting is 50, which

ensures the best possible detection of feedback for the majority of applications. For

speech-only applications you can raise the feedback detection threshold towards 100,

which enables the algorithm to detect and remove feedback even more quickly. Vice

versa, lower values provide for a more stable feedback suppression responding less to

wanted feedback-like signal portions produced by guitars or keyboards.
In FILTER LEARN mode, feedback is generated and suppressed automatically.

Whenever it detects feedback, the DSP110 selects the filter parameters automatically

to efficiently remove the feedback. As the filter is set to the frequency detected, this

mode is ideally suitable for suppressing constant feedback frequencies produced by

“fixed” microphones, e.g. those used on drums. Once set, the filters automatically enters

lock mode, i.e. the frequency remains fixed but width and depth of the filter are still

being adapted to the signal. The filter width is enlarged whenever the feedback frequency

shifts slightly. If feedback persists, gain is reduced even more and kept low to prevent

feedback from recurring.
All microphones that are moved during a performance (e.g. hand-held vocal

microphones) are usually susceptible to varying feedback frequencies, which should

be suppressed in automatic search mode (entered when you power up the SHARK).

Much like in FILTER LEARN mode, a filter automatically determines the ideal settings

for all parameters, in order to suppress feedback. However, once all filters have been

set, the filter first activated gets reset to automatic search mode. Thus, the SHARK

makes sure that there is always one free filter to identify and remove new feedback

frequencies. If your music contains wanted feedback elements (e.g. guitar feedback),

the SHARK will suppress these too, because it is impossible from a physical point of

view to distinguish wanted from unwanted feedback. Section 2.1.6 provides some

information on how to get around this physical problem.

2. APPLICATIONS

Advertising