Mic Splitter Curiousity

[Frank Stearns]

I'm rebuilding my splitter (new connectors this time; it's 40 years
old), and noted 150 ohm resistors in series with the direct outs,
both legs. There are also 150 ohm resistors in series on each leg of
each iso out (two iso outs per channel).

I'm wondering what these series resistors are for -- only thing I can
think of is that any output -- direct or iso -- can go into a dead
short and all the other outputs will continue to work (though it
seems a little odd that the isos have these).

I've looked online and see many different variants in the design of
transformer iso'd splitters. Some have these series resistors and some
don't; only have them on the direct out; use values other than 150 ohms,
etc.

I'd perhaps like to remove those resistors from the direct outs, but
don't want to mess up anything.

Comments as to the purpose of these resistors?

Thanks in advance,

Frank
Mobile Audio
--

[Phil Allison[_4_]]

Frank Stearns wrote:
>
>
> I'm rebuilding my splitter (new connectors this time; it's 40 years
> old), and noted 150 ohm resistors in series with the direct outs,
> both legs. There are also 150 ohm resistors in series on each leg of
> each iso out (two iso outs per channel).
>
> I'm wondering what these series resistors are for -- only thing I can
> think of is that any output -- direct or iso -- can go into a dead
> short and all the other outputs will continue to work (though it
> seems a little odd that the isos have these).


** I think you are dead right.

The isolated windings need them because a short across one would be reflected to the other windings on the same tranny and drop the level 26dB or more.

The "isolation" is really only at DC voltages, for phantom power and not the actual signal. Achieving both requires an active splitter and multiple transformers.


..... Phil

[[email protected]]

On Thursday, December 3, 2015 at 2:14:36 AM UTC-5, Phil Allison wrote:
> Frank Stearns wrote:
> >
> >
> > I'm rebuilding my splitter (new connectors this time; it's 40 years
> > old), and noted 150 ohm resistors in series with the direct outs,
> > both legs. There are also 150 ohm resistors in series on each leg of
> > each iso out (two iso outs per channel).
> >
> > I'm wondering what these series resistors are for -- only thing I can
> > think of is that any output -- direct or iso -- can go into a dead
> > short and all the other outputs will continue to work (though it
> > seems a little odd that the isos have these).
>
>
> ** I think you are dead right.
>
> The isolated windings need them because a short across one would be reflected to the other windings on the same tranny and drop the level 26dB or more.
>
> The "isolation" is really only at DC voltages, for phantom power and not the actual signal. Achieving both requires an active splitter and multiple transformers.
>
>
> .... Phil

doesn't the multiple winding audio transformer isolation also prevent ground loops.

If the device was just for isolation of DC phantom power, that could be accomplished simply using DC blocking caps and no need for an expensive transfomer

Mark

[Scott Dorsey]

Frank Stearns > wrote:
>I'm rebuilding my splitter (new connectors this time; it's 40 years
>old), and noted 150 ohm resistors in series with the direct outs,
>both legs. There are also 150 ohm resistors in series on each leg of
>each iso out (two iso outs per channel).
>
>I'm wondering what these series resistors are for -- only thing I can
>think of is that any output -- direct or iso -- can go into a dead
>short and all the other outputs will continue to work (though it
>seems a little odd that the isos have these).

What transformers are in use?

These may have been added so that the microphone sees about the same impedance
through the splitter as without the splitter, because the transformers are
too low ratio to use properly for the application.

They might also have been added because the transformers are touchy about
loading.

If you have high ratio transformers that are comparatively immune to loading,
there's no reason to have any of this.

Also, as you point out, it means when something goes horribly wrong at the
truck and everything is shorted, the FOH guys don't lose their feed.

For a general discussion, permit me to recommend Jensen application note #5.

Before removing them, apply a 1kc square wave through a 500 ohm resistor to
simulate an SM-57, plug preamps into the direct and indirect outputs, and
look at the signals on each output with a scope. If you short the resistors
does it ring more or less?
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Phil Allison[_4_]]

wrote:

>
>
> > The "isolation" is really only at DC voltages, for phantom power
> > and not the actual signal. Achieving both requires an active
> > splitter and multiple transformers.
> >
>
>
> doesn't the multiple winding audio transformer isolation also
> prevent ground loops.

** Nope.

Many splitter boxes have ground lift switches on XLR pin 1s for that.


> If the device was just for isolation of DC phantom power,
> that could be accomplished simply using DC blocking caps
> and no need for an expensive transformer


** Correct, but using DC blocking caps creates possible surge current issues that transformers do not.



.... Phil

[Frank Stearns]

(Scott Dorsey) writes:

>Frank Stearns > wrote:
>>I'm rebuilding my splitter (new connectors this time; it's 40 years
>>old), and noted 150 ohm resistors in series with the direct outs,
>>both legs. There are also 150 ohm resistors in series on each leg of
>>each iso out (two iso outs per channel).
>>
>>I'm wondering what these series resistors are for -- only thing I can
>>think of is that any output -- direct or iso -- can go into a dead
>>short and all the other outputs will continue to work (though it
>>seems a little odd that the isos have these).

>What transformers are in use?

Good question -- they're supposedly early Jensens (before Jensen was Jensen). What's
in there is unmarked, nothing written on the mu cans at all, but the profile is
*exactly* like the PC-mount two-iso-winding Jensens, same pin-out (model # escapes
me, but I looked this up some time ago, when I first got this thing -- for a song --
used and very abused from a Seattle radio station).

I've actually taken an iso out for recording a time or two when I worked with a PA
company who did not know me and they demanded the directs. I found the iso sound was
very good, and other PA folks have never complained when they took the iso side. So,
these transformers might in fact be a Jensen design.

It's a 10-channel Unisync, with many serious build flaws, but I've gotten around
those with a little rework. (Wouldn't mind finding another one of these to refurb,
but they seem to be impossible to find. Most are probably now in land fills. But I
was pleased to have saved well north of a couple grand had I gone out and bought
something new with Jensens in it.)

>These may have been added so that the microphone sees about the same impedance
>through the splitter as without the splitter, because the transformers are
>too low ratio to use properly for the application.
>
>They might also have been added because the transformers are touchy about
>loading.

Could very well be.

>If you have high ratio transformers that are comparatively immune to loading,
>there's no reason to have any of this.

>Also, as you point out, it means when something goes horribly wrong at the
>truck and everything is shorted, the FOH guys don't lose their feed.

>For a general discussion, permit me to recommend Jensen application note #5.

>Before removing them, apply a 1kc square wave through a 500 ohm resistor to
>simulate an SM-57, plug preamps into the direct and indirect outputs, and
>look at the signals on each output with a scope. If you short the resistors
>does it ring more or less?

Good idea, will do. Thanks.

Frank
Mobile Audio
--

[Phil Allison[_4_]]

wrote:

> > >
> > >
> > > > The "isolation" is really only at DC voltages, for phantom power
> > > > and not the actual signal. Achieving both requires an active
> > > > splitter and multiple transformers.
> > > >
> > >
> > >
> > > doesn't the multiple winding audio transformer isolation also
> > > prevent ground loops.
> >
> > ** Nope.
> >
> > Many splitter boxes have ground lift switches on XLR pin 1s for that.
> >
> >
>
> But they wouldn't be able to provide that feature if it were not for
> the transformer providing isolation for pins 2 and 3 ????


** The signal wires from a low Z, balanced mic are already isolated from ground - normally coming direct from the voice coil or in some cases via an impedance matching transformer. Pin 1 merely provides a ground for the metalwork.

Perfectly possible to connect one mic to several balanced inputs in parallel. Ground lift may be needed to break hum loops between various units.

Using using a splitter transformer ensures that phantom supply voltages will not transfer between outputs and input nor any common mode hum voltage.


.... Phil

[Mike Rivers[_2_]]

>>> Many splitter boxes have ground lift switches on XLR pin 1s for
>>> that.
>>>>>
>>>>>
>>>
>>> But they wouldn't be able to provide that feature if it were not
>>> for the transformer providing isolation for pins 2 and 3 ????

The "ground lift" switch only isolates pin 1 on a connector from pin 1
on another connector. Since in most devices enclosed in a metal box
(including mic splitters), all pins 1 are connected together through the
box unless they are "ground lifted." Pin 1 is normally connected to the
cable shield. When you "lift" a ground, what you're doing is preventing
what's connected to pin 1 on the other end of the cable from being
connected to whatever's on the other end of a cable plugged into another
connector on the splitter.

If the cable goes to a microphone, pin 1 won't go to the brown dirty
stuff under the house. But if it goes to a mic preamp, most likely it
will end up being connected to earth, or at least, to system ground.
That's what makes (or breaks) ground loops.

On 12/3/2015 10:40 PM, Phil Allison wrote:
> ** The signal wires from a low Z, balanced mic are already isolated
> from ground - normally coming direct from the voice coil or in some
> cases via an impedance matching transformer. Pin 1 merely provides a
> ground for the metalwork.

In the case of a singled ended balanced output (and some microphones
have this configuration), pin 3 (unless you're arguing with Scott Dorsey
who believes that Pin 3 is hot) is connected to ground (the same
reference point as the output voltage) through a resistor equal to the
source impedance feeding Pin 2. So this configuration may not be
completely isolated from ground.


--
For a good time, visit http://mikeriversaudio.wordpress.com

[[email protected]]

On Thursday, December 3, 2015 at 10:40:29 PM UTC-5, Phil Allison wrote:
> wrote:
>
> > > >
> > > >
> > > > > The "isolation" is really only at DC voltages, for phantom power
> > > > > and not the actual signal. Achieving both requires an active
> > > > > splitter and multiple transformers.
> > > > >
> > > >
> > > >
> > > > doesn't the multiple winding audio transformer isolation also
> > > > prevent ground loops.
> > >
> > > ** Nope.
> > >
> > > Many splitter boxes have ground lift switches on XLR pin 1s for that.
> > >
> > >
> >
> > But they wouldn't be able to provide that feature if it were not for
> > the transformer providing isolation for pins 2 and 3 ????
>
>
> ** The signal wires from a low Z, balanced mic are already isolated from ground - normally coming direct from the voice coil or in some cases via an impedance matching transformer. Pin 1 merely provides a ground for the metalwork.
>
> Perfectly possible to connect one mic to several balanced inputs in parallel. Ground lift may be needed to break hum loops between various units.
>
> Using using a splitter transformer ensures that phantom supply voltages will not transfer between outputs and input nor any common mode hum voltage.
>
>
> ... Phil

I agree with you that the mic source is inherently balanced and isolated.
And there is no issue when one mic is connected to one console.

But when you connect one mic to two consoles without an isolator, the two grounds of the two __consoles___ can get tied together and form a loop.

It can be a big loop if one console is in a remote truck for example.

I thought that was the reason for using an isolated mic splitter and why a multi winding transfomer is used to provide galvanic isolation between the two consoles.

It also isolates the phantom.


Mark

[Mike Rivers[_2_]]

On 12/4/2015 8:57 AM, wrote:
> I agree with you that the mic source is inherently balanced and
> isolated. And there is no issue when one mic is connected to one
> console.
>
> But when you connect one mic to two consoles without an isolator, the
> two grounds of the two __consoles___ can get tied together and form
> a loop.

This is exactly the reason for the "ground lift" switch. As long as you
have a balanced signal you don't need the shield connected to chassis
ground, it's just a good idea to do so if it doesn't cause other problems.


--
For a good time, visit http://mikeriversaudio.wordpress.com

[JackA]

On Friday, December 4, 2015 at 8:57:18 AM UTC-5, wrote:
> On Thursday, December 3, 2015 at 10:40:29 PM UTC-5, Phil Allison wrote:
> > wrote:
> >
> > > > >
> > > > >
> > > > > > The "isolation" is really only at DC voltages, for phantom power
> > > > > > and not the actual signal. Achieving both requires an active
> > > > > > splitter and multiple transformers.
> > > > > >
> > > > >
> > > > >
> > > > > doesn't the multiple winding audio transformer isolation also
> > > > > prevent ground loops.
> > > >
> > > > ** Nope.
> > > >
> > > > Many splitter boxes have ground lift switches on XLR pin 1s for that.
> > > >
> > > >
> > >
> > > But they wouldn't be able to provide that feature if it were not for
> > > the transformer providing isolation for pins 2 and 3 ????
> >
> >
> > ** The signal wires from a low Z, balanced mic are already isolated from ground - normally coming direct from the voice coil or in some cases via an impedance matching transformer. Pin 1 merely provides a ground for the metalwork.
> >
> > Perfectly possible to connect one mic to several balanced inputs in parallel. Ground lift may be needed to break hum loops between various units.
> >
> > Using using a splitter transformer ensures that phantom supply voltages will not transfer between outputs and input nor any common mode hum voltage.
> >
> >
> > ... Phil
>
> I agree with you that the mic source is inherently balanced and isolated.
> And there is no issue when one mic is connected to one console.
>
> But when you connect one mic to two consoles without an isolator, the two grounds of the two __consoles___ can get tied together and form a loop.
>
> It can be a big loop if one console is in a remote truck for example.
>
> I thought that was the reason for using an isolated mic splitter and why a multi winding transfomer is used to provide galvanic isolation between the two consoles.
>
> It also isolates the phantom.
>
>
> Mark

GALVANIC?

Jack

[Scott Dorsey]

> wrote:
>
>I agree with you that the mic source is inherently balanced and isolated.
>And there is no issue when one mic is connected to one console.
>
>But when you connect one mic to two consoles without an isolator, the two grounds of the two __consoles___ can get tied together and form a loop.

Yes, this is bad news.

>It can be a big loop if one console is in a remote truck for example.
>
> I thought that was the reason for using an isolated mic splitter and why a multi winding transfomer is used to provide galvanic isolation between the two consoles.

The most common way this is done is with a high-Z to low-Z transformer
bridged across the microphone... so the microphone ground is connected to
the ground of the direct output... and the isolated output that comes off
the transformer has its own ground. The guy with the direct output supplies
the phantom.

The problem with this is that people fight over who gets the direct and who
gets the iso. It is a political issue.

One way around that is to have a 1:1:1 transformer with two secondaries so
_everybody_ gets an isolated output, but that then means a phantom supply
has to be integrated into the splitter.

>It also isolates the phantom.

This is both a good and a bad thing, politically. I have had the PA guys
throw a fit at me because I had phantom turned on in the truck for a given
mike or didn't have it turned on for a given mike. If I am recording fewer
channels than the PA guy has, I may be repatching between sets and so my
numbering changes and the channels I need to have phantom on change too.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Phil Allison[_4_]]

wrote:

>
> >
> > > > >
> > > > > doesn't the multiple winding audio transformer isolation also
> > > > > prevent ground loops.
> > > >
> > > > ** Nope.
> > > >
> > > > Many splitter boxes have ground lift switches on XLR pin 1s for that.
> > > >
> > > >
> > >
> > > But they wouldn't be able to provide that feature if it were not for
> > > the transformer providing isolation for pins 2 and 3 ????
> >
> >
> > ** The signal wires from a low Z, balanced mic are already isolated from ground - normally coming direct from the voice coil or in some cases via an impedance matching transformer. Pin 1 merely provides a ground for the metalwork.
> >
> > Perfectly possible to connect one mic to several balanced inputs in parallel. Ground lift may be needed to break hum loops between various units.
> >
> > Using using a splitter transformer ensures that phantom supply voltages will not transfer between outputs and input nor any common mode hum voltage.
> >
> >
> >
>
>
> I agree with you that the mic source is inherently balanced and
> isolated. And there is no issue when one mic is connected to one
> console.
>
> But when you connect one mic to two consoles without an isolator,
> the two grounds of the two __consoles___ can get tied together
> and form a loop.

** A balanced mic can feed two consoles, be grounded at the first one and link only pins 2 & 3 to the other. Usually works just fine.


> It can be a big loop if one console is in a remote truck for example.

** In a case like that, the use of splitting transformers is probably essential. The two consoles may be on different AC power systems and operated by separate organisations so the chances of a problem are much higher.


.... Phil

[Scott Dorsey]

Phil Allison > wrote:
>
>** In a case like that, the use of splitting transformers is probably essential. The two consoles may be on different AC power systems and operated by separate organisations so the chances of a problem are much higher.

65 VAC measured between the truck ground and the stage ground at the 9:30
club at one gig back in the nineties, using a Simpson 260.

Needless to say the building has since been torn down for numerous fire
code violations.
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Mike Rivers[_2_]]

On 12/4/2015 7:36 PM, Phil Allison wrote:

> ** A balanced mic can feed two consoles, be grounded at the first
> one and link only pins 2 & 3 to the other. Usually works just fine.

That's the classic "ground lift." Of course if neither console had a
"Pin 1 Problem" you wouldn't have to lift grounds at all, because there
wouldn't be any signal on the ground of the source end, and if there
was, it wouldn't go anywhere harmful on the destination end. But
nothing's perfeck.

--
For a good time, visit http://mikeriversaudio.wordpress.com