sowter 3575

[Matt Syson]

Hi
Strictly you should be using a 600 Ohm version as a 'send' transformer (close to the desk output) and 10K as a 'receive' as the higher impedance will suffer greater HF signal loss if it then has a load of cable attached to it. There is also a greater chance of it 'ringing' in the audio band.
A 600 Ohm type would be less 'fussy' about what is at the other end of the cable.
If your cables are only a few metres or so you can probably get away with either.
Matt S

[gouge]

thanks for your time matt.

i've got some jensen output xformers that i can use for the sends and will probably source some api's as well.

will take your advice and keep the sowters for the returns only.

cheers.

Originally Posted by Matt Syson

Hi
Strictly you should be using a 600 Ohm version as a 'send' transformer (close to the desk output) and 10K as a 'receive' as the higher impedance will suffer greater HF signal loss if it then has a load of cable attached to it. There is also a greater chance of it 'ringing' in the audio band.
A 600 Ohm type would be less 'fussy' about what is at the other end of the cable.
If your cables are only a few metres or so you can probably get away with either.
Matt S

[Matt Syson]

Hi
If you already have the Sowters you could simply try one. As most 'modern' gear has a high input impedance the send impedance is not as critical as it was when you may have to 'drive' 600 Ohms.
I can't remember what the specified 'load' resistance is for these particular transformers but you should check and maybe add extra to suit the loading of the return input resistance. It is probably about right anyway.
Matt S

[gouge]

hi matt,

i don't have the sowters as yet. i'm not sure what the specified load resistance is?

this is the info from the datasheet.

General Purpose Bridging Transformer
For reduction of most kinds of noise caused when equipments are connected. For use where it is important not to load the driving circuit. Ideal as an input transformer for a valve amplifier to provide a balanced, isolated line input which will then accept balanced or unbalanced sources. Unbalanced to balanced conversion and vice versa. For helping to ensure Electromagnetic Compatibility (EMC) compliance.
High impedance presents minimal load to source even at 15 Hz. Mumetal core for minimal harmonic distortion. Internal electrostatic shield for high common mode noise rejection. Mumetal can for magnetic shielding. Secondary centre tap for phase splitting etc.
RATIO 1:1ct
MAXIMUM LEVEL 0.5 % THD at 50 Hz (40 ohm source) +26 dBu
DISTORTION at 50 Hz (Source 40 ohms) 0.05% at +20 dBu
INSERTION LOSS 10k ohm source 10k ohm load 1.08 dBu
INPUT/OUTPUT BALANCE >60 dB
FREQUENCY RESPONSE 6k4 ohm source 10k ohm load +/- 0.25 dB 15 Hz to 60 kHz


should i implement them as per the jensen AS089 schematic and use a damping network on the unbalanced outputs?

[Matt Syson]

Hi
Look at the spec for the Ghost's inputs and if they are 10K you are fine as it is. If it is a lot more (20K or above) then perhaps put a resistor in parallel with the (ghost) input to bring it nearer to 10K. It does not need to be 'exact' as it will vary depending on the gear that is sending the signals to a small extent.
Having to consider the loading on a transformer is one of the downsides to what is otherwise a most useful 'component'.
Matt S

[gouge]

thanks again matt,

i'm going to have to guess. i had a look in the manual and it calls up a 2k impedance on the insert sends but doesn't reference the return.

my guess is 15k as the line inputs on the channels are 15k.

[Matt Syson]

Hi
15K is probably OK but you could try the effect of a 30K resistor 'for fun'.
You may not actually notice any difference.
Matt S

[gouge]

great stuff matt,

will try a few different values. "for fun"

we are doing it primarily to get some more transformers into the mixdown path and to balance the groups inserts to simplify the patching as 99% of the outboard gear is balanced.

the thing i'm most looking forward to is trying the different transformers on the sends to see what the subtle differences are with the sound.

[gouge]

i was hoping to take this a little further and today threw together a couple of schematics. based on the fact that i really don't know what i'm doing i was hoping to ask a few further questions.

the 2 top diagrams are the insert send boxes. top one is 1:1 output and the middle one is 1:2 output.
bottom diagram is the return box. i cobbled the diagrams together from here and the jensen schematics.
i'm not sure i got the transformer wiring correct either.

i guess my first question, (if anyone is keen to comment) is did i get it totally wrong?

i was also curious about the grounding on the insert send diagrams.
the other question i had is regards the 680p cap on the insert return diagram. i got that directly from the jensen website. is it there to filter out hf noise from the parallel resistor or the transformer or both or something completely different.

any thought or comments very much appreciated.

[Matt Syson]

Hi
Most of them are more complicated than you need, unless you want the ability to have +4 or -10 levels available on a switch.
Go with the bottom diagram, it is all you need.
The resistor and 680pF cap are there to 'tame' the very high frequencies on that particular transformer. Sowters don't need it if you are working into around 10K resistance.
The bottom circuit has a 'fround lift' switch, you may not even need that. Put it in a metal box which should be grounded obviously but 'float' all wiring except the pin 1 of XLRs. Use insulated jacks or use insulation so the 'sleeve' doesn't get grounded to the box.
It's all 'passive' so getting it wrong won't kill anything, it just won't work!
Matt S

[Mixermend]

On the bottom drawing the 51 ohm earth lift resistor should have the small value capacitor in parallel, not in series - as shown. The function of the capacitor is simply to provide a low impedance at high frequencies.

[gouge]

thanks again guys,

will tweak those items.

the pad on the output diagrams is there only to allow us to drive the transformer harder. initially i'll just hook up 2 transformers and listen to it. then add stuff if/when needed.

[Matt Syson]

Hi
The first two connection diagrams show the switch in a place that still allows the transformer to be fed by the 'send'. Even though you have 'bypassed' the transformer it is still being fed signal so will affect it, although less than when you are using it 'signal through it'.
You should rearrange the switch so it is completely disconnected if that is your aim.
Matt S

[gouge]

thanks matt,

i wondered if that was also the case with the pad. it's not completely removed from the circuit even when switched out.

not sure i know what the answer is?

[Matt Syson]

Hi
Yes the pad is also 'in' so adding to the load on the transformer. You should really rearrange the switching completely so transformer / pad are completely disconnected unless you wamt it, and alter pad impedances to suit.
Matt S

[gouge]

is the answer using 3pdt switching instead?

[Matt Syson]

Hi
Actually the pad switching is fine although the resistances may benefit from a change. Try it as it is for now but it is loading the transformer quite hard.
A 3 pole switch is a 'simple' way to disconnect the transformer but if you don't mind a but od capacitive 'unbalance' I think it can be managed with a 2 pole switch.
Matt S

[gouge]

thanks matt,

i've had a long think about the first switch and decided to ditch it.
i think you're right, it just adds an extra complication.

i've convinced myself that it's better to just have the trannies permanently in the path. if they don't do the job then they shouldn't be there so the switch is pointless.