GSSL input and output transformers question...

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

finbase

Well-known member
Joined
Jan 6, 2014
Messages
47
Location
Espoo, Finland
Hi folks!

I'm going to build two gssl units, one "regular" (since I'm still a n00b I want to do that first...) and one with input and output transformers, for "colour".

I've done a lot of search in this forum, and I've found a couple very interesting mods for both inputs and outputs. Nevertheless, the projects I'm talking about are different, so I'm still a bit hesitant...

The input section would derive from the Jensen mod schematic I found, but the actual transformer would be Cinemag CMLI-15/15B (interchangeable with the JT-11P-1). The output section would derive from the mod I found from Luny Tune's passive EQ + GSSL thread. The transformer I'm going to use for the output is CMOQ-2S.

Please see the attached pictures for more detailed "insight" into the modifications. (Output mod attached to the following message.)

The question is: If I do the described modifications in the both ends of the GSSL schematic and use the CMLI-15/15B and CMOQ-2S transformers, will the unit perform properly? Or should I be aware of some further problems?

finbase
 

Attachments

  • SSLJensen_in_outputmod-1.jpg
    SSLJensen_in_outputmod-1.jpg
    54.8 KB · Views: 152
To add some "color" and still retain circuit integrity, may I suggest an Ampex -02 matching transformer for the input, turned around for voltage loss, API 2520's (or similar) for the gain blocks, and a step-up output transfomer.  You do not need the gain inherent in the output iron but it adds a "bigness" to the sonics.  Though I have drawn in  a 2503, any vintage 1:2 or 1:3 should suffice, including WE 111c, UTC LS-30, Triad HS-56.
The vca in the SSL compressor is current output and requires an inverting op-amp.
 

Attachments

  • in_out.jpg
    in_out.jpg
    36.7 KB · Views: 119
gridcurrent said:
The vca in the SSL compressor is current output and requires an inverting op-amp.

That's actually one of the reasons why I'm a bit hesitant of doing the mod I described in the first post. The "Jensen Input mod" actually makes the first op-amps (pardon the unprofessional term) non-inverting. Although I'm still quite a novice as far as the electronics are concerned, it stirred up my fears... :)

 
So here's the original schematic with the intended mods, drawn with Microsoft Paint... So my apologies for the lack of artistic impression. For example, the output transformer is drawn poorly, since I'm not 100% sure of the ultimate wirings.

The left "path" is drawn with the possible mods, the right one is as the original.

Here's the "highlited" notes:

1.) Input transformer is Cinemag CMLI-15/15B (Jensen JT-11P-1 equivalent).
-1 : 1 Winding Ratio;
-Input impedance 18k with 15K load;
-Maximum input level +20 dBu at 20 Hz;
-Voltage Gain -2.2 dB
-Phase Shift at 20 kHz -5 degrees

2.) Due to Jensen's suggestion the op-amp is non-inverting...

3.) The reduced output resistors, due to the output transformer

4.) Output transformer is Cinemag CMOQ-2S
-Quadfilar Windings, ratio 1 : 1
-Input impedance 680ohm
-Maximum input level +22 dBu at 20 Hz;
-Voltage Gain approx -2.78 dB
-Phase Shift at 20 kHz -5 degrees

So the question still is, would this circuit work? Or have I fallen into an asinine reverie?
 

Attachments

  • ssl_schematic_WithMods.jpg
    ssl_schematic_WithMods.jpg
    813.7 KB · Views: 103
Maybe use my updated schematic for the latest pcb revision as shown in http://groupdiy.com/index.php?topic=47.msg720178#msg720178
You might add some decoupling caps close to line receivers NE5534 supply pins, add anti-latchup diodes and use 100pF caps instead of silk screen error 0.1uF, located between 1K placeholders in DBX202C substitution circuits.
Your implementation should work (too lazy to google your type of transformers datasheet, but left input - will not be hard tied to 0V reference voltage).
 
What you want to do will work, one way or another.  But it will gain you nothing as far as color. The Jensen mod was meant to be a technical improvement over the 5534 line receiver, mainly better CMRR.  A modern 1:1 transformer should have no color at all.  Same thing on the CMOQ-2S. As CJ once said, the 2503 (api's version of CMOQ2S) is flat "from DC to daylight".  Obviously he's exaggerating, but it makes a point.

Consider starting with some Edcor trafos, til you get the circuit working.  If you want color I'd go 10K:600 on the input and 600:10K on the output.  The voltage gain, combined with Edcor's winding methods may give you all the color you need.


 
Hi folks!

I'm going to build two gssl units, one "regular" (since I'm still a n00b I want to do that first...) and one with input and output transformers, for "colour".

I've done a lot of search in this forum, and I've found a couple very interesting mods for both inputs and outputs. Nevertheless, the projects I'm talking about are different, so I'm still a bit hesitant...

The input section would derive from the Jensen mod schematic I found, but the actual transformer would be Cinemag CMLI-15/15B (interchangeable with the JT-11P-1). The output section would derive from the mod I found from Luny Tune's passive EQ + GSSL thread. The transformer I'm going to use for the output is CMOQ-2S.

Please see the attached pictures for more detailed "insight" into the modifications. (Output mod attached to the following message.)

The question is: If I do the described modifications in the both ends of the GSSL schematic and use the CMLI-15/15B and CMOQ-2S transformers, will the unit perform properly? Or should I be aware of some further problems?

finbase
Hi,
had any outcome with the mod?
any info, doc (connections, parts to add-remove,etc..) about, to share ?
thanks.
 
I did once, put in some telecom line-repeater transformers "repeater coils", sounded in a way that I decided to not bother with spending a switch for bypassing them
 
Back
Top