GSSL HELP THREAD!!!

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

So, what does this mod do to the psu? I need to know before I mod my GSSL because it sounds soooo sweeeet and I don't want to screw it up in any way. Do other people have that problem? Is the original PSU design defective??
 
all that I know is that the original psu design is wrong and the clone does not always power up correctly.. that mod affect only the psu section, so it's not going to damage your gssl  ;)

Anyways, I still can't understand why I have my meter showing different gain reduction as I change the ratio...  too much compression at 2:1, rather than at 10:1...all the resistors seem to be ok.. could it be something that I've broken, like my 78L12 ? I don't know if this could affect the ratio...

sorry if I ask again...I'm just getting crazy with all this work  ;D
 
For all I know, it is mentioned everywhere in this forum that the ration DOES affect the threshold. This happens to everyone. My gssl does it too, but I read that it is normal  ???
I just re-adjust the threshold each time I change the ratio to get the same amount of compression and keep the new ratio character.
Anyways, the controls seem to be stuck on 4:1, 10 ms Att., Auto Rel. with 4dB of reduction for me  ;D ;D (sounds lovely)
 
Hi all !

I´m new although I´ve been reading this for some time.

I´ve completed my first GSSL now and have troubles:

When I switch it on, the 10Ohm resistor which is in between the 1000uf and the 79L12 goes up in smoke. I´m using 1/4 watts resistor there.

I did the following:

- checked the board for shorts - nothing found
- disconnected the 15V-regulators - same problem
- disconnected the 78L12 - same problem
- changed rectifier, changed 1000uF cap, changed 15V regulators - same problem
- checked all resistors and caps which are in the 12V-area of the circuit - no problem found

In between this steps i always changed the 10Ohm Resistor and the 79L12 before testing again - had 5 pieces of each

I kept testing three hours and now I´m out of parts - will get new 78L12 tommorow.

It would be great if someone here has a hint for me in the meantime, i think i don´t see the wood for the trees ???

I have seen other people here had the same problem but it seems like there can be multiple reasons for that.

Other information : I use a 2x15V 10VA toroid, i have no ICs on the board yet, the panel board wasn´t connected yet, the board is not
installed in a case yet.

Thanks in advance for any help

Arne


 
ytsestef said:
For all I know, it is mentioned everywhere in this forum that the ration DOES affect the threshold. This happens to everyone. My gssl does it too, but I read that it is normal  ???
I just re-adjust the threshold each time I change the ratio to get the same amount of compression and keep the new ratio character.
Anyways, the controls seem to be stuck on 4:1, 10 ms Att., Auto Rel. with 4dB of reduction for me  ;D ;D (sounds lovely)
Check a couple pages back in this thread where I posted a link for proper calibration procedures.  If your GSSL is properly calibrated then there is no change in Gain Reduction or output when toggling between ratios.  If you're using the resistance values as labeled on the pcb, then you will have the problem that you have.  The values on the pcb are for the THAT2150 or DBX202.  If you're using THAT2180 or 2181, then you need to change some resistance values.  Go to http://homepage.mac.com/marten.thielges/gssl/calibration.html and http://diy.fischerworks.com/gssl_vca.shtml
 
regularjohn said:
ytsestef said:
For all I know, it is mentioned everywhere in this forum that the ration DOES affect the threshold. This happens to everyone. My gssl does it too, but I read that it is normal  ???
I just re-adjust the threshold each time I change the ratio to get the same amount of compression and keep the new ratio character.
Anyways, the controls seem to be stuck on 4:1, 10 ms Att., Auto Rel. with 4dB of reduction for me  ;D ;D (sounds lovely)
Check a couple pages back in this thread where I posted a link for proper calibration procedures.  If your GSSL is properly calibrated then there is no change in Gain Reduction or output when toggling between ratios.  If you're using the resistance values as labeled on the pcb, then you will have the problem that you have.  The values on the pcb are for the THAT2150 or DBX202.  If you're using THAT2180 or 2181, then you need to change some resistance values.  Go to http://homepage.mac.com/marten.thielges/gssl/calibration.html and http://diy.fischerworks.com/gssl_vca.shtml

yea, i changed the 3,9k to 5,1k and removed all the pin4-nc-related stuff but still there is some slight change. Anyway, I have an ever-so-slightly more annoying problem. I use the behringer meter with the 20dB scale (the one from diypartsupply) and have used the 2k resistor (as specified for 20dB scale meters) but still when the meter shows 4dB reduction, 2dB actual reduction happens, when meter shows 12dB reduction, 6dB actual reduction happens, etc. It is always showing double values from the actual ones. Should I try a 3,9k resistor in place of the 2k?
 
my gssl seems to be modded corectly for using that2180, and I still have the meter showing gain reduction without audio passing through it.. is it a ratio problem, maybe caused by some broken component? ...i have the turbo board connected to the ratio, could this be the problem? 
 
ytsestef said:
yea, i changed the 3,9k to 5,1k and removed all the pin4-nc-related stuff but still there is some slight change. Anyway, I have an ever-so-slightly more annoying problem. I use the behringer meter with the 20dB scale (the one from diypartsupply) and have used the 2k resistor (as specified for 20dB scale meters) but still when the meter shows 4dB reduction, 2dB actual reduction happens, when meter shows 12dB reduction, 6dB actual reduction happens, etc. It is always showing double values from the actual ones. Should I try a 3,9k resistor in place of the 2k?

You also need to remove the 10K resistors that bridge pins 3 and 5 in the audio and sidechain VCA's.  You need to remove the 68R resistor from pin 4 of the audio vca's.  You need to remove the 47R from pin 4 of the sidechain VCA.  You need to replace the 1M resistors just outside of the audio VCA circuits with a 680K for the 2181, and remove it for the 2180.  And you need to replace the 10K resistor on pin 4 of the audio vca's with a wire link if you have the 2181 and remove it if you have the 2180.
As for the meter resistor, use a 3.3K for 20dB scale if you have the behringer style meters.
 
3.3K, eh?

Well, at first i tried 2K in it always shows double the actual reduction and I figured that 4K will be ok. Then I read that 4.7K is the ideal for behringer meters, then 3.3k (both are equally spaced values from 4K  :D )
Oh well, I'll buy both tomorrow and see which one gets closer. I'll also try the PSU mod (GOD, unstable problems in electronics are frustrating!) and I'll report back to you guys. Thanks a lot!

Hopefully, painting is going to get finished tomorrow too, so maybe I'll post a picture or two!
 
ytsestef said:
regularjohn said:
ytsestef said:
For all I know, it is mentioned everywhere in this forum that the ration DOES affect the threshold. This happens to everyone. My gssl does it too, but I read that it is normal  ???
I just re-adjust the threshold each time I change the ratio to get the same amount of compression and keep the new ratio character.
Anyways, the controls seem to be stuck on 4:1, 10 ms Att., Auto Rel. with 4dB of reduction for me  ;D ;D (sounds lovely)
Check a couple pages back in this thread where I posted a link for proper calibration procedures.  If your GSSL is properly calibrated then there is no change in Gain Reduction or output when toggling between ratios.  If you're using the resistance values as labeled on the pcb, then you will have the problem that you have.  The values on the pcb are for the THAT2150 or DBX202.  If you're using THAT2180 or 2181, then you need to change some resistance values.  Go to http://homepage.mac.com/marten.thielges/gssl/calibration.html and http://diy.fischerworks.com/gssl_vca.shtml

yea, i changed the 3,9k to 5,1k and removed all the pin4-nc-related stuff but still there is some slight change. Anyway, I have an ever-so-slightly more annoying problem. I use the behringer meter with the 20dB scale (the one from diypartsupply) and have used the 2k resistor (as specified for 20dB scale meters) but still when the meter shows 4dB reduction, 2dB actual reduction happens, when meter shows 12dB reduction, 6dB actual reduction happens, etc. It is always showing double values from the actual ones. Should I try a 3,9k resistor in place of the 2k?
The 3,9K to 5,1K change is only for the audio-VCAs. The sidechain-VCA pin5 is connected to -12V and not to -15V.
For the ratio you change the 100K (used for DBX2150) between TL074-pin8 and TL072-pin6 to 120K or 127K (for That218x).
Take a look at the latest gssl schematic, component placement, .. (still containing minor errors and discrepancies to pcb, like the 10Ks @ NE5532 non-inverting inputs, the mixed up 91K/750K @ auto-release on pcb, .., but getting closer). Use a 4,7K trimmer (wiper dialed @ center for 1st powerup) and adjust it for the needed value for your meter to match gain reduction with needle movement.
 
Harpo said:
ytsestef said:
regularjohn said:
ytsestef said:
For all I know, it is mentioned everywhere in this forum that the ration DOES affect the threshold. This happens to everyone. My gssl does it too, but I read that it is normal  ???
I just re-adjust the threshold each time I change the ratio to get the same amount of compression and keep the new ratio character.
Anyways, the controls seem to be stuck on 4:1, 10 ms Att., Auto Rel. with 4dB of reduction for me  ;D ;D (sounds lovely)
Check a couple pages back in this thread where I posted a link for proper calibration procedures.  If your GSSL is properly calibrated then there is no change in Gain Reduction or output when toggling between ratios.  If you're using the resistance values as labeled on the pcb, then you will have the problem that you have.  The values on the pcb are for the THAT2150 or DBX202.  If you're using THAT2180 or 2181, then you need to change some resistance values.  Go to http://homepage.mac.com/marten.thielges/gssl/calibration.html and http://diy.fischerworks.com/gssl_vca.shtml

yea, i changed the 3,9k to 5,1k and removed all the pin4-nc-related stuff but still there is some slight change. Anyway, I have an ever-so-slightly more annoying problem. I use the behringer meter with the 20dB scale (the one from diypartsupply) and have used the 2k resistor (as specified for 20dB scale meters) but still when the meter shows 4dB reduction, 2dB actual reduction happens, when meter shows 12dB reduction, 6dB actual reduction happens, etc. It is always showing double values from the actual ones. Should I try a 3,9k resistor in place of the 2k?
The 3,9K to 5,1K change is only for the audio-VCAs. The sidechain-VCA pin5 is connected to -12V and not to -15V.
For the ratio you change the 100K (used for DBX2150) between TL074-pin8 and TL072-pin6 to 120K or 127K (for That218x).
Take a look at the latest gssl schematic, component placement, .. (still containing minor errors and discrepancies to pcb, like the 10Ks @ NE5532 non-inverting inputs, the mixed up 91K/750K @ auto-release on pcb, .., but getting closer). Use a 4,7K trimmer (wiper dialed @ center for 1st powerup) and adjust it for the needed value for your meter to match gain reduction with needle movement.
Thanks a lot Harpo! Most of these thing have been done. I'll do the rest tomorrow.
and what about the mixed up 91k/750k?? should they be the other way around???  ??? ???
 
hi harpo & ytsestef

i think the 91k & 750K are flipped on the schemo but on the right order on the PCB...
correct if i'm wrong ;)

but i don't get what's up with the 10K on the 5532 ?

thanks
 
well, I did everything there was to it to adapt the circuit to the 2180 vca's. Everything is normal now, when changing ratio, the reduction changes just a bit, as does the wave's fx g384 plugin, ratios are correct.

resolved the meter problem, too, using a 4,7 resistor.

The problem with the 12v rail disappearing upon startup sometimes persists, though.
I changed the two 1000μF electrolytics to 2200μF, that reduced the problem a lot.
Now it does it 1 out of 6 times or so, whereas with the 1000μF it did happen a lot more frequently (like 50% chance each time).
I also tried the psu mod that angelo suggested (with the 2 ceramics 100nF/50V and the 1N4007 diode), and it did nothing noticable. Should I remove the mod since I can't notice any difference? Or is it actually better to leave it on?

and ultimately, what am I going to do with the powerup problem? I think just going with bigger caps is plain wrong, although it is a workaround, this is a problem that has to be fixed, not camouflaged.
 
bassculture said:
i think the 91k & 750K are flipped on the schemo but on the right order on the PCB...
correct if i'm wrong ;)

but i don't get what's up with the 10K on the 5532 ?
edit ;)

Inverting output buffer from schematic shows a 10K between NE5532 non-inverting input pin5 and 0V (amp stages are flipped, so this is pin3 @ pcb), contrairy to the pcb overlay showing a 470 ohm instead, I'd use a 0 ohm/wire link but YMMV.
 
So I have to switch the resistors next to the release switch and put 750k near the 47μF cap and 91k near tha 6,8μF?
Is this confirmed? Has anyone done it? What happens if I don't? I use the auto release very often so this is important!! :eek: :eek:

What about the 10k/470Ω?
Is the schematic or the pcb layout wrong?? Maybe pcb is updated and schematic outdated? Maybe not?

Wow! I didn't know about any of these. Is there a META that contains all errors and mods needed?

EDIT: Answered my own question by reading your post again, obviously the pcb is the correct one, since you'd use wire link, and 0 ohm is closer to 470 ohm than 10k.  ::)
 
Hi Guys,
i have recently built my 3rd gssl and have a problem with the make up gain. i have read through around 100 pages of this thread and used the search but cannot find any info.

the problem is that the make up gain does not seem to work until the pot is around 1/2 the way clockwise.
can anyone point out where about i should look. thanks
pagel
 
Hi experts and beginners (like me) :'(

I am looking for good connectors for the
2pcs 5POLE. PIN connectors, PCB, 1/10" (2.54mm pitch)
1pcs 10POLE. PIN connectors, PCB, 1/10" (2.54mm pitch)
1pcs 10POLE connector, Cable, 1/10" (2.54mm pitch)
2pcs 5POLE connector, Cable, 1/10" (2.54mm pitch)
16pcs CRIMP TERMINAL for pcb connectors

I cannot find theses anywhere

is there anyone who know where to find theses?

thank you
Alain.
 
Alain said:
Hi experts and beginners (like me) :'(

I am looking for good connectors for the
2pcs 5POLE. PIN connectors, PCB, 1/10" (2.54mm pitch)
1pcs 10POLE. PIN connectors, PCB, 1/10" (2.54mm pitch)
1pcs 10POLE connector, Cable, 1/10" (2.54mm pitch)
2pcs 5POLE connector, Cable, 1/10" (2.54mm pitch)
16pcs CRIMP TERMINAL for pcb connectors

I cannot find theses anywhere

is there anyone who know where to find theses?

thank you
Alain.

You can get them from www.mouser.com
Part # 651-1725685 (5 pin pcb Molex terminal)
And, 651-1725737 (10 pin pcb Molex terminal)

Not sure about the others.  If you look around mouser's site I'm sure you'll find em
 
The most widely used ones are manufactured by Molex.
I found them in farnell.
Too lazy to search for 5pin and 10pin but it is essentially the same, the number of pins changes only. Here are three-pinned versions of what you need (I used these for the Turbo Board):

PCB Connectors 1/10": http://export.farnell.com/molex/22-27-2031/header-square-pin-0-1-3way/dp/9731156&PRED=True
Cable Connectors 1/10": http://export.farnell.com/molex/22-01-2035/crimp-housing-0-1-3way/dp/143127?in_merch=true&MER=i-9b10-00001460
100pcs Crimp Terminals: http://export.farnell.com/molex/08-50-0032/crimp-pin-22-30awg-pk100/dp/9773789?in_merch=true&MER=i-9b10-00001460

Good luck.
 
ipagel said:
the problem is that the make up gain does not seem to work until the pot is around 1/2 the way clockwise.
Pin "E" of the molex conector is probably varying voltage between 0 and +12V correctly, else check wiring at the bypass switch.
Doublecheck resistor values connected to pin 6 of the TL072. (620K Rin and 100K Rfb)
Good luck
 
Back
Top