GSSL HELP THREAD!!!

[museic]

Hey Chrome Heart, Thanks for the welcome and the parts numbers!! Helps a lot.

I've got another question...

Is there page about making this build true bypass? Didn't see one in the META.

 

[onlymeeee]

Hi,

I've built a gssl with the that 2181b and turbo board. (using the same caps in kev's partlist)  I am pleased with the results.
However it's much much more punchy than the original. Although this is gonna be perfect for drums, I'd like a softer one more like the original for the mix bus.

Is it the dbx 202x that will give me this sound?  Has anyone else noticed this?

 

[feza]

Have you tried this version? There're two versions of the GSSL if I am correct. The Panasonic and the Nichion's. The difference lies in the signal capacitors. The panasonic is more mid scooped and extended low with high frequency. The Nichion caps gives a smoother sound overal. This is quoted from Greg's posting.

http://members.cox.net/ram3n/Greg_GSSLwPW.pdf

I'm building that version. About a week the unit is build and surely let you people know. 

 

[JW]

onlymeeee, is he describing what to do if you have DBX202's,  or is he describing the changes back to DBX202's, meaning I shouldn't reverse the following?
"If I remember rightly the original layout was done with the DBX202 package in mind, so the resistor values on the PCB will be correct except for the fact that ratio resistor is changed to 127k (120k is close enough) and the resistor feeding the VCA is changed to 27k because the balanced output gives +6dB gain."
(In actual fact the value should be 30k if you want unity gain).

Thanks

 

[onlymeeee]

What to do when you are using 202s.

As far as number 2.  How does it sound?? Are you getting correct ratios?  Unity gain?

 

[JW]

Wait a second. I didn't remember reading this on Jakob's site the first time but he says this concerning using THAT 2181's instead of DBX 202's which contain the 2150 chips:

"As this compressor was originally designed for using the - now obsolete - DBX2150 VCA's, there is a change in one resistor that is necessary to obtain the correct ratios - look here to see the component in question - the 100K resistor marked has to be replaced by a 127K resistor to compensate for higher input current sensitivity on the new chips."

So, it would appear that I need to go back to the 100K resistor since I have the 202's.

As for the 15K resistors by the VCA's, I think this was just a general mistake, and that they need to be 30K each for unity gain, regardless of the type of VCA you're using. Correct?

Sorry to beat this dead horse.

The meter appears to be working quite well. I wish I could know exactly it's specs though, so I could find out what value the resistor across the meter needs to be. (330R is the original specified.) Anybody know how to figure this out?

 

[feza]

Hey,

I have a question with the power switching. Having the SPDT on-on switch for the power do I connect the mains power "live" wire to the top of the three poles from the SPDT, the middle pole connected with the transformer primary 230v wire? Then the "earth" wire of the mains connected to the chassis and as last the "neutral" from the transformer to the neutral pin on the mains inlet?

Just can't get my head around this. Thanks for the reply!     

 

[Harpo]

yes, switching the unit 'on' when switch toggle directing upwards. If you want it the other way round, turn the switch or connect to the center- and bottom pole. (you can easily ohm it out).

 

[Kir_For_Pleasure]

Have my gssl up and running now, really nice smashing machine !!!
Just wants to thank Gyraf for sharing this project and Gustav for the pcb work !
NI KILLAR ÄR GRYMMA !!!

 

[feza]

yes, switching the unit 'on' when switch toggle directing upwards. If you want it the other way round, turn the switch or connect to the center- and bottom pole. (you can easily ohm it out).

Thanks!

Finally the last parts will arrive in the coming days. Probably have this build in the same week out of enthusiasm! 

 

[feza]

Hey !!!!!! 

Turned the unit on and all the voltages are correct. Although the chips are not in the socket yet and have not heard one single sound out of the GSSL, it is still euphonic this moment. 

I'll keep you people updated! 

 

[feza]

Hey fellow builders. Measured all the voltages and everything is correct!

I was inserting the GSSL in my Mackie mixer, but no sound. Guessing that the XLR's are not correct. I followed the plus and minus on the PCB section BUT was thinking that the pins on the actual Neutriks must be different. What I did was connecting the HOT (pin 2) and COLD (pin 3) on the Neutriks as illustrated on the XLRs themselve. So I am not getting it entirely the idea but knowing not doing it right. Just don't get it right now. ;D

Working full 3 days building it, just one little thing. Please can someone comment? 

 

[Chrome Heart]

When you say "inserting", are you using the "inserts" on the Mackie? To make it less complicated for test, why not just run it in line and see if your getting signal and forget about using the inserts for now.

 

[feza]

Thanks for the reply!

I'll solder the meter and bypass swith as well. Still I am quite convinced that it must be the Neutriks.

Will report as fast as possible! 

 

[feza]

Well. When the left input channel is fed I have signal and the meter works and bypass as well. But no output signal. And the right input does not affect the meter also. What I am doing wrong (still guessing) is that the Neutrik plug which is connected with the mainboard needs to be soldered some other way. I now have the Neutrik plugs soldered this way :

Hot (+):  Pin 2 (red)
Cold (-): Pin 3 (transparant)

But here comes the error probably. All the Neutriks (connected on the mainboard's I/O) are wired this way. But the wires in the Molex are wired like on the component layout of Gyraf. Probably that is the problem. It fases each other out in some way due by wrong Neutrik wiring. The leads connected on the actual unit needs to be wired that way, not the unit itself?  I just need to be sure before re-solder the plugs. This is the first time soldering XLR's and can not find any info regarding the wiring in such situation, other then the above pin reading. 

I'm tired. Worked three days straight on it and is almost finished, just a few things before mounting in the case. Many thanks for helping me!   

 

[Chrome Heart]

Easy. Solder pin 2 to + on the board and pin 3 to - and pin 1 to ground. Pay very close attention to how it is layed out on the PCB. If you have things phased correctly you will have no problem here. If you still have a problem it is somewhere else. You have to be very patient soldering this board, there is a high possibility of solder bridging. Check, recheck and then check again. Dont rush!

 

[Piotr]

Hi to all of you !

I was wondering which Lorlin switches you are using and recommending for the GSSL, BBM (break before make) or MBB (make before break) ? Didn't find the answer to this question elsewhere, thanks !!!

Piotr

 

[feza]

An update. I removed the 5K1 resistors as well in the channel VCA's. Re-soldered them and have 3v on the output pin. The right input and left output channels do receive/give signal from XLR's.  But the other two not, some kind of muffled and lowpass filter sound if connected when using the good output and "bad" input. When using both bad I/O it is just plain hum with very silent sound. I read that with the muffled sound it could be an wrong cap like mistaken the pF for an nF, that is not the case here.

I've measured all the opamps and VCA's with the TL's and still the voltages are correct. Reading +12 and -12 volts at the threshold and correct voltage at the makeup-gain. Those pots work. What doesn't work are the ratio, attack and release functions. I do hear some kind of collapsed sound when the played track has a peak, with a loud bassdrum for example. With that peak the sound will collapse and cuts almost ifself off for 2 second and returns as normal, like some gate effect. When the meter still was attached I saw this huge peak. It just doesn't work properly.

Reading  a lot but just can not find some solution yet. Thanks for your help! 

http://www.groupdiy.com/index.php?topic=5324 <--- Tracing the signals is a good idea. I'll keep you updated! 

 

[feza]

Yo,

I've made some measurements but still can not find the problem. In the schematics I traced the opamp signals but they are still correct at the end, or am I doing something wrong? What is the best way tracing errors with my I/O problem? To re-state my problem :

The left input and right output socket do not give proper audio. The sound is barely audible when connecting the two faulty sockets. But if the right input and left output is connected there is signal. Not totally right (collapsed sound with transient peaks) but there is sound. I tried swapping the opamps of the working I/O but the same results. Re-measured the opamps once again and they are all correct, including VCA's. Soldered again some joints which I thought were not right, same deal. Just weird symptoms.

Any help is very, very and very appreciated!         

Okay#1 : I measured 6 volts on the left input! That is not correct, right? Every component in that path has 6 volts with close to 7. Don't know what to do with that problem, but will try at least. 
Okay#2 : Fried the opamp 5534. Accidentally touched two pins with the probe. But traced that 6~7 volt "input" that was causing the malfunction. So I bought new opamps and give it another spin.

Still can anybody give me some more insight about it, anyone?   

 

[feza]

Hey,

Below is the area which has the high voltage. I've measured all components and all is well to specification. Perhaps I am looking wrong at tracing the error? The left input plus and minus (hot and cold) pins receive the 6~7 volts. It has to be something. The high voltage comes from pin 7 which is the VCC+ on the chip, and the +-15v or still correct. 

Thanks for your help!

Right when disconnecting the XLR input Molex the voltage raise to +15 volts! I just do not know where that is coming from?

What I did is lifting one resistor leg around the whole signal path of the left input, no result. Same thing with the 100nF and 22uF, all right. Next stop was the right output channel and did the same procedure as the left channel, no result and all correct. But now you know that the high voltage is from, probably, a +15v rail!  ;D   

*Corrected to right pin from the 5534