I picked up an early-1980s 10-channel mic splitter (one direct, two iso
outs), a Uni-sync "Trouper Series" model MS1003. Transformers look like
Jensens, but the visible portions of the cans are unmarked so who knows
for sure. (Can't see the tops of the cans; I'd nearly have to destroy the
unit to get to them.)

There is a peculiar resistor usage. Each output, *including* the supposed
direct out, has 75 ohm resistors in series (one on each side of the line).
On the direct side, the transformer pick off is before the resistors.

What are these supposed to do?

The ground lifts are "hard" lifts; perhaps the series 51 ohm resistor/.1
cap to ground should be added to the float side.

It does appear that per Bill Whitlock's paper on splitters that channel 1
ties the XLR chassis ground to XLR pin 1 on channel 1 while channels 2-10
float the XLR chassis ground, and pin 1 is simply fed through.

Also per that paper, perhaps I should add the 2.3K damping resistors
across the iso outs.

I drew blanks on various web-searches on this critter. Thanks in advance
for any advice,

Frank Stearns
Mobile Audio
--

"Jim Gregory" writes:

"Frank Stearns" wrote in message
...
I picked up an early-1980s 10-channel mic splitter (one direct, two iso
outs), a Uni-sync "Trouper Series" model MS1003. Transformers look like
Jensens, but the visible portions of the cans are unmarked so who knows
for sure. (Can't see the tops of the cans; I'd nearly have to destroy the
unit to get to them.)

There is a peculiar resistor usage. Each output, *including* the supposed
direct out, has 75 ohm resistors in series (one on each side of the line).
On the direct side, the transformer pick off is before the resistors.

snipped myself...

Is there a centre-tapped phantom +ve path from only one direct sec to the
pri ct?
The secs may have the 2 x 75ohm line resistors added to approach matching of
load - and to be less onerous on the tiny pri voltage.

Hi Jim -

Not sure I'm parsing what you're asking. This is a totally passive device,
so there is no local phantom supply with the unit.

There's no center tap that I can see. Looking "bottoms up" at the
transformers (PC board mount), I see two rows of four connections. On one
row of four, the outer most connections (call them 1 and 4) receive the
lines from pins 2 and 3 of the input XLR. The inner connections, call them
2 and 3, are tied together and to XLR pin 1. I'm assuming this is
transform housing ground and an internal transformer ground. (This seems
to be the case -- continuity exists here to the transformer casing but
"infinite" resistance to all other pins on the transformer.)

The second row of four, call them connections 5-8, are wired like this:
5 & 7 go to pins 2 & 3 of the first ISO output XLR, while 6 & 8 go to pins
2 & 3 of the second ISO out XLR.

Does that provide any additional useful data?

I thought perhaps the 75 ohms might be providing some sort of bridging,
but it seems like they might cause more problems? I dunno. Perhaps with
modern higher-Z preamps these resistors aren't that important...?

Thanks for taking the time to reply,

Frank Stearns
Mobile Audio


--

"Frank Stearns" wrote in message
...
"Jim Gregory" writes:

"Frank Stearns" wrote in message
...
I picked up an early-1980s 10-channel mic splitter (one direct, two iso
outs), a Uni-sync "Trouper Series" model MS1003. Transformers look like
Jensens, but the visible portions of the cans are unmarked so who knows
for sure. (Can't see the tops of the cans; I'd nearly have to destroy
the
unit to get to them.)

There is a peculiar resistor usage. Each output, *including* the
supposed
direct out, has 75 ohm resistors in series (one on each side of the
line).
On the direct side, the transformer pick off is before the resistors.

snipped myself...

Is there a centre-tapped phantom +ve path from only one direct sec to the
pri ct?
The secs may have the 2 x 75ohm line resistors added to approach matching
of
load - and to be less onerous on the tiny pri voltage.

Hi Jim -

Not sure I'm parsing what you're asking. This is a totally passive device,
so there is no local phantom supply with the unit.

There's no center tap that I can see. Looking "bottoms up" at the
transformers (PC board mount), I see two rows of four connections. On one
row of four, the outer most connections (call them 1 and 4) receive the
lines from pins 2 and 3 of the input XLR. The inner connections, call them
2 and 3, are tied together and to XLR pin 1. I'm assuming this is
transform housing ground and an internal transformer ground. (This seems
to be the case -- continuity exists here to the transformer casing but
"infinite" resistance to all other pins on the transformer.)

The second row of four, call them connections 5-8, are wired like this:
5 & 7 go to pins 2 & 3 of the first ISO output XLR, while 6 & 8 go to pins
2 & 3 of the second ISO out XLR.

Does that provide any additional useful data?

I thought perhaps the 75 ohms might be providing some sort of bridging,
but it seems like they might cause more problems? I dunno. Perhaps with
modern higher-Z preamps these resistors aren't that important...?

Thanks for taking the time to reply,

Frank Stearns
Mobile Audio


--
.
I was trying to suggest....
If you did find a middle point of a winding on its own tag, and ditto for
another separate winding, then, even if turns ratios are different, if those
2 ct midway points are linked, you can advance the phantom volts +ve sent
from a mixer to the sec onto the pri where it would nourish the capacitor
mic, assuming -ve returns go to pin 1s on m & f XLRs concerned. But maybe no
ct here as you reported.
The 4 x 75 ohms are increasing reflected z protection, I think
Jim

Frank Stearns wrote:
I picked up an early-1980s 10-channel mic splitter (one direct, two iso
outs), a Uni-sync "Trouper Series" model MS1003. Transformers look like
Jensens, but the visible portions of the cans are unmarked so who knows
for sure. (Can't see the tops of the cans; I'd nearly have to destroy the
unit to get to them.)

There is a peculiar resistor usage. Each output, *including* the supposed
direct out, has 75 ohm resistors in series (one on each side of the line).
On the direct side, the transformer pick off is before the resistors.

What are these supposed to do?

I remember seeing these things many years ago, and I was told that they
were for RF rejection. That sort of thing CAN be a useful strategy for
improving stability on very-high-Z circuits but I can't imagine it would
really be very useful for a low-Z mike application.

A number of companies made transformers basically shaped like that. The
basic can design came from UTC originally, I think, and guys like Jensen,
Cine-Mag, Reichenbach and so forth tended to adapt it.

It does appear that per Bill Whitlock's paper on splitters that channel 1
ties the XLR chassis ground to XLR pin 1 on channel 1 while channels 2-10
float the XLR chassis ground, and pin 1 is simply fed through.

Also per that paper, perhaps I should add the 2.3K damping resistors
across the iso outs.

Depends on the transformers. It's possible an RC constant would be better,
it's possible nothing would be. Run a 1 KC square wave through it, see
how the sequre wave looks. If adding a shunt resistor or RC network
reduces the ringing, use it.
--scott


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

(Scott Dorsey) writes:

I picked up an early-1980s 10-channel mic splitter (one direct, two iso
outs), a Uni-sync "Trouper Series" model MS1003. Transformers look like
Jensens, but the visible portions of the cans are unmarked so who knows
for sure. (Can't see the tops of the cans; I'd nearly have to destroy the
unit to get to them.)

-snippage-

Also per that paper, perhaps I should add the 2.3K damping resistors
across the iso outs.

Depends on the transformers. It's possible an RC constant would be better,
it's possible nothing would be. Run a 1 KC square wave through it, see
how the sequre wave looks. If adding a shunt resistor or RC network
reduces the ringing, use it.

Thanks, Scott.

I did in fact run square waves and was pleased to see virtually no ring at
all and very good HF response (maybe the xformer cans are hiding hard Y
connections???). The ring spike that did appear was extremely narrow and
wasn't a series of jiggles like some of the junk box xformers I've got.

There was a bit of LF loss, wild guess of about 1-2 dB below 50 hz. Adding
the shunt lost a small amount of level but lessened the LF loss.

Used it for a gig last week and the sonic results seemed fine but with
nothing to compare it to it's hard to tell for sure. But, at $10/channel
for that level of performance I'm not going to complain!!

Frank Stearns
Mobile Audio

--