Manley STEREO VARIABLE MU LIMITER / COMPRESSOR - 3/2004 - Manual (Page 11)

NOTES ON MASTERING MODIFICATIONS
The Mastering Version of the Variable MU Limiter Compressor uses expensive Greyhill rotary switches with gold
contacts where conductive plastic pots were used. The steps are determined with a large number of 1% precision metal
film 1/2 watt resistors. The best conductive plastic pots only have 10% or 20% tolerance. The ten fold improvement in
precision helps a great deal in left-right matching. There is a subtle audible improvement with stepped switches as well.
Audiophile HI-FI often uses that technique to wring the last drop of performance out of a pre-amplifier. And I bet you
wanted stepped switches mostly for resets.
The INPUT LEVEL is a five position switch with a generic optimum setting of “0” in the 12:00 position. Each step in
either direction is 2 dB. For reference, Unity Gain is “0”.
The Output Attenuators are in half dB steps. Reference Unity is -11.5 or fully counter-clockwise. You might think of
these as “gain makeup”. They are marked technically, in that “0” or fully clockwise has zero attenuation in the circuit.
The tube circuit actually has 15.5 dB of gain. The Input attenuator at “0” removes 4 dB and the Output attenuator re-
moves the last 11.5 producing “unity”. With a little compression the “gain make-up” available with the Output Attenu-
ator is very handy.
The Threshold is in half dB steps calibrated to LIMIT mode. In Compress the steps are approximately 1/4 dB. There are
24 steps so LIMIT gets a 12 dB range and Compress has a 6 dB range of adjustment. In some cases it is common to use
the Input Attenuator to find a good starting point. Some Mastering engineers find using the combination of Input, Output
and Threshold to achieve a little different “drive”. Another good reason for stepped gains.
The Attack Time has been slightly extended in both directions compared to a regular Variable-MU and divided into 11
steps. The approximate values are: CW (fast)15mS, 20, 25, 30, 35, 40, 50, 60, 70, 80, 90mS (slow) CCW.
The Recovery has always been a 5 position switch. The times are: 200mS, 400mS, 600mS, 4S, 8 Seconds.
NOTES ON M/S MODIFICATIONS
The M/S modification allows various approaches to Mid/Side recording, playback and processing. There are separate
switches for both input and output. Normal STEREO operation is with both switches pushed towards the right. M/S
mode is with the switches to the left. The left switch changes the way the transformer inputs are wired (instead of an
“active circuit”) to derive M/S. The right switch changes the output wiring.
When both the Input and Output switches are in M/S, then if one feeds normal left and right signals - the output will
also be left and right. The signal would be “encoded” then “decoded”. This can be useful because the compressor can
process mid and side differently. Now the “Channel One” Threshold, Attack and Recovery are for the mid, sum or mono
part of the sound - “Channel Two” Threshold, Attack and Recovery are for the side or difference part of the sound. If
the Link switch is SEP then adjustment of “width” is possible using the Threshold controls and some gain reduction. To
“widen” simply compress more in Channel One. In LINK mode the adjustment of width is not possible but one can use
more of the “Mid” (or “Side”) to control overall Limiting. This is useful for material meant for stereo broadcast. FM
splits the signal into M/S and transmits as M/S. Broadcast engineers are generally more concerned that the mono com-
ponent is solid because it transmits much further - and that relates to audience size and perceived quality.
To “decode” mics in the M/S configuration one can either switch the input or output to M/S. Generally the input would
be switched to M/S and the compressor and all after it will now have left and right.
To “encode” program for FM broadcast and bypass the transmitter encoder, either the input or output switches can
be used. If the unit is to be used for compression then it is normal to follow the output with a very good Limiter. This
approach sounds louder and less “processed” than multi-band processors. It also allows a finer degree of control to the
critical mono signal.
11