GSSL HELP THREAD!!!

[MeToo2]

Also, the picture I posted was taken with ratio @ 10.
With ratio 4 the 10Khz oscillation is less prominent.
With ratio 2 the oscillation disappears but it appears anyway as "soft" clipped on the tops (just like the photo, but without the little waves).

What attack and release timing settings have you got when making those traces?

As you said, the side chain control voltage should be pretty much DC on long release times.
S/C CV should go virtually straight down (-ve) on short attack times, when any peaks come in.
S/C CV should discharge with a classic CR discharge with a time constant >> 1/200 Hz back up towards 0V when there are no peaks/signal present.

TL074 pin 14 should clearly contain significant audio.
But after the diode, the S/C CV at point C should not contain any significant audio signal on long release times, regardless of any oscillation.
Point C & D should be -ve wrt 0v.

Checked diode orientation and tantalum cap orientation in the s/c filters?
Checked for shorts around this area?
Leaky caps? duff diode?

 

[qwerty8787]

Timing should have been longest attack and auto release. But I cannot see any significant changes when changing them.

Even bypassing the 2 SSC and Turbo board (taking back everything to standard version) I get the same behaviour.

Orientation of tantalums on front panel pcb is ok.

Checked for shorts between poles in flat cable connecting main to control board.
Checked for shorts in main pcb.

 

[MeToo2]

Timing should have been longest attack and auto release. But I cannot see any significant changes when changing them.

Even bypassing the 2 SSC and Turbo board (taking back everything to standard version) I get the same behaviour.

Orientation of tantalums on front panel pcb is ok.

Checked for shorts between poles in flat cable connecting main to control board.
Checked for shorts in main pcb.

It looks like something is (rapidly) discharging your timing chain, so that basically the CV is modified audio, rather than it being peak detected and filtered to near DC.
Checked the 1n4148 diode on TL074 pin 14? Has that gone short circuit?

 

[gyraf]

Maybe a missing wire link on the front panel pcb - somewhere close to where the meter attaches. This has been the problem in more than one case in this thread

Jakob E.

 

[inputhone]

Hey Folks i have some issues with my ssl clone. first it startet to get distorted from time to time, repowering it helped. then when i decided to check it through i suddenly have very loud distortion on one channel and the other is very quiet.

i wanted to do the step by step procedure, but i am not sure what i am doing wrong.

Feed both gssl inputs with same source, maybe from a keyboard or frequency generator, level set to about 0dBV (1V RMS, 1.414V pk), frequency set in range of 100 ... 400Hz to allow a more reliable measuring with a cheap meter that won't measure higher frequency testtones.

Input stage: Measure both outputs of the line receivers, NE5534 pin6. They should measure the same 1V AC. Case not, check for shorts, same 22K resistor values or replace the broken opamp.

i lunch my tonegenerator, sine wave, connect ground (generator) to ground-in and link it with negativ-in then connect positive (generator) to positiv-in with each input, feed 100 hz at 2,8Vpp and i measure on my pinout of 5534 pin 6 only 0,4VAC. what is wrong? checked resistor values (all good) replaced the ops (working ones) still same. am i wring it wrong?

 

[inputhone]

ah, shit i found out that simply one vca is broken.  :-\ exchanging the broken one brings all back to normal. voltages are good everywhere.

but still: why do i get so different measurements when doing the first step of the suggested check, mentioned above? any ideas on that? what do i do wrong?

and: how is it possible that this vca broke? didn't do anything but use it from time to time? anything special that i should check?

thanks folks! 

 

[qwerty8787]

It looks like something is (rapidly) discharging your timing chain, so that basically the CV is modified audio, rather than it being peak detected and filtered to near DC.
Checked the 1n4148 diode on TL074 pin 14? Has that gone short circuit?

Just checked it: correct orientation and tested with DMM.

Maybe a missing wire link on the front panel pcb - somewhere close to where the meter attaches. This has been the problem in more than one case in this thread

Jakob E.

Currently I'm not at home, so I cannot check that until Friday.
I'm 90% sure that I putted that link, anyway I really hope I'm wrong....  :-X
I'm so tired of this endless troubleshooting. The good thing is that by now I know much better how this compressor (and the VCAs) works. Hope someday I could use this thing!  :

MeToo2, Gyraf, thanks for helping me! 

 

[qwerty8787]

Maybe a missing wire link on the front panel pcb - somewhere close to where the meter attaches. This has been the problem in more than one case in this thread

Jakob E.

No way. I checked for the missing link and it's there. That's not the problem  :'(

Any other thoughts?

 

[qwerty8787]

the f#@§ing release rotative switch were somehow mounted wrong. So there wasn't any connection to the release net.
Now it works! Finally!!

I think I have to fine tune the meter response putting in a trimmer instead of resistor and make some measurements.

Also I feel like there isn't so much GR when I'm putting down hard the threshold. I already putted the 47K resistor on legs, but I think the gain reduction should reach greater ranges. I have to pump up on the input to reach peaks in GR of 15-20dB. But when it's bypassed it will distort the next stage. not so good.

In the 2x SSC boards I shunted the 2 input pins, this already gave me more gain reduction. What can I do to increase it a little more?

Many thanks again to everyone helped me 

 

[jim.jenca]

Hey everyone,
im new to the whole DIY thing.  I decided to go with the gssl for my first build but i've ran into an issue while getting it up and running.  It seems that it will only compress audio that is hard panned either left or right and decreases in compression as it gets closer to the center eventually not compressing at all on things that are panned directly in the center, like the kick drum.  Does anyone have any ideas on how to fix this problem?  It would be greatly appreciated. 
Thanks,
Jim

 

[Harpo]

Jim, double check input side hot/cold XLR wires internally or externally. One side is out of phase, so summed signal at sidechain in cancels out common (centered) to both L/R signals.

 

[jim.jenca]

Harpo,
Thank you so much.  This was exactly the issue and has been resolved.  Now my GSSL is up and sounding awesome!

 

[Xab13r]

Hi guys,

I'm trying to complete my first SSL Buss Compressor. Right now I'm having some doubts regarding the use of THAT2181B VCAs.

I've followed the very useful guide to tweak the board for the 2181s by Fischer http://diy.fischerworks.com/gssl_vca.shtml.
I've chosen to use the Super SideChain board from PCBGrinder.com

Here my question:

According to the Fischer's page I should use a 120k resistor in place of the 47k couple (in VCAs output section) - but for use the board with the SSC I'm supposed to remove them and use their spots to plug-in the SSC board. Do I have to do any modification on the SSC board to be sure the VCAs are correctly used?

Thanks in advance.

 

[Harpo]

According to the Fischer's page I should use a 120k resistor in place of the 47k couple (in VCAs output section)

Keep both 47Ks for the GSSL and leave them out when using the SSCF board. (The 120K was ment for the 1st.pink row ratio "adjustment". For adjusting ratio, use a maybe 91K in series with a 50K trimmer/rheostat, trimmers initial value about centered instead of the 100K or 127K* from later pcb revisions.)

but for use the board with the SSC I'm supposed to remove them and use their spots to plug-in the SSC board. Do I have to do any modification on the SSC board to be sure the VCAs are correctly used?

No other mods required on the SSCF board.

 

[Xab13r]

Thank you Harpo. It's been very helpful

 

[tnerbsivraj]

The left channel seems to work great in all regards but the right channel only outputs the bypassed signal. When I fed a signal to the right channel alone it appears on the meter to be compressing and varying with changes of threshold, ratio etc... but the output stays the same as if in bypass.

Right channel VCA does not get a sidechain control signal - check output voltages at the 5534 driving the VCA control input, compare L and R while compressing.

Then look for errors around R VCA

Jakob E.

Great, that seems to be it exactly. I'm getting 0v on pin 6 and 2 of 5534 of the right channel (contrary to the left channel) although I haven't figured out why yet. Seems like appropriate readings everywhere else but will keep looking... The help is greatly appreciated

 

[phancak]

How is current draw GSSL???
Thanks

 

[Harpo]

How is current draw GSSL???
Thanks

How is ltr/100km my car??? (and I don't tell you what car it is, ideling, driving up or down hill, full throttle, ...)
Do you really expect anyone to pull his GSSL out of a rack, open and disassemble it, cut some pcb traces or lift parts to measure current draw at whatever part or spot you didn't specify or ask for, maybe the +15V, -15V, +12V, -12V rails or AC mains and come up with some numbers that probably will be different from yours because the other build maybe uses light bulbs for meter illumination, status LEDs, uses different type of opamps or VCAs, might have built in additional circuits like turbo or sscf, might have been measured with signal present, ...

 

[phancak]

How is current draw GSSL???
Thanks

How is ltr/100km my car??? (and I don't tell you what car it is, ideling, driving up or down hill, full throttle, ...)
Do you really expect anyone to pull his GSSL out of a rack, open and disassemble it, cut some pcb traces or lift parts to measure current draw at whatever part or spot you didn't specify or ask for, maybe the +15V, -15V, +12V, -12V rails or AC mains and come up with some numbers that probably will be different from yours because the other build maybe uses light bulbs for meter illumination, status LEDs, uses different type of opamps or VCAs, might have built in additional circuits like turbo or sscf, might have been measured with signal present, ...

I thought base scheme GYRAF. Current draw is base value.

 

[MeToo2]

I thought base scheme GYRAF. Current draw is base value.

It might help if you said why you need the info on current draw. Transformer sizing? Heat?

I know in my build, the lamps for the switches were the biggest current draw, followed by the relays on the SSC board, and neither of those are part of the Gyraf schematic/spec at all.