GSSL HELP THREAD!!!

[DBarbarulo]

hi Jakob,
thanks for the help registering here!
Yes sorry, 072 and 074... just going mad with numbers during this journey.
I,  say, "hundredchecked" the bypass switch config, also swapped the hardware switch with another 5A rated i had in the studio thinking of a bad unit. I followed strictly Gustav's graphic in his guide. Unless there's something i miss visually, my wiring looks 1:1 as the manual.

 

[DBarbarulo]

Just as a guess: if soldering is fine, continuity is fine, rails are fine what is able to do a mess with controls and meter?
Now, after unsoldering and resoldering leads (10p) to the control board (no shorts etc.) i have the meter directly influenced by threshold (with no audio on input so is moving by it's own) when in bypass. If compression "active" i can move the meter with the makeup gain... always without audio passing...

 

[Script]

Just a guess: your soldering is not fine

 

[DBarbarulo]

thanks @Script, but looks like is fine and  continuity (in my knowledge) supports the theory that is fine.
But you are right the board behaves like i have 25 shorts around. I checked all the sockets, from top to bottom, followed traces on the control board, switches etc. and can't find any clue. Are you excluding some bad IC?

 

[Script]

The meter tracks GR (gain reduction).

If no input is connected., but the needle moves when turning THR or MAKEUP pots, then something is fundamentally wrong.

-- THR only moves the needle when audio signal is present.
-- MAKEUP pot should never move the needle.

Ergo, you have shorts or wrong connections, maybe causing internal feedback ?

 

[DBarbarulo]

I know, looks like is all wrong but can't figure out what since my wiring looks 1:1 as Gustav's guide unless there's something to be seen i'm not able to see or my board has something different than the one in the guide.
Now spent an hour looking at traces with backlight. Checked 100 times polarized cap's orientation 10 pin to 10 pin pad's insulation and continuity from control pcb to board and subsequent elements. Checked whipes off circuit and are fine. Checked if there's something on chassis but 0.
With all ICs pulled out the behavior is the same, tension.
Now i'm cutting the molex connector to solder vires directly.

 

[DBarbarulo]

Ok,
i figured it all at the end.

I would like to leave some hints to absolute beginners like me:
symptoms were crazy meter, unresponsive threshold pot and control board, makeup gain not linear etc. A mess.
When my troubles occurred i had several reports about "check for shorts" since the situation appeared like major faults in wiring or soldering but i followed strictly the book.
No errors.
I did it anyway, probing, removing molex etc.
Nothing.
So i started thinking about blown/wrong components etc.
Not the case.

A "good soldering" is not only the precision of "points" but also a psycho level of cleaning the pcb after.

I did a "quite good" job (but not enough), used flux, heat probed station, good tin compound and great care with handling, no blobs, no cold joints etc. Did some reflows thinking about a bad joint.

Nothing.

After the soldering i cleaned the pcb with 95% iso alcohol and rubbed away the flux with a toothbrush and some paper and cue tips to dry some areas. 

Troubles... "check for shorts".

So a step up i started looking at the soldering points one by one with a 10X loupe glass.
Nothing... "check for shorts".

So a step up i started looking with 10X on a light source (with light from behind the pcb) to have a greater contrast of traces and joints with transparency.
Nothing... "check for shorts".

Yesterday night i used a 25 to 200X usb microscope, for fun,  to look around and noticed some dark gunk totally invisible at lower magnification with a glass in some spots. So today i did a pass of Philips 339 CCS and seriously brushed all the joints with a brass brush.

Guess what?
The unit works perfectly.
Probably a thin debris was sticking somewhere in a critical point of the VCA section, believe me INVISIBLE at 10X magnification with proper light. I looked at the sidechain zone for days, probed, unsoldered, measured, resoldered, xrays, roadmap, ICs...
N-O-T-H-I-N-G.

So to any unlucky beginner: clean that damned job with a stiff brush and a solvent without scraping the insulation naturally but with much more power than you think an electronic board should take.   

DB

 

[Gustav]

After 30 mails with the vendor and 150kg of frustration i realized with a 10x glass that i had wrong chips... 

Long story short i had the missing 4 5534s, installed them and the unit started to work properly like magic

You seem to have forgotten the point where I asked you if you had put a fuse in it, and you assured me you had, but we had to exchange 3 mails before getting back to the fact that you hadnt.

You seem to have forgotten the part where I asked you to remove all ICs and make sure voltage were correct before proceeding, to insert active components, and they werent, which means there certianly was a problem with your soldering work, as well as me sending you the wrong 5534.

And the part where I spent time going over your soldering and found a short on the input from the toroid, and a short on your regulator for you.

You also took the time to send me a mail to explain to me, that the only problem was the wrong chips, and you now had a 100% working unit, which seems to have not been the case, since you came on here to look for help.

I am glad your unit is working.

Gustav

 

[Script]

seriously brushed all the joints with a brass brush

Uomo, that is hardcore abuse and not to be imitated. But we got how frustrating it must have been. And as long as you promise you won't start sand-blasting your boards... 

Maybe next time check for connectivity meticulously with a meter using the trace overlay downloadable from Gyraf's website.

Happy you have a working unit  8)

 

[DBarbarulo]

You seem to have forgotten...
Gustav

Not at all.
Did you read somewhere mentions or comments directed at you  to justify again an aggressive tone?
I've been kind, here and before.
I am a lot more kind than that when i'm in the vendor position.
Please, as i suggested, about the fuse put a mention (or a picture) in your guide or put one in the supply. A customer will understand. Also i'll include too a basic written guide in a kit that you advertise as "beginner friendly" (that i read as "no degree of engineering required") showing say basics of how to probe and verify rails before putting chips etc. A minimal troubleshooting guide will be helpful too.


Thanks

DB

 

[DBarbarulo]

Uomo, that is hardcore abuse and not to be imitated. But we got how frustrating it must have been. And as long as you promise you won't start sand-blasting your boards... 

Maybe next time check for connectivity meticulously with a meter using the trace overlay downloadable from Gyraf's website.

Happy you have a working unit  8)

The suggestion is "clean vigorously"... ;-) use a stiff non abrasive brush and a good cleaning solution.

if you want laugh i downloaded the overlay long ago and edited it on a Surface with a digital highlighter to trace every "tricky" situation where potential proximity of pads may lead to a short . Believe me or not i spent days with a 10x loupe and beeper searching for that "heavily  wrong" situation. Also trying to collect spare parts (with suppliers closed) to try a new 074 or a 072...

I was convinced about a bad chip since when first connected the right 5534s the unit worked properly for 1 hour or so. SO excluding a blown component, or a damage to power supply the only reasonable thing was a gone bad chip.
Probably a debris was doing an intermittent short and thermal cycle of the board was refining the contact...

Yo! 

 

[Script]

Just happy you have a working unit now  8)

 

[DBarbarulo]

Now in the next days i'll receive by mail another quite complete set of ICS, and a cargo of caps  ;D

Edit:

i saw too people cleaning pcbs with ultrasonic cleaner... i have one! but i  think is supposed to clean an unpopulated board...
I can try next time to put a finished board inside the cleaner.... maybe disconnecting power chord before. 

 

[DBarbarulo]

by the way,
the 2:1 ratio has a very long knee catching well below 4:1 and 10:1.

 

[DBarbarulo]

And as long as you promise you won't start sand-blasting your boards... 

Honestly?
I grinded with a dremel and a jeweler's bit (the one to put point diamonds in rings) the space between 2 pads under the sidechain VCA that had some dark debris under the insulation! Next time i'll use the blaster!!  ;D

 

[Script]

I can try next time to put a finished board inside the cleaner.... maybe disconnecting power chord before.

Might be an idea should stores run out of light bulbs due to hoarding or cut off supply chain    ;D

Spare ICs never a bad idea.

Enjoy the unit !

 

[DBarbarulo]

here it is!

 

[Gustav]

Not at all.

Cool, I got a different impression from your post, where you were refering to me, so I wanted to see if you forgot.

Glad you are happy with your first DIY build, and you got it working.

Gustav

 

[DBarbarulo]

Cool, I got a different impression from your post, where you were refering to me, so I wanted to see if you forgot.

Glad you are happy with your first DIY build, and you got it working.

Gustav

Hi gents...

After 30 mails with the vendor and 150kg of frustration i realized with a 10x glass that i had wrong chips... 

Long story short i had the missing 4 5534s, installed them and the unit started to work properly like magic BUT for 1 hour or so...

I think is, as you correctly say, just an impression. Troubleshooting the post, too many words and phrases are missing to acquire the meaning you felt.

Thanks again

DB

 

[fandusss]

Hi guys,

I've had a GSSL change in sound over the years, as things do. Less brilliance on the top-end. Low mids are somewhat muddy. It muddles the transients. Looking to restore some Hi-Fi qualities to the unit, for lack of a better term. Before I go swapping things unnecessarily, can anyone point me in the direction of the usual suspects in these units?

Running turbo mod, side chain mod, THAT 2180B vca's. Should I start with the electrolytics? Opamps? Any advice from an experienced person here who has done this before would be greatly appreciated.

Thanks!