GSSL HELP THREAD!!!

[ruckus328]

OK, you guys just fried my brain lol.  I've been at it for 14 hours.  Need to call it quits and get some sleep.  You made me realize I totally screwed something up though (I think), I'm summing be L/R sidechains before the hpf.  I guess it would be better to do it:

HPF > Turbo switch > Sidechain ?

That way like you said, the turbo switch will dictate if it's a stereo or mono key input. 

Never did any hardware external inputs before now, just ITB, so I still don't quite have my hands wrapped around this yet, hopefully it will click in the morning.

 

[Harpo]

Hey Harpo what's keeping you up at 4:20 AM?

Probably the same as you @ 4:22 AM, drowning in work.

Oh what about a balanced CV insert with unbalanced option and TRS/XLR combo jack?
So you'd have really clean tracking. I can imagine it might matter there, if only little...

Why not. You could go crazy with adding functions until you run out of frontpanel space. (remember my little different scf at the german department some years ago? variable hpf, derived lpf, filter blend, some key-listen functions,..?)

 

[livingnote]

Haw yeah...so many ideas, 24h day time limit, 8h sleep necessary

You know, after doing the SC send thing myself - question is, why bother?

What you're doing is essentially sending a copy of the signal back out again, but
debalanced and then rebalanced. While it might provide a CM scrubbing unit, there
really isn't too much that speaks against just doing it as a straight through from
input, because once it's back out you can do whatever with it anyway...

What you have for turbo plus SC is something like:

IN L ->  Sidechain Filter 1 -> SC VCA 1 -> Detector 1 -> Timining Network -> CV to VCAs
                                                       ^
IN R ->  Sidechain Filter 2 -> SC VCA 2 -> Detector 2  |

And without turbo in you just have:

IN L ->  Sidechain Filter 1 -> SC VCA 1 -> Detector 1 -> Timining Network -> CV to VCAs
                             ^
IN R ->  Sidechain Filter 2  |

...which is owing to that the turbo is in there and you'd have to chuck the HPF for Turbo mode
like many do (and comment that it doesn't matter nuts), if you're going with mono HPF, which
would be the standard:

IN L ->  Sidechain Filter 1 -> SC VCA 1 -> Detector 1 -> Timining Network -> CV to VCAs
      ^
IN R  |

And then for Turbo mode:

IN L -> SC VCA 1 -> Detector 1 -> Timining Network -> CV to VCAs
                                ^
IN R -> SC VCA 2 -> Detector 2  |

 

[ruckus328]

But what about? :

For Mono s/c

IN L ->  Sidechain Filter 1 -> SC VCA 1 -> Detector 1 -> Timining Network -> CV to VCAs
      ^
IN R  |


And:


For Stereo s/c

IN L ->  Sidechain Filter 1 -> SC VCA 1 -> Detector 1 -> Timining Network -> CV to VCAs
     
IN R ->  Sidechain Filter 1 -> SC VCA 1 -> Detector 1 -> Timining Network -> CV to VCAs


Why wouldn't it work hooking it up like this?

PS: Sent you email.

 

[livingnote]

Well Mono S/C as you have it there is the classic Gyraf, and the stereo thing you have
is a stereo turbo with an extra timing network, thing is - why two? OTOH behavior
would be the same, the only merit being that you could now go full dual mono with it,
but with stereo you would have to sum the CVs before you hit the VCAs with them
anyway (SC VCAs and Sig VCAs), or your stereo image is gonna do za funny stuff.

In effect, the higher voltage charged cap in either timing net is gonna say where goes,
which in effect is the same as the higher-of-two function performed before the timing
caps. Redundant for stereo operation (unless I'm missing something).

Dual mono is another one of those critters. Like, why? There's so little change in the
workload for building two stereo ones or a dual mono one (the whole control thing
gets doubled, you get a relay mess and for mono it's gonna behave right anyway if
you just plug a mono signal into a stereo unit). A matter of practicality more than
anything, why put almost the same time and money into a dual mono thing when
for little more you could have two stereo ones?

(oh check it out I wrote "Timining", haw)

 

[tv]

Of course it -would- work, but in this case, you would have a "dual-mono" compressor, i.e. two independent compressors in one box.

You could "gang" the two independent channels -at-the-timing-capacitor- ("oxford" style), but then you would have double CV path doing "same thing" for two VCA channels, possibly introducing offsets/errors/pot-tracking errors etc.

In such case, you would have to get rid of the now redundant "second timing network" and "CV circuit" (edit AND feed BOTH VCAs from same CV source, to reduce errors.

The choice is of course yours. Much ado for nothing, if you ask me (but you didn't).

 

[ruckus328]

Drr....I'm a moron.  Sorry, I'm going on about 4 hours of sleep.  Hopefully tonight I can get some rest finally.  I screwed up on my diagram.  What I meant was putting the turbo switch before the high pass filter, (but NOT dual mono).

Like this:

Mono s/c

IN L ->  Sidechain Filter 1 -> SC VCA 1 -> Detector 1 -> Timining Network -> CV to VCAs
         ^                                          
IN R ->  


Stereo s/c

IN L ->  Sidechain Filter 1 -> SC VCA 1 -> Detector 1 -> Timining Network -> CV to VCAs
                                                                     ^
IN R ->  Sidechain Filter 2 -> SC VCA 2 -> Detector 2

Isn't that how dual ssc's are usually wired?

 

[livingnote]

Yep, there you go. Either sum before the hipass or thereafter for non-turbo,
and for turbo just sum CV's at point "C" as it would be.

 

[Biasrocks]

Dual mono is another one of those critters. Like, why? There's so little change in the
workload for building two stereo ones or a dual mono one

For one, you're costs increase significantly over a dual mono build.

You'll need an additional, case, power supply, internal mounting hardware & wiring looms, resistors,
ic's, etc, additional connectors, xlr, sidechain, power, etc. That easily adds another $175-$200 to a build.

Also additional space in the rack.

It's also very desirable to have the ability to compress two instruments using the same box.

This discussion really belongs in the redesign thread.

http://www.groupdiy.com/index.php?topic=38940.120

Mark

 

[livingnote]

Ah yeah, come to think of it it would really be cool to have something like
2 somethings through 1 compressor, I just keep seeing me wanting to set
everything separately. There's one for you, Mike, basically just design one
SSL x2 and then relay-etize at proper positions.

That said, let's get outta here, wrong thread indeed.

 

[tv]

This would make more sense if the whole circuit would be "dual mono" - in terms of power supply - as well ...

 

[livingnote]

I mean what would make a lot more sense in my world would be.... Hang on...

<runs off to his little Beast thread>

 

[MikeFFG]

Hey I'm figuring out this whole transformer wiring thing, think I got it, but what rating of fuse should I get?  That's assuming it is necessary to have a fuse, is it? Thanks

 

[mokkinger]

hi guys,
having a little problem here: I have a GSSL with turbo and dual Sidechain Board here, all wired up correctly, it worked fine for about 2 years now... since some weeks I have a white noise turning up after some minutes, sometimes right at startup, and it disappears again after some minutes or so... it´s only in the right channel. It can´t be the sidechain filter since it also appears when SC is turned off, and also appears in GSSL mode (without turbo board)... I guess it is a damaged capacitor, but I am not sure... how and where would you guys start searchin for the problem?
really would appreciate any suggestions how to find and fix this noise...
thanks & cheers,
mark 

 

[mokkinger]

hmm, okay. remove the That, easy. how do I bypass the VCA section? could you point me to the right direction please?
thanks!

 

[littleant]

Ugh, okay please don't laugh...or do...it's pretty bad:

I soldered my opamps directly to the pcb.  It was one of those days...

So in leu of this:

Would it be better to leave them on?  Will they still work properly?

Or would it be better to try to de-solder them?  Will I damage them by doing this?

Any advice would be greatly appreciated, bearing in mind I already feel like a prize idiot

 

[mac]

as long as they are pin correct - should be fine..

i only use sockets so I can swap and change them - incase of failures or I want to try different compatible opamps..


Mac

 

[littleant]

Thank you Mac!

 

[Alessio]

Hi everybody,
I've finished my first gssl compressor but i have some problems with the voltages.
I'm testing the voltages without the components only on the sockets.
TL074
pin 1: +2,8V
pin2: +2,8V
pin3: 0
pin4: +12,17V
pin5: 0
pin6: +2,6V
pin7: +2,7V
pin8: +3V
pin9: +2,6V
pin10: 0
pin11: -12,21
pin12: 0
pin13: 0
pin14: 0

I've never seen something like this in this forum and are weeks that I'm looking...
I've also wrong voltages on the TL072 but at this point I think the problem is before. I've tested 79L12 and 78L12 and they work, same thing with 7915 and 7815...
I cant understand where the problem is.

Thanks for your help...


Alessio

 

[ruckus328]

Hi everybody,
I've finished my first gssl compressor but i have some problems with the voltages.
I'm testing the voltages without the components only on the sockets.
TL074
pin 1: +2,8V
pin2: +2,8V
pin3: 0
pin4: +12,17V
pin5: 0
pin6: +2,6V
pin7: +2,7V
pin8: +3V
pin9: +2,6V
pin10: 0
pin11: -12,21
pin12: 0
pin13: 0
pin14: 0

I've never seen something like this in this forum and are weeks that I'm looking...
I've also wrong voltages on the TL072 but at this point I think the problem is before. I've tested 79L12 and 78L12 and they work, same thing with 7915 and 7815...
I cant understand where the problem is.

Thanks for your help...


Alessio

I'm confused why you think you have a problem?  The only pins that matter as far as supply voltages are 4 & 11, in which you do have the correct voltages.  The other pins will vary depending on the signal you're feeding the compressor, but without your chips installed and without feeding the compressor a signal at a specific RMS Level your Voltage readings on the other pins are completely useless.