ARX MSX 4 Supplement

FIG.1: As soon as a monitor console is added, the
microphone’s loading impedance is halved.
amonitoringconsoleisaddedacross the F.O.H.mixer,
microphone loading drops to 600 ohms (fig.1). When
additional mixers are added for recording and/or
broadcasting, the loading falls further, down to 300
ohms when 4 mixers’ inputs are connected. The load
impedance seen by microphones is often even lower
at high frequencies, above 5kHz, due to the capaci-
tance of many metres of multicore cable, and also
the RF filtering capacitors inside each console
How excess loading affects
Microphones.
The extra loading of multiple mixers has two effects:
First, it reduces the microphone’s output level by
Dynamic Microphone behaviour
Everyday dynamic microphones are broadly as-
sumed to have an impedance around 250 ohms. In
reality, the impedance of many widely used micro-
phones fluctuates widely at resonant points across
theirfrequency range, upto2500ohms(over 10 times
higher) and down to 50 ohms*, or one fifth of the
nominal.Theseimpedance fluctuations aren’taprob-
lem when microphones are plugged into a single
mixer, presenting a load that’s close to the 1200 ohms
specified for most microphones.
The trouble starts when more than one mixer needs
to be connected across each microphone.As soon as
* eg. some models by AKG, EV, Shure and Sennheiser.
Application Notes
Application Notes
MSX 4
Active Microphone Splitter
www.arx.com.au

between 3 and 6dB. This isn’t much of a problem by
itself, but if any of the mixers is for some reason un-
plugged or switched from ‘Microphone’ (1200 ohms
inputimpedance)to ‘Line’ (10k bridging impedance),
this sudden change in microphone level can trigger
feedback/howlround in the PA system, requiring
some frenzied adjustments.
Second, and more important perhaps, the micro-
phone’s frequency response is changed. At the mi-
crophone’s resonant points, the response (with loads
below the rated 1200 ohms) may dip or peak by sev-
eral extra dB, causing a substantial change in the way
the microphone sounds. This upsets the engineer’s
judgement as to which microphone to use, and re-
quires extra EQ’ing effort in an attempt to recover
the intended sonic characteristics.
Because the response changes can involve changes
in quite sharp dips and peaks, not many console EQ
sections will be capable of fully or satisfactorily com-
pensatingthe results ofthemicrophone’sexcess load-
ing.Again, if the ‘Microphone/Line’ settings of any
of the 2,3 or 4 mixers across the microphone are sub-
sequently changed, the frequency response will be
upset, possibly causing howlround, and certainly
causing the microphone’s sound to change again, re-
quiring further EQ’ing. Even without Microphone/
Line switching, level and tonal changes can occur
when pad switches are operated, as in some consoles
theirsetting significantly alters theinputload imped-
ance.
Overall, as the PA crew have little control over what
broadcasting and recording engineers do with the
consoles in their trucks, this explains why broadcast and
recording link-ups are regarded with some trepidation...
Why Microphone Splitter Transformers
don’t help.
Microphone ‘splitter’ transformers are commonly
employed when microphones are connected to more
than one or two mixers. They usually have 1:1 ratios,
eg. 1:1:1:1 for a three-way split. It’s a common belief
that they alleviate the mixers’ loading effects. This
howeveriswrong, unless the transformer is combined
with active electronics.
Usually for reasons of cost, splitter transformers are
stand-alone, passive devices. If so, they can only ap-
portionthemicrophone’s impedance equally between
the mixers, exactly as would happen without the
transformer! Splitter transformers are still useful, as
they galvanically isolate the grounds breaking poten-
tial loops, and also restricting the potential for dam-
age and death if any circuits are accidentally con-
nected to the mains power. However, Ground isola-
tion can be achieved more cheaply by breaking shield
connections with switches, or using low value resis-
tors for a “soft float”.
Since good transformers are highly expensive, and
more affordabletransformers willalwaysadd signifi-
cant frequency response, phase and distortion aber-
rations of their own, it’s really best to avoid them
unless galvanic isolation is essential for safety, for
example, in outdoor events where wet weather is
possible, or where insurance requires it.
Benefits of an Active Split
Because an Active Microphone Splitter makes each
feed immune to changes caused by the loading pre-
sentedbyadditionalmixerconnections, one of its psy-
chological high notes is that harassed live sound en-
FIG.2: An Active Microphone Splitter allows several
consoles to be used without affecting microphone loading.

gineers can quit worrying about the mobile truck !
Use of anActive Microphone Splitter close to the mi-
crophones also improves sound quality - on at least
3 counts:
First, each microphone’s signal level is brought up
to line level before it has travelled away from the
stage, so the signal level rides 10-20dB higher above
hum, RF hash and signal breakthrough that gets
picked up along the main body of multicore cabling.
Meanwhile, the noise level (or S/N ratio) of the mi-
crophone splitter is as good as the best console mi-
crophone inputs.
Second, pickup in the multicore is typically reduced,
particulary at RF, as theActive Microphone Splitter’s
output impedance is much lower (at 50 ohms) at all
audio frequencies and up to 1 MHz and above, and
is more consistent than any dynamic microphone.
Third, by avoiding the complex loading effects of
multiple consoles, each microphone’s frequency re-
sponse and sonics are improved - back to what the
maker intended - and less EQ’ing will be required.
Finally, passing microphone signals through trans-
formers (with their well known transient, phase, am-
plitude and distortion aberrations) can be avoided
in most cases.
It’s not a complete ‘win-win’ situation, though, and
a possible downside ofActive microphone splitting,
compared to Passive splitter boxes, is a higher risk
of unreliability, since a mains power supply, and also
more parts, are required.
But, by usingtwoPower Supplies in tandem wehave
effectively a second, ‘redundant’powersupplywhich
will cut-in instantaneously if the first supply goes
down. This successfully overcomes the first objec-
tion. Meanwhile, the extra parts are all low power,
small-signal,highqualityelectronics, certainly no less
reliable than the microphone amplification in pre-
mium mixing consoles.
Inside the ARX MSX4.
The ARX MSX 4 microphone splitter contains high
quality components in a carefully considered signal
path. This provides Dynamic microphones with a
constant,ideal1200ohmload, irrespective of whether
the pad switch is in or out.
And, unlike many consoles, the switchable Phantom
power is applied in a way that prevents speaker and
nerve destroying bangs and clicks.
Powerful RF input filtering removes both common-
mode and differential nasties at ultrasonic frequen-
cies and above. High common-mode rejection at the
inputis achieved with precisionfixedparts; there are
no vulnerable trimpots. Similarly, independent dif-
ferential outputs have been specified to avoid the
need for fussy output CMR trimming, required in
conventional, cross-coupled balanced drivers. With-
out the latter’s feedback based topology, sonics are
vastly improved.
Connecting the MSX 4
Theoriginalsignal from the microphone is connected
into the Input connector on the rear panelof the cho-
sen channel. From there it can go any or all of four
separate ways.
1: To the main Frontof houseconsole,out oftheMain
connec-
tor on the
rear
panel
2: To the
Monitor
console
(or a sec-
ond Main console) out of the Monitor connector on
the rear panel
3: To either of the two front panel splits, for connec-
tionto remote trucks,OBvans, recording feeds,press
feeds, etc. In normally supplied configuration these
two splits are electronically balanced, identical to the
Main and Monitor feeds. However, they can be op-
tionallyfittedwithisolatingtransformerswhere com-
plete signal isolation is required.
Phantom power can be switched to the mic input
from the front panel, and a 20dB pad can be switched
in to cope with ultra hot signals.
Each channel can be tested with the ‘Listen’ button,
which is used in much the same way as the PFL
switch on consoles. Pressing the switch in causes the
signal on that channel to appear at the headphone
socket on the MPS 1 Power Supply.
To avoid confusion the Listen switch is non-latching,
so only one channel’s signal at a time will be heard.

MSX 4 Options
The two front panel splits are available with isolating
transformer balanced outputs if required. Ideally these
should be installed at the time of ordering the unit(s),
but they are available as a retrofittable kit. ContactARX
directly or the dealer at your point of sale for further
informationonobtaining a tranformer balancing kit. The
kit has complete details on the installation, testing and
ground lift wiring of the transformers.
Note: A reasonable amount of technical knowledge is
required for this retrofitting, so we recommend that it
be done by a qualified technician
The internal gain of the MSX 4 is set at +10dB. This is
the optimum figure for obtaining the best signal to noise
ratio. However, for applications that require the inter-
nal gain to be unity, contactARX directly for the techni-
cal application notes.
Note: Once again, a reasonable amount of technical
knowledgeisrequiredforthischange,sowerecommend
that it be done by a qualified technician
Power Connections
The MSX 4 is designed to be used with the MPS 1 Dual
Power Supply only. This unit has two completely inde-
pendent power supplies inside, and both 6 pin connec-
tors should be connected to the MSX 4 with the sup-
plied leads.
If more than one MSX 4 is being used, then the
MPS 1 should be connected as the following dia-
gram shows.
IMPORTANT
Check that the AC Power at the wall is in the same
voltage range as that printed on the fuse holder doors,
before connecting the MPS 1 to the AC supply.
Application Notes by Colin Park and Ben Duncan