SSLtech
Well-known member
Rochey and I have been working on a couple of developments for the GSSL, one of which -while not yet ready for announcement (a couple more weeks, we expect)- should significantly improve the performance. -As with the 'turbo', it's designed to be easily retrofittable to any existing GSSL, or incorporated into a new build.
I've been taking a long hard look at the GSSL's noise and distortion behaviour, and one annoying quirk of the GSSL has been hampering my measurements; You'd expect the GSSL to be at it's quietest when switched to 'bypass', and to distort the least when switched to 'bypass' (in reality it doesn't actually 'bypass' anything -it's really a "compression enable/disable" switch)... -But this is not what happens at all. -In actual fact the HIGHEST distortion figures for the GSSL are measured when the compression is DISABLED by the front panel switch.
But I've discovered what's happening, and how to make it stop. -And you can do it for FREE!!! -All it takes is popping the top, swapping a couple of wires and adding a third wire.
One of the two halves of the bypass switch (the part which only has TWO wires going to it) requires a slight modification.
Referring to this pdf:
On the component overlay (second page) two wires going to the "bypass switch" are marked "Defeat makeup when out". One connects to Pole 'A', and the second connects to contact A2.
Firstly, REVERSE these two wires. -If you (as I've done) noticed that it's just a simple "make" connection and possibly just connected them in no particular order, you need to make sure that the wire at the board end which is physocally CLOSEST to the 750K resistorneeds to be connected to the POLE (common contact) of the switch, and the other wire to the second contact.
Now, finally solder a wire to GROUND from the other contact "A1". -There's already a handy ground available at contact 1 of pole 'C' if you need it... ("C1" on the rotary switch as shown).
That's it.
Distortion in bypass is now a thing of the past. -I know that most people don't seem to care about distortion in bypass so long as the performance is unaffected while the unit is switched 'in', but I get distortion readings up to 0.755% worst case (when the unit is in 'bypass', ratio 10:1, threshold set to max) which drops to 0.007% on the unit currently on the bench with this modification.
It only takes a minute to do, and it fixes something which I've always found slightly bothersome, but which has now made my life MUCH easier.
Enjoy!
-Keith
I've been taking a long hard look at the GSSL's noise and distortion behaviour, and one annoying quirk of the GSSL has been hampering my measurements; You'd expect the GSSL to be at it's quietest when switched to 'bypass', and to distort the least when switched to 'bypass' (in reality it doesn't actually 'bypass' anything -it's really a "compression enable/disable" switch)... -But this is not what happens at all. -In actual fact the HIGHEST distortion figures for the GSSL are measured when the compression is DISABLED by the front panel switch.
But I've discovered what's happening, and how to make it stop. -And you can do it for FREE!!! -All it takes is popping the top, swapping a couple of wires and adding a third wire.
One of the two halves of the bypass switch (the part which only has TWO wires going to it) requires a slight modification.
Referring to this pdf:
On the component overlay (second page) two wires going to the "bypass switch" are marked "Defeat makeup when out". One connects to Pole 'A', and the second connects to contact A2.
Firstly, REVERSE these two wires. -If you (as I've done) noticed that it's just a simple "make" connection and possibly just connected them in no particular order, you need to make sure that the wire at the board end which is physocally CLOSEST to the 750K resistorneeds to be connected to the POLE (common contact) of the switch, and the other wire to the second contact.
Now, finally solder a wire to GROUND from the other contact "A1". -There's already a handy ground available at contact 1 of pole 'C' if you need it... ("C1" on the rotary switch as shown).
That's it.
Distortion in bypass is now a thing of the past. -I know that most people don't seem to care about distortion in bypass so long as the performance is unaffected while the unit is switched 'in', but I get distortion readings up to 0.755% worst case (when the unit is in 'bypass', ratio 10:1, threshold set to max) which drops to 0.007% on the unit currently on the bench with this modification.
It only takes a minute to do, and it fixes something which I've always found slightly bothersome, but which has now made my life MUCH easier.
Enjoy!
-Keith
